Co-opted directors and corporate climate risk disclosure

Today corporate boards are recognising the importance of managing climate risks, a key emergent form of business risk. Climate risks comprise of potential impacts arising from changes in natural weather patterns, the incidence of natural disasters and the technological, market, policy and legal repercussions associated with such risks (O’Dwyer and Unerman, 2020; Schiemann and Sakhel, 2019). Climate risk information is crucial for stakeholders, particularly as it influences investors’ decision-making (Bingler et al., 2022a; Kölbel et al., 2024). It can also impact companies’ operations and resulting in significant stranded assets. For instance, in 2016, a comprehensive evaluation of climate risk disclosure conducted by ConocoPhillips, Shell Plc. and Total led to their early withdrawal from Canadian oil sands. In contrast, ExxonMobil’s failure to inadequately assess and disclose climate risks resulted in the company writing off 3.5 billion oil sand reserves ^[1]. Despite the significant consequences arising from the lack of climate risk disclosures, corporate commitment to climate disclosure remains limited ^[2].

Previous research has focused on corporate voluntary environmental disclosures more generally, neglecting corporate climate-related risk disclosure. Studies examining the relationship between board structure and corporate voluntary environmental disclosures have yielded mixed results. One strand of literature suggests that the reinforcement of climate governance (Bui et al., 2020), the establishment of environmental committees (Liao et al., 2015; Peters and Romi, 2014), the enlargement of boards (Liao et al., 2015; Tauringana and Chithambo, 2015) and the inclusion of gender-diverse boards (Ben-Amar et al., 2017; Haque, 2017; Liao et al., 2015; Benjamin et al., 2020a, 2020b) enhance corporate disclosure. Alternatively, a substantial number of studies have found negative and/or no correlation between a well-structured board, captured via board independence, gender diversity, board meeting and board size, etc. and corporate voluntary environmental disclosure (Prado-Lorenzo and Garcia-Sanchez, 2010; Rupley et al., 2012; Cordeiro and Sarkis, 2008; Delmas et al., 2013; Haque, 2017) ^[3].

We attribute the differences in results to two main factors: the focus on voluntary corporate environmental disclosure and the neglect of management−director connections. Voluntary disclosures may lack credibility because of limited stakeholder (auditors and legislators) scrutiny (Farooq, Azantouti, and Zaman, 2024; Farooq, and de Villiers, 2020) allowing managers to conceal information (Li et al., 1997). In contrast, mandatory climate risk disclosures are more likely to undergo rigorous internal (managers and boards) and external scrutiny (regulators and auditors), ensuring greater reliability and trustworthiness (Tang and Demeritt, 2018). However, this may not always be the case as managers may exercise caution with mandatory disclosures to avoid revealing strategic information to competitors or disclosing sensitive information that might increase perceived risks, such as the risk of litigation or financial instability (Kays, 2022). The nature of such disclosure − voluntary versus mandatory − thus carries different rewards and incentives, significantly varying across firms based on their level of governance (Zaman et al., 2022).

In the case of mandatory disclosures, managers may rely on co-opted directors (directors appointed after the incumbent CEO) to withhold or obscure mandatory information. Despite its relevance, studies often overlook how directors’ connections to management can undermine board effectiveness related to climate risk disclosure. For instance, Zaman et al. (2021a, 2021b) highlight that directors aligned with management tend to limit disclosure to protect management interests. This intentional concealment of information, particularly environmental risk-related, can deter investments (Mattison and Mints, 2019; Mueller and Sfrappini, 2022) and expose companies to shareholder activism or legislative action (Durkin, 2024). Although Ghafoor et al. (2023) examined the association between co-opted directors and climate risk exposure, finding that the presence of such directors elevates climate risk disclosure, they did not explore whether this exposure translates into actual disclosure.

Given that managers may withhold or obscure less favourable information (Zaman et al., 2021a, 2021b; Kays, 2022), it is important to examine the role of co-opted directors in this context. Our research addresses this gap by examining the association between co-opted directors and mandatory climate risk disclosure. Specifically, we aim to answer the following research questions (RQs):

To address the above RQs, we examine an interesting case involving the US Securities and Exchange Commission’s (SEC) climate disclosure guidelines. This case highlights the divergence between the requirement to disclose climate risks and the inadequacies in regulatory enforcement for non-compliance within the US context. The US SEC mandates a materiality-based approach to climate risk disclosure, requiring public companies to identify and disclose material climate risks in their 10-K filings (SEC, 2010). Failure to comply may result in the SEC asking for a letter of explanation from the company. However, a materiality-based approach to reporting, combined with the complex and uncertain nature of climate risks and the SEC’s inadequate oversight, allows for significant managerial discretion in such disclosures (Peters and Romi, 2013). Such a setting underpins a critical ethical concern, as managers’ personal interests can influence the extent and nature of corporate climate risk disclosures. Opportunistic managers, motivated by agency considerations, may obscure these disclosures (Kim and Lu, 2011), as revealing climate risk information can increase the reporters perceived risks (Kölbel et al., 2024). In this context, the board of directors’ (BOD) role in fulfilling their fiduciary duties of transparent information dissemination to stakeholders becomes even more ethically significant.

We argue that co-opted directors, potentially influenced by loyalty to the CEO who appointed them, may neglect their fiduciary and ethical responsibilities to stakeholders by enabling the CEO and his/her team of executives to obscure climate risk disclosures. Consequently, a substantial presence of co-opted directors on a corporate board may undermine climate risk disclosure. This aligns with recent co-option literature, which suggests that such directors are associated with poor managerial monitoring (Coles et al., 2014; Zaman et al., 2021a, 2021b), a reduced likelihood of implementing clawback provisions (Huang et al., 2019), excessive managerial risk-taking (Lee et al., 2021) and overinvestment in inefficient research and development (Harris et al., 2019).

Alternatively, the presence of co-opted directors might positively impact climate risk disclosure. Due to reputational concerns, fear of losing legitimacy and career considerations, these directors may act as diligent monitors. Consequently, a significant proportion of co-opted directors could reduce board-management conflict and enhance management’s commitment to stakeholders, leading to greater disclosure. Supporting this view, Nguyen et al. (2021) find a positive relationship between corporate innovation and co-opted directors. Similarly, Bhuiyan et al. (2022) and Harris and Erkan (2021) note that co-opted directors help decrease a company’s cost of capital and improve earnings management.

To test these competing arguments, we select a comprehensive sample of US-listed companies from 2006 to 2018 using climate risk disclosure data from Kölbel et al. (2024), who capture such disclosures by using Bidirectional Encoder Representations from Transformers (BERT) ^[4] for language understanding on 10-K filings. This analysis provides original evidence indicating that companies with a large proportion of co-opted directors are less likely to disclose climate risk information in their 10-K filings. These findings are statistically and economically significant, holding across sensitivity tests, including industry adjustments, the Global Financial Crisis, SEC guidelines and endogeneity controls like propensity score matching (PSM) and difference-in-differences (DiD) analyses. Overall, we found a board’s composition with a single standard deviation rise in the presence of co-opted directors corresponds to a notable 4.2% reduction in climate risk disclosure ^[5].

Exploring how co-opted directors influence climate risk disclosure, we identify managerial risk-taking as a key mechanism and empirically test its impact. Our findings reveal that a substantial presence of co-opted directors significantly increases managerial risk-taking, leading companies to disclose less information about their climate exposure. However, the influence of co-opted directors on climate risk disclosure is rather contingent on certain conditions. Our moderation analyses show that high internal information asymmetry and weak external monitoring intensify the negative influence of co-opted directors on climate risk disclosure. Even industries with greater environmental sensitivity, which are typically associated with stronger disclosure practices, are not exempt. In such cases, co-opted directors often prioritise management’s interests over transparency, further restricting climate risk disclosures.

We contribute to the growing literature on climate risk disclosure from three perspectives. Firstly, we extend research on mandatory material climate risk disclosures by addressing their antecedents through a stakeholder-agency lens. Unlike prior studies focused on voluntary environmental disclosures (e.g. Ben-Amar et al., 2017; Liao et al., 2015), we examine mandatory disclosures, highlighting how co-opted directors facilitate the concealment of material climate risks, thereby overcoming biases associated with incomplete voluntary reporting. Secondly, we contribute to the discourse on board effectiveness by empirically demonstrating that co-opted directors, known for prioritising managerial interests (Cassell et al., 2018; Zaman et al., 2021a, 2021b), reduce the extent of mandatory climate disclosures, revealing their influence on regulatory compliance. Thirdly, we advance policy debates by providing evidence that co-opted directors impede the impartiality of independent boards and disproportionately reduce transitional risk disclosures under the SEC’s 2010 guidelines. These findings underscore the need for stricter governance reforms to limit CEO influence in board appointments and enhance climate risk transparency.

The remaining paper is organised as follows: Section 2 outlines the theoretical framework and hypotheses, Section 3 details the methodology, Section 4 presents the empirical results, Section 5 provides additional analyses, Section 6 discusses robustness tests, Section 7 offers the conclusion and discussion and Section 8 highlights the implications of the study.

Climate risk includes potential impacts arising from changes in natural weather patterns, the incidence of natural disasters and consequential technological, market, policy and legal changes (O’Dwyer and Unerman, 2020; Schiemann and Sakhel, 2019). It can be divided into two main categories ^[6]. Physical risks arise from changes in weather patterns or extreme weather events such as floods, droughts and wildfires. These events can deplete water resources and lead to declines in agricultural production, air pollution, biodiversity loss, infrastructure damage and loss of life. These risks can negatively impact corporate financial performance by contributing to stranded assets, disrupting business operations and compromising supply chains (Chenet et al., 2019; TCFD, 2017). Transitional risks, on the other hand, arise from society’s transition to a low carbon economy and the resulting regulatory reforms that require companies to adopt low-carbon technologies, practices and processes ( Kazemian et al., 2024; Chenet et al., 2019; O’Dwyer and Unerman, 2020; TCFD, 2017).

Climate risks have the potential to cause severe damage to economies and businesses worldwide. For example, according to Martinich and Crimmins (2019) sectoral impact models, the USA alone could lose $520bn across 22 sectors due to climate change. In addition, the International Labour Organisation (ILO) of the United Nations predicts that if global warming continues its current trajectory with a predicted temperature rise of 1.5°C, 80 million jobs could be lost by 2030 ^[7]. These alarming predictions have led to increased interest from global and national policymakers in climate risk reporting, resulting in countries worldwide issuing regulations and policy guidelines on climate risk disclosures ^[8].

In 2010, the US SEC issued policy guidelines mandating listed companies to disclose information on material climate risks (SEC, 2010). These guidelines are based on materiality principles, and the SEC issues comment letters to companies found to be non-compliant. Coburn and Cook (2014) have evaluated the effectiveness of these guidelines and reported a notable decline in SEC enforcement regarding poor corporate climate risk disclosures.

While climate risk disclosures are mandatory, the regulations allow for considerable managerial discretion, including decisions on whether a risk is material. This discretion, coupled with uncertainties about enforcement (Coburn and Cook, 2014; Peters and Romi, 2013), poses challenges for corporate climate risk disclosure. Scholars agree that managerial discretion is largely influenced by the composition of the board of directors (Zaman et al., 2021a, 2021b). It is plausible that substantial presence of co-opted directors influences companies’ climate risk disclosure compliance. Nevertheless, most research in this area has been limited to voluntary climate risk disclosures by companies participating in the Carbon Disclosure Project (Lewis et al., 2014; Schneider et al., 2018). Extant studies’ find that voluntary climate risk disclosures are affected by the presence of large institutional shareholdings, corporate financial performance and managerial characteristics (Bolton and Kacperczyk, 2021; Lewis et al., 2014). In addition, the literature has noted that climate governance strength (Bui et al., 2020); environmental committees establishment (Peters and Romi, 2014); board size (Tauringana and Chithambo, 2015); and board gender diversity (Ben-Amar et al., 2017; Liao et al., 2015) also influence such disclosure. Unfortunately, despite the relevance of co-opted directors with climate change risks, none of the above studies have made any effort to examine their association, with exception of Ghafoor et al. (2023). Importantly, Ghafoor et al.’s (2023) study was limited to climate change risk exposure rather than the provision of mandatory climate risk disclosure in 10-K filings, with findings indicating that co-opted directors increase companies’ exposure to climate risk. However, whether and how co-opted directors affect companies’ compliance with mandatory climate risk disclosure regulations remained unanswered in the extant literature.

It is important to investigate whether co-opted directors, who may have closer ties to the company’s management, are more likely to influence the withholding of climate risk information to maintain the status quo. Alternatively, they might push for greater disclosure to benefit the company and its stakeholders in the long term. Understanding the influence of co-opted directors on mandatory climate risk disclosures is valuable for both academic research and corporate practice.

In contemporary literature, corporate governance (CG) discourse has been significantly shaped by agency theory and, more recently, by stakeholder-agency theory, with a specific emphasis on safeguarding stakeholder rights (Aguilera et al., 2015). Stakeholder-agency theory contends that separating control and ownership gives rise to agency conflicts, particularly stemming from divergent objectives (Hill and Jones, 1992) between managers (acting as agents) and multiple owners (acting as principals). For instance, stakeholders tend to prioritise long-term company value and performance. In contrast, managers driven by agency incentives like rewards and career considerations generally emphasise short-term performance, often leading to excessive risk-taking. This discrepancy between stakeholder and management objectives creates heterogeneity (Jain and Zaman, 2020).

Consequently, self-interested managers, owing to their access to information, can exploit their positions to damage other stakeholders’ interests (Jensen and Meckling, 1976) and deliberately obscure the flow of information (Campbell and Slack, 2011; Jensen and Meckling, 1976). To limit such managers, the stakeholder-agency paradigm emphasises strengthening formal CG mechanisms primarily through well-structured and independent boards (Jain and Zaman, 2020).

A well-structured board, especially one dominated by independent directors, is more effectively poised to oversee managerial conduct and ensure regulatory compliance (Brickley et al., 1997; Jain and Zaman, 2020) with climate risk disclosure. While undertaking their fiduciary duties, directors are shielded by the business judgement rule ^[9] so long as they ensure that their actions are aligned with the optimal interests of their companies and stakeholders. However, scholars argue that the BoD faces multiple constraints. For example, the lack of skills/expertise, limited resources and CEOs’ involvement in their selection process may impede their monitoring ability. We contend these arguments and argue that even if directors are given all the necessary resources and possess relevant skills and expertise, a CEO’s involvement in their selection process would remain an inherent limitation to their impartiality (Zaman et al., 2021a, 2021b). Therefore, in the presence of weak monitoring managers will be encouraged to pursue their own agendas, leading to poor regulatory compliance regarding climate risk disclosure, which remains largely at managerial discretion, exacerbating the agency conflict between companies and their stakeholders. Accordingly, we contend that stakeholder-agency theory provides a suitable framework to examine the association between climate risk disclosure and co-opted directors. We have developed the theoretical model, as illustrated in Figure 1, to address the research questions of this study.

Previous literature on CG has reflected on the CEO’s involvement in the BoD’s appointment in numerous ways, including the direct selection of the candidate, through a mock search process, or influencing the directors’ nomination committee (see Clune et al., 2014). Despite the 2002 Sarbanes–Oxley Act’s requirement for greater board independence, the CEO’s influence in the selection process is still exerted (Carcello et al., 2011, p. 748). For example, Clune et al. (2014) interviewed the New York Stock Exchange (NYSC) listed companies’ nomination committee chairs, in which one of the interviewees stated that [t]he reality is, we don’t have a [director] search procedure. The CEO personally selected the last three directors, and they went through a simulated process. Accordingly, selecting CEO-influenced directors compromises their impartiality and efficacy in decision-making (Coles et al., 2014). The co-opted directors’ loyalty is extended to their appointees rather than impartially monitoring managers and protecting stakeholders’ rights (Zaman et al., 2021a, 2021b).

Prior studies characterise such directors as deficient overseers of managers and argue that they support the managerial status quo and are less likely to dismiss a CEO in instances of corporate malpractice (Zaman et al., 2021a, 2021b). The extent of support for the status quo and management protection depends on the proportion of co-opted directors on a board (Coles et al., 2014). Furthermore, such directors may also wrongly strengthen managers’ confidence that they can effectively avoid facing adverse outcomes due to their misconduct (Zaman et al., 2021a, 2021b). This is because, when the violation occurs, management may influence these directors to manipulate or obscure internal records, creating challenges for legal authorities to establish in court that malpractices occurred (Arlen and Carney, 1992).

Similarly, accounting scholars have investigated the impact of CEO influence on director appointments across various accounting measures. For example, the research by Krishnan et al. (2011) indicates when social connections exist between a CEO/CFO and board members, companies are more likely to engage in greater earnings management. Likewise, Dey et al. (2011) also examine the impact of CEOs’ social connections with independent directors on financial reporting quality, proxied by accrual quality, restatement and analyst forecast error. Their findings establish that the substantial presence of socially connected directors impedes financial reporting quality. The findings of Huang et al. (2019) also indicate that the presence of co-opted directors decreases a company’s propensity to adopt clawback provisions. More recently, Zaman et al. (2021a, 2021b) provided empirics of the involvement of co-opted directors in stakeholder violations. Overall, the literature concludes that co-opted directors’ poor monitoring increases stakeholder-agency conflicts and exacerbates managerial non-compliance.

We contend that companies may compromise compliance with climate risk disclosure by leveraging the influence of friendly directors. The SEC’s guidelines on climate risk disclosure follow a principles-based approach, allowing managerial discretion in identifying and disclosing material climate risks (SEC, 2010). In this context, it is the BoD’s responsibility to ensure that companies adopt transparent procedures for disclosing risks related to climate change. However, disclosing such risks has significant implications for companies’ operations, providing managers with incentives to avoid full disclosure. For example, disclosing material climate risks can incur additional costs (Brammer and Pavelin, 2006; Verrecchia, 1983), affect financial performance (Scanlan, 2021; Qiu, Shaukat and Tharyan, 2016) and increase legal scrutiny (Scanlan, 2021), thus raising the probability of litigation (Matsumura, Prakash and Vera-Muñoz, 2024). In addition, revealing these risks may turn investors away and negatively impact the company’s share price, undermining the CEO’s efforts to maximise profits. Although disclosing material risks is mandatory, management can determine whether a risk is material, allowing opportunistic managers to seek board support in obscuring such information.

Consequently, these directors may be more inclined to support efforts to obscure climate risk disclosures, in an effort to mitigate legal and reputational risks, over the need to promote transparency and accountability to stakeholders. For instance, some companies in the oil and gas sector have claimed that despite their intent to follow the task force on climate-related financial disclosures (TCFD) recommendations, contractual, practical or legal reasons limit their ability to fully reveal climate risks to their stakeholders (WBCSD, 2018). Therefore, due to the CEO’s influence in their selection, co-opted directors may prioritise facilitating managers’ efforts to obscure disclosure, over their fiduciary responsibility to stakeholders, thereby exacerbating stakeholder-agency conflicts. As a result, the significant presence of co-opted directors on the board could reduce climate risk disclosure, leading to our first hypothesis:

Recent literature on co-opted directors provides empirical evidence that a substantial presence of such directors on corporate boards encourages managerial risk-taking (Lee et al., 2021). Extending this, we argue that excessive risk-taking may tempt boards to engage in activities with high environmental risks and/or demonstrate a lack of interest in investing in environmental risk reduction systems and processes. Our intuition aligns with the prior literature that provides empirical evidence of a positive association between managerial risk-taking and poor financial reporting quality. For instance, Chakrabarty et al. (2018) indicated that managers with higher risk-taking (captured through greater option Vega) obfuscate information to their stakeholders by providing less readable disclosure. Armstrong et al. (2013) also echoed that higher option Vega is associated with increased misreporting in financial statements.

In addition to associating managerial risk-taking with financial reporting quality, some studies present evidence of managerial risk-taking’s influence on corporate sustainability outcomes. For instance, Bouslah et al. (2018) investigated the impact of managerial risk-taking incentives on corporate stakeholder-related controversies, including environment-related wrongdoings, suggesting such risk-taking significantly increases corporate involvement in socially irresponsible activities. Similarly, Mayberry (2020) demonstrates that risk-taking discourages firms’ sustainability initiatives. Considering both standpoints – the possibility of co-opted directors inciting managerial risk-taking and the correlation between managerial risk-taking and increased probability of reporting quality impairment (Armstrong et al., 2013; Call et al., 2016; Chakrabarty et al., 2018; Kim and Lu, 2011) as well as non-compliance (Bouslah et al., 2018; Jain et al., 2023), we expect such directors to influence climate risk disclosure via managerial risk-taking, formulating the second hypothesis for our study.

Our previous hypotheses articulate a negative correlation between the presence of co-opted directors and climate risk disclosure, with managerial risk-taking as a potential mediating mechanism. We further our exploration by examining whether the information environment within a company influences co-opted directors to obscure climate risk information. Existing research on board co-option suggests that co-opted directors exploit information asymmetry to protect management interests over those of stakeholders (Zaman et al., 2021a) and information asymmetry creates a less transparent environment (Brown and Hillegeist, 2007; Healy et al., 1999; Heflin et al., 2005). Theoretically, information asymmetry is a significant factor in stakeholder-agency conflict (Jain and Zaman, 2020), as opaque information environments provide opportunities for management to engage in rent-seeking activities (Kerr, 2019). Empirical evidence indicates that the negative impact of co-opted directors on stakeholder interests is more pronounced in companies with poor information environments, where information opacity can promote corporate malpractice, such as tax avoidance (Kerr, 2019).

Therefore, it is imperative to test the influence of a company’s internal information environment on the relationship between co-opted directors and climate risk disclosure. Specifically, we hypothesise that in companies with poor information environments, the ability of co-opted directors to obscure climate risk information is enhanced due to greater information asymmetry. This environment allows co-opted directors to act more effectively in favour of management and against stakeholder interests.

We extend our theoretical framing by examining the impact of external monitors, such as analysts and institutional investors, on the association between co-opted directors and climate risk disclosure. Our motivation stems from the ongoing debate among CG scholars, as underscored by Aguilera et al. (2015). These authors argue that internal CG mechanisms, such as directors’ functions, should not be assumed to operate independently of the external environment. This is crucial because, although both board directors and external monitors share the common objective of minimising agency conflicts, they possess distinct characteristics interconnected with company performance. For instance, institutional investors, due to their investment horizon, are more interested in the company’s market performance (Zaman, 2024). On the other hand, company directors, as exemplified by a well-structured board, prioritise long-term organisational performance (Zaman et al., 2022). Given the substantial variations in roles, responsibilities and expectations between internal and external monitors, it becomes crucial for us to examine the behaviour of co-opted directors regarding climate risk disclosure in the context of heightened external monitoring.

Previous research has recognised the importance of external monitors, including a higher number of institutional investors and analysts, in mitigating agency conflicts and fostering responsible business practices (Jain and Zaman, 2020; Liu, 2014). In addition, the literature on information asymmetry highlights that rigorous external monitoring by such institutions can reduce managerial tendencies to withhold bad news (Zaman et al., 2021b). Studies have found that external monitors influence corporate reporting practices (Roberts et al., 2006; Solomon and Solomon, 2006; Solomon et al., 2011). Similarly, external monitors have been found to influence the behaviour of co-opted directors regarding corporate violations (Zaman et al., 2021a) and investment (McCahery et al., 2016). Also, external monitoring by institutional investors reduces opportunistic behaviour by managers and mitigates the adverse effects of high board co-option on insider trading (Rubin and Smith, 2009; Rahman et al., 2021). Therefore, we anticipate that a lower representation of external monitors could lead to a higher level of board co-option and a closer relationship between the board and CEO, which may have significant negative implications for climate risk disclosure.

Prior studies have consistently emphasised that climate change risk is closely linked to industrial heterogeneity (Cho and Patten, 2007; Benjamin et al., 2023). These studies illustrate that companies in environmentally sensitive industries are particularly vulnerable to climate change risk compared to other industries. This vulnerability stems from their heavy reliance on natural resources (Bui et al., 2020; Carvajal et al., 2022; Nadeem et al., 2021) and the heightened potential for litigation, which may increase managers’ incentives to enhance information asymmetry. Accordingly, previous studies have reported that managers in companies operating in environmentally sensitive industries engage in higher bad news hoarding than those in other industries (Bui et al., 2020; Carvajal et al., 2022; Jain and Zaman, 2020). Likewise, Clarkson et al. (2004) noted a higher level of opaqueness in climate disclosure among companies in these industries. These companies face increased covert environmental liabilities, resulting in greater financial obligations related to climate compliance.

In addition, stakeholders, including investors, are becoming more assertive and sophisticated in their expectations of companies’ management of climate-related issues. As a result, there is a growing demand for increased disclosure on climate-related risks (Solomon et al., 2011), particularly from companies operating in environmentally sensitive industries (Chelli et al., 2019; Passetti et al., 2018). Considering the mounting pressure on companies’ management to report on climate issues and their reluctance to disseminate such information, the role of co-opted directors becomes increasingly critical. We argue that, due to their loyalty to the company’s management, co-opted directors may be inclined to support actions that result in lower disclosure of climate risks. Accordingly, we anticipate that the negative association between co-opted directors and climate risk disclosure will be more pronounced for companies operating in environmentally sensitive industries compared to those in less sensitive industries.

The data for our study comes from a variety of sources, including Kölbel et al. (2024), Lalitha Naveen’s ExecComp, Bloomberg and Compustat. Climate risk disclosure data for S&P500 companies was obtained from Kölbel et al. (2024) ^[10]. Lalitha Naveen’s website ^[11] was used as the source of data on board co-option (Coles et al., 2014). CG and board-level variables have been retrieved from Bloomberg. Financial data for companies was obtained from the Compustat database. ExecuComp was used to gather CEO compensation information and create measures of managerial risk-taking (CEO Vega). We winsorised explanatory variables, except the dummies, at the 1st and 99th percentiles to remove outliers. Our final sample consists of 2,975 observations taken between 2006 and 2018 ^[12].

We acknowledge that previous research on corporate climate change disclosures has relied on three main measures, including ESG disclosure ratings from databases like Bloomberg, mapping climate risk disclosure through companies’ participation in voluntary programmes such as the carbon disclosure project (CDP), and carbon emissions disclosure (see Zaman et al., 2022). However, these measures suffer from certain limitations, such as the fact that companies may strategically abstain from disclosing information that could compromise investors’ decisions or that they deem to be worthless. In addition, voluntary disclosures are not enforceable or subject to mandatory audit, so the information provided may not be accurate or reliable. In contrast, the SEC’s climate risk disclosure guidelines require companies to disclose information on climate risks that are material to their business in their 10-K filings, making them directly accessible to stakeholders. Therefore, we prefer a regulatory approach and use companies’ regulatory disclosures to the US SEC as their main source of data (Hain et al., 2022; Kölbel et al., 2024).

Previous studies on corporate disclosure mainly used bag of words approaches, where researchers used pre-defined keywords to identify relevant information in corporate filings (see, Torreggiani and De Giacomo, 2022). Although this approach helped to transform text into numerical data, it had limitations such as poor identification of relevant sentences and false positives with low accuracy (Bingler et al., 2022a; Varini et al., 2020). To overcome these issues, recent studies have started using advanced AI-based pre-trained neural language models such as BERT to map corporate climate risk disclosure from 10-K filings. Compared to traditional keyword approaches, recent studies have shown that BERT is more effective in identifying and analysing corporate climate risk disclosures (see Kölbel et al., 2024; Hain et al., 2022 and Bingler et al., 2022a).

Following the findings of recent stream of literature (see Bingler et al., 2022b; Hain et al., 2022), we used climate risk disclosure of Kölbel et al. (2024), who capture climate risk disclosure from 10-K reports using BERT. More specifically, Kölbel et al. (2024) performed a three-step process to capture climate risk disclosure from 10-K filings using BERT. Firstly, they trained BERT on the thousand sentences selected related to climate risks that include physical and transition climate risks from the sample reports provided in the TCFD guidelines. Secondly, they used BERT to perform two tasks; identification of climate-relevant information (sentences) topics from Item 1A of 10-K filings; and differentiation of climate-relevant information (sentences) related to physical or transition climate risks. These two tasks have generated the raw score for each sentence of SEC filings that reflect the probability of that sentences being related to climate risk, physical risk or transition risk. Finally, they calculated the aggregated score by following a two-step procedure; (i) created a binary variable that take the value of “1” if the raw score by each sentence is above the pre-specified probability threshold i.e. 0.80 and (ii) averaged the binary variables values of 1 or 0 for a given document ^[13], representing the aggregate climate risk score of 10-K filings i.e. Total Climate Risk Disclosure, Physical Climate Risk Disclosure and Transition Climate Risk Disclosure. Notably, some recent studies have validated these scores by using various techniques and across all such techniques, BERT has outperformed traditional approaches by large margins (Bingler et al., 2022a; Hain et al., 2022).

The metric for board co-option is predicated on the ratio of directors co-opted by an existing CEO. Directors who become part of a company’s board subsequent to the commencement of an incumbent CEO’s tenure are considered co-opted (Coles et al., 2014). We used four distinct proxies to capture board co-option. The first proxy, labelled as “Co-option,” quantifies the proportion of co-opted directors in the total count of directors on a BOD. The second proxy, designated as “Co-option (TW)” or tenure-weighted co-option, measures proportion of the cumulative tenure of co-opted directors in the combined tenure of all directors. This proxy encapsulates the influence of directors’ tenures in the co-option process, acknowledging that directors with longer service are likely to wield greater impact on board decisions. The third proxy, denoted as “Co-option (Independence),” measures the proportion of co-opted directors who hold independent status, relative to the board size. These independent co-opted directors possess enhanced influence over boards proceedings owing to their autonomous status. The fourth proxy, named “Co-option (TW Independence)” represents the tenure-weighted evaluation of co-opted independent directors. It quantifies the proportion of total tenure of co-opted independent directors relative to the cumulative tenure of all directors.

We use three sets of control variables that have the potential to influence corporate disclosure in general and corporate climate risk disclosures in specific. The first one includes the characteristics of BODs. Previous studies have found that climate voluntary disclosure is influenced by various factors, such as the size of the board (Ln_BSIZE), the presence of female directors on the board (BGD) and CEO board representation (CEO Board Member) (Carvajal et al., 2022; Rupley et al., 2012). To account for the influence of CG on climate disclosure, we also included a control variable based on the Bloomberg CG Index (GOV Index), following prior literature (Haque, 2017; Peters and Romi, 2014; Schiemann and Sakhel, 2019). In addition, we controlled for firm performance [Sales Growth, return on assets (ROA), and market-to-book ratio (MTB)], company size (FSIZE), leverage (Leverage), capital expenditure (CAPX_AT), cash holding (CASH) and research and development (R&D_AT). We also controlled for merger and acquisition activity (M&A) and corporate age (Ln_FAGE), as these factors may affect a company’s disclosure choices. See  Appendix for detailed definitions of the variables.

The descriptive statistics for the variables used in the study are presented in Table 1. The sample mean for the total climate risk disclosure score is 0.106, indicating that overall climate risk disclosure is relatively low, despite the mandatory disclosure requirements set by the SEC. The average mean score for the two components of climate risk disclosure reveals that companies tend to disclose more information on transition risks (0.068) compared to physical risks (0.029). The descriptive statistics for the dependent variables are equivalent to those reported by Kölbel et al. (2024). The descriptive statistics for the explanatory variable suggest that on average 42.1% of total directors are co-opted (Co-option). Tenure weighted co-opted directors in our sample companies represent 24.2% of the total number of directors on board. In terms of independent co-opted directors, the proportion of co-option (TW Independence), and co-option (Independence) are 20.4% and 36.7%, respectively. The average mean value of board co-options proxies is comparable with those documented in prior studies such as Zaman et al. (2021a, 2021b). In terms of our control variable, we find on average 18.3% of total directors in our sample are female directors with on average 11 directors on board (before natural logarithm). Approximately 95.5% of CEOs in our sample hold board membership. In terms of governance, we find our sample companies have moderate (a mean score of 61.7%) CG practices on average. All financial control variables align with those documented in the extant literature (Zaman et al., 2021a, 2021b).

We have also conducted a Pearson correlation test to rule out any multicollinearity issues which can occur when independent variables in a regression model are highly correlated with each other. The results showed a negative correlation between the number of co-opted directors and climate risk disclosure. However, the correlation coefficient of all the independent variables remained below the 0.80 threshold, indicating that there were no major concerns with multicollinearity in the analysis ^[14].

We begin our analysis by presenting our findings in Table 2. To investigate the effect of co-opted directors on climate risk disclosures, we estimate the following regression model:

where corporate Climate Risk Disclosure is captured by using BERT, an artificial intelligence (AI)-based algorithm for natural language processing (NPL), on 10-K filings (see Section 3.2.1). Board co-option is proxied by fours variables (see Section 3.2.2). Incorporating state, industry and year fixed effects, Control comprises a vector encompassing all control variables outlined in Section 3.2.3. To mitigate potential biases arising from time-invariant factors jointly impacting both dependent and independent variables, we include state and industry fixed effects. In addition, year fixed effects aid in addressing shared macroeconomic fluctuations. Details of these variables can be found in Appendix while the findings are presented in Table 2.

The results from Models (1) to (4) show that the coefficient estimates for board co-option on climate risk disclosure are significant and negative at the 1% level, indicating that having a higher ratio of co-opted directors and independent co-opted directors on the board reduces the level of climate risk disclosure. These findings have practical significance as they suggest that a one standard deviation increase in board co-option is associated with a 4.2% decrease in climate risk disclosure. These results are consistent with stakeholder-agency theory and support H1, which suggests that co-opted directors appointed through CEO patronage fail to fulfill their fiduciary duty to comply with SEC guidelines and exacerbate stakeholder-agency conflicts by enabling managers to withhold climate risk information from stakeholders.

The finding related to co-opted independent directors is noteworthy as it not only echoes the monitoring inadequacies associated with board co-option but also provides an explanation for the mixed findings of prior studies on independent directors’ effectiveness (see Coles et al., 2014; Horváth and Spirollari, 2012; Zaman et al., 2021a). Although Boivie et al. (2016) have theoretically mapped the barriers that hinder boards of directors’ impartial monitoring ability (e.g. lack of skills and external job demands), the current study provides empirical support that their allegiance to management is an important factor that impedes their mandatory compliance ability, resulting in lower climate risk disclosure.

In terms of control variables, the significant negative relationship between gender diversity and climate risk disclosure is in line with the literature that echoes female directors as risk averse. Since disclosure of information increase the probability of litigation, therefore, female directors avoid such disclosure. In terms of overall CG, the study finds companies with better CG score encourages such disclosure. All other control variables findings are also consistent with the prior literature on corporate environmentalism (Zaman et al., 2021a, 2021b).

In the abovementioned results, we have determined that the co-opted directors and independent co-opted directors are negatively associated with corporate climate risk disclosure. Our theoretical underpinning assumes that the disengagement of the principal (shareholders) in decision-making stimulates agency conflicts – particularly those arising due to differences in objectives (Brennan and Solomon, 2008).

Previous literature has examined the impact of CEO characteristics on the relationship between board co-option and climate-related risks. For example, Zhang and Liu (2022) have demonstrated that the personal experiences of CEOs can determine their insights about climate change and the pro-environmental behaviour of the companies they run. In this section, we systematically examine whether managerial risk-taking serves as a mediating factor in the association between co-opted directors and independent co-opted directors, respectively. CEO Vega is used to capture manager risk-taking behaviour (Zaman et al., 2021a, 2021b).

This mediation has been tested by applying four methods:

1. Causal step regression of Baron and Kenny (1986);

2. Goodman (1960) tests;

3. Sobel (1982) tests; and

4. Aroian (1947) tests.

The study first applied Baron and Kenny’s four-step regression.

The first step involves regressing the climate risk disclosure measure (dependent variable) against the co-opted directors (independent variable) to assess the relationship. In the next step, CEO Vega (mediator) is regressed on the co-opted directors (independent variable) to verify the effect of independent variable on the mediator. The third step involves regressing the climate risk disclosures measure on CEO Vega to check the influence of mediator on the dependent variable. In the last step, the climate risk disclosure measure is regressed on both co-opted directors (independent variable) and CEO Vega (mediator) to identify if the association between the co-opted directors (independent variable) and climate risk disclosure (dependent variable) is influenced by the inclusion of this mediator. The insignificant results will demonstrate a full mediation and a lower magnitude of the coefficient describes partial mediation. To execute this analysis, we followed a four-step process, the details of which are outlined, and the results reported in Table 3.

The results in Panels A1 and B1, Models (1) and (4) support the study baseline results, demonstrating that a decrease in climate risk disclosure for firms due to the presence of both co-opted directors and independent co-opted directors. Similarly, the significant positive relationship in column (2) of Panels A1 and B1 suggests that the high level of co-option and independent directors’ co-option significantly elevates the level of risk taking by company managers. Panels A1 and B1 [column (3)] demonstrates negative significant coefficients on CEO Vega, suggesting that the high level of managerial risk-taking discourage firms to disclose climate-related information to stakeholders. In Panel A1, column (4) the coefficient of co-option is negative, significant and smaller in magnitude than the coefficients of co-option in column (1), illustrating that CEO’s capacity to take risks partially mediates the relationship between board co-option and climate risk disclosure. Column (4) of Panel B1 shows that the coefficient of independent co-opted directors is negative, significant and smaller in value as compared to the values reported in column (1), indicating that managerial risk-taking partially mediates the association between independent co-opted directors and climate risk disclosure.

Baron and Kenny’s (1986) stepwise causal regression demonstrates partial mediating effect of managerial risk-taking on the relationship between board co-option, independent board co-option and climate risk disclosure. Thus, the Sobel (1982), Goodman (1960) and Aroian (1947) tests were used. The t-test values from equations (2b) and (2c) were used to construct these tests. The results in Panels A2 and B2 exhibit significant p-values of all the tests at 5% level of confidence, demonstrating the significant difference in coefficients.

To capture companies’ internal information environment, we rely on linguistic obfuscation, which is the deliberate use of jargon and complex language structure with the aim of confusing, distracting and perplexing readers (see Courtis, 2004; Fabrizio and Kim, 2019). Accordingly, we use Bonsall IV et al. (2017) BOG index to capture the company’s internal information environment. The index entirely represents the guidelines of the US Securities and Exchange Commission (SEC) concerning transparent stakeholders’ communication. It relies on an exclusive collection of over 200,000 phrases. Several studies have used this index to capture linguistic obfuscation (Rjiba et al., 2021). A high score on the index signifies greater linguistic obfuscation and indicates a poor information environment. More specifically, we apply a subsampling approach using a binary variable that takes the value of 1 if the BIG Index value is greater than the cross-sectional median, i.e. high linguistic obfuscation, and otherwise zero, and vice versa for low linguistic obfuscation. We re-run equation (1) on two subsamples, i.e. high linguistic obfuscation and low linguistic obfuscation. We report the results in Panel A of Table 4.

Our results show that there exists a negative relationship between co-opted directors and corporate climate risk disclosures for companies characterised by high linguistic obfuscation, suggesting that the co-opted directors use the prevailing environment of information asymmetry to protect management interests, potentially to the detriment of company’s stakeholders. These provide support for H3, that in companies with poor information environments, the ability of co-opted directors to obscure climate risk information is enhanced due to greater information asymmetry, allowing co-opted directors to act in favour of management and against stakeholder interests more effectively.

To capture the extent of external monitors, we followed Aggarwal, Cao, and Chen’s (2012) approach and used a composite SUM based on the mean decile ranking spanning the three external monitors, which include the presence of institutional followings, stock traders and analyst followings. A higher value of SUM reflects greater external monitoring. We then used a subsampling approach where the value is set to 1 if the SUM value is greater than the cross-sectional median, indicating high external monitoring, and 0 otherwise, and vice versa for low external monitoring. We re-ran equation (1) on two subsamples: high external monitoring and low external monitoring. The results are reported in Panel B of Table 4.

Our results in Panel B indicate that the negative relationship between co-opted directors and climate risk disclosure is more pronounced in companies with a higher presence of external monitors. This supports our H4. These findings suggest that when external monitoring is low, co-opted directors have greater opportunities to support managerial information obfuscation. This, in turn, is associated with a tendency to withhold climate risk information from company stakeholders.

We followed prior literature to classify the companies as environmentally sensitive and environmentally insensitive industries (Clarkson et al., 2004; Benjamin et al., 2020a, 2020b; Benjamin, Biswas, Wellalage, and Man, 2023). We re-run equation (1) on two subsamples, i.e. environmentally sensitive industries and environmentally insensitive. The results are reported in Panel C of Table 4.

Our results indicate that co-opted directors have a negative and significant relationship with climate risk disclosure in companies operating within environmentally sensitive industries, confirming H5. In contrast, this relationship is less pronounced in industries that are less environmentally sensitive. This suggests that in environmentally sensitive industries, co-opted directors’ allegiance to management leads to lower levels of climate risk disclosure, thereby withholding crucial information from company stakeholders.

Prior studies that investigate the impact of climate risks on different corporate outcomes have used a homogeneous approach by focusing either on physical risks (Schiemann and Sakhel, 2019) or transition risks (Tang and Demeritt, 2018). Therefore, the literature mapping heterogeneity in corporate climate disclosures remains limited. Previous literature comparing the types of climate risk disclosure across companies found a significant variation between physical and transition climate risk disclosures (Elijido-Ten and Clarkson, 2019). This variation is due to the market’s risk perception associated with different types of climate risk disclosure. The market response to transition risks is generally more severe than physical risk disclosures. For example, on the one hand, Dunz et al. (2021) documented that banks increased the interest rate for high-carbon firms after a rise in carbon tax (transition climate risk). On the other hand, the results of Garbarino and Guin (2021) suggest that lenders consider ex-post extreme events (e.g. physical climate risk) as one-off occurrences and do not incorporate such events into their mortgage valuation. Since firm stakeholders perceive each type of climate risk differently (Kölbel et al., 2024), we may observe a different response from co-opted directors across physical and transition climate risks disclosures. To investigate whether the co-opted directors exhibit differences across physical and transition climate risks, we repeated our baseline analysis, replacing the study’s composite measure of climate risk disclosures with physical and transition risk disclosures. The results are reported in Table 5. Note that the calculation of physical risk disclosures and transition risk disclosures is discussed in Section 3.1.

Panel A of Table 5 showcase the results of the influence of co-opted directors on physical climate risk, while Panel B reports their impact on transition climate risk. The findings show that the estimate of the coefficient of board co-option on transition climate risk in all models (i.e. 1–4) are more significant and more negative at the 1% level compared to the estimates of physical climate risk models. This result may be due to the fact that transition risks tend to exhibit an increased likelihood of being linked with heightened risk perception as emerging regulations are targeting these risks more intensively, compared to physical risks (Kölbel et al., 2024). Therefore, it appears that co-opted directors are more inclined to support management in obscuring transition climate risk disclosure, compared to physical climate risk.

In this section, we conducted several robustness tests to validate that the negative association between co-opted directors and climate risk disclosure is not dependent on conditions such as SEC regulations stringency, climate risk measures and financial crises. Firstly, we examined SEC regulatory stringency, which is important due to the additional policy guidance issued by the SEC in 2010. This guidance mandates companies listed on stock exchanges to disclose information regarding material climate risks (SEC, 2010). These guidelines follow a materiality-based approach, where the SEC issues comment letters to companies deemed non-compliant. Such regulatory changes have the potential to influence the behaviour of co-opted directors, thus impacting their disclosure choices. Therefore, we aim to examine whether the heterogeneity in climate risk disclosure regulations affects co-opted directors and climate risk disclosure. To do so, we use a split-sample approach and re-run equation (1) on two sub-samples: the pre-SEC 2010 Climate Risk Guidelines and the post-SEC 2010 Climate Risk Guidelines. Secondly, climate risk disclosure may vary across industries, with some industries disclosing more information compared to others. Therefore, we used industry adjusted climate risk disclosure score. Finally, tight financial position induced by the global financial crisis (GFC) may hamper companies board appointment and investment in green technologies (Zaman et al., 2021a, 2021b), subsequently influences our baseline results. Therefore, we excluded global financial crisis from the sample and re-run equation (1). Table 6 shows the documentation of all three robustness results.

Our results reported in Panel A1, and A2 of Table 6 show that the coefficient estimates of co-opted directors, and independent co-opted directors on climate risk disclosure of companies remains consistently significant and negative in both the pre-SEC 2010 Climate Risk Guidelines and the Post-SEC 2010 Climate Risk Guidelines. However, in the absence of specific SEC Climate Risk Guidelines, the negative coefficient on the relationship between co-opted directors, and independent co-opted directors on climate risk disclosure becomes more pronounced. Similarly, the qualitatively similar results reported in Panels B and C indicate that our baseline results are not prone to industrial variation in climate change risk disclosure and the global financial crisis.

The association between board co-option corporate climate risk disclosure might suffer from endogeneity biases. To ensure our results are not prone to such biases, we used three tests (i) propensity matching score (PSM), (ii) reverse causality and (iii) difference in difference (DiD) analysis.

In this section, we use PSM to address likely selection bias occurring from function misspecifications and characteristics of firms. We create two groups. Treatment groups comprise companies with a board co-option proportion exceeding the upper quartile. Control firms are selected as matched counterparts from the lower quartile of board co-option proportions, achieved through PSM using the nearest-firm approach with replacement. The same set of control variables as detailed in Table 2 is used for matching. The findings of the PSM analysis are presented in Table 7.

Table 7 (Panel A) showcases the univariate average comparisons between characteristics of the treatment and control group of firms, along with their respective t-statistics. Panel B displays the results of the PSM regressions on the matched sample. The results in Panel A of Table 7 suggest that the matching variables’ average values between treatment and control firms exhibit qualitative similarity. Nonetheless, a distinct pattern emerges concerning the average value of climate risk disclosure which remains negative and exhibits statistical significance between the control and treatment firms. This provides initial evidence that companies in the treatment group marked by substantial presence of co-opted directors provide significantly less disclosure compared to control group firms, while keeping all other covariates consistent. To further confirm the relationship between co-opted directors and climate risk disclosure, we conducted equation (1) analysis on the PSM matched sample. The significant negative relationship observed between the co-opted directors and climate risk disclosure in Panel B indicates that our results are not prone to functional misspecifications. This finding strengthens the validity of our findings regarding the relationship between co-opted directors and climate risk disclosure.

Recent reviews of CG suggest that reverse causality is a major concern that many prior studies have confronted (see Jain and Jamali, 2016; Zaman et al., 2022). We acknowledge that our baseline results have the potential to suffer from reverse causality concerns. This is because the regulatory pressure for higher climate risk disclosure may influence companies’ board appointments. Therefore, to ensure that such an issue does not influence our baseline results, we followed prior literature and identified companies with zero changes in climate risk disclosure across a two-year period and rerun the equation (1) on such a sample. By doing so, we aim to minimise the probability of climate risk disclosure influencing board appointments, especially when co-option remains unchanged over the two-year period. The results are reported in Table 8.

The coefficient estimations on all proxies of co-opted directors across Models (1) to (4) remain negative and significant, thus affirming the nonexistence of reverse causation within our framework.

Finally, to ensure our results are robust and support causal interpretation, we use difference in difference approach using CEO exogenous turnover as event. We specifically mapped the co-opted director and climate risk association two-year post CEO exogenous turnover. Our approach is consistent with Bernile et al. (2017), who examine the change in corporate policies following the turnover of exogenous CEOs. We argue that if our hypothesis of board allegiance towards their CEO holds true in the context of climate risk disclosure, then exogenous turnover of CEOs should diminish CEO-board co-option nexus and potentially lead to an improvement in climate risk disclosure. We received the CEO turnover data from Gentry et al. (2021) to identify firms with CEO exogenous turnover and created two groups. Treatment groups consist of firms two-year following CEO exogenous turnover. Control firms refer to the matched firms from the pre-two-year CEO exogenous turnover period. Control firms are matched through PSM using nearest-firm approach with replacement. We then multiply the two-year POST CEO Turnover, which is an indicator variable that takes the value 1 for the two-year period following CEO exogenous turnover and 0 for the pre-two-year CEO exogenous turnover period, with all four proxies of board co-option:

1. Co-option;

2. Co-option (TW);

3. Co-option (Independence); and

4. Co-option (TW Independence).

We performed DiD regression on matched sample and report the result in Table 9.

Panel A reports the univariate results for post-PSM matching. We find a significant increase in climate risk disclosure in companies following two years after CEO turnover, while all other covariates remain qualitatively similar across both groups. This result indicates that the lower number of co-opted directors, resulting from CEO exit, contributes to an improvement in climate risk disclosure, providing initial support for our baseline hypothesis. The DiD regression results reported in Panel B provide confirmatory evidence that an improvement in climate risk disclosure is associated with an exogenous reduction in board co-option. In summary, across all three endogeneity tests, we find qualitatively consistent results with our baseline, suggesting that co-opted directors are significantly negatively associated with climate risk disclosure.

In 2010, the SEC introduced climate risk disclosure guidelines, a significant regulatory step intended to improve transparency in corporate reporting of material climate risks. These guidelines mandated firms to disclose climate-related risks within existing rules but left a critical gap – the determination of “materiality” was entrusted to managerial discretion. This caveat created a duality in the regulatory framework − on the one hand, a mandate for transparency, and on the other, the potential for strategic obfuscation. Such discretion opened the door for boardroom dynamics, particularly the influence of co-opted directors, to play an important role in shaping the disclosure of climate information to stakeholders. Co-opted directors, often aligned with managerial interests (Coles et al., 2014), are uniquely positioned to exploit regulatory ambiguities, potentially withholding disclosures to prioritise managerial objectives over stakeholder needs. While much of the prior literature highlights the drawbacks of co-opted directors, portraying them as lax monitors (Zaman et al., 2021a, 2021b), some scholars have identified potential benefits from their appointment. For example, Nguyen et al. (2021) found that co-opted directors can insulate managers from career concerns associated with innovation risks, fostering an environment conducive to innovative activities. Considering these contrasting traits of co-opted directors and the unique regulatory context created by the SEC’s 2010 climate risk disclosure guidelines, we investigated, using a stakeholder-agency theory framework, whether and how co-opted directors influence climate risk disclosures.

Our findings reveal that co-opted directors lower the overall extent of climate risk disclosure, with the negative association being more pronounced for transition risks compared to physical climate risks. We argue that transition risks, due to their close ties to regulatory changes and geopolitical developments, carry more immediate consequences for managerial accountability and are therefore more likely to be obscured under the influence of co-opted boards. This tendency for suppression aligns with agency theory, as co-opted directors often prioritise managerial interests over stakeholder needs, leveraging the discretionary nature of the SEC’s 2010 climate disclosure guidelines to strategically limit transparency. Informing the “how” aspect of our research question, we find that this relationship is partially mediated by managerial risk-taking – illustrating that such directors enable decisions that prioritise short-term managerial objectives over long-term accountability. While part of our findings corroborates Nguyen et al.’s (2021) observation that co-opted directors exacerbate risk-taking by fostering innovation, we extend the discussion by revealing that this risk-taking can negatively influence climate risk disclosure − a relationship largely overlooked in the literature.

We also recognise that the relationship between co-opted directors and climate risk disclosure is not uniform but rather contingent upon various organisational and industry-specific conditions (Coles et al., 2014; Zaman et al., 2021a, 2021b). To capture this complexity, we extended our analysis to observe how contextual factors shape the influence of co-opted directors on disclosure practices. Our findings reveal that the negative impact of co-opted directors is not uniform but is amplified in specific contexts. In firms characterised by greater linguistic obfuscation (poor internal information environment), the use of complex or vague language compounds the opacity introduced by co-opted directors, making it increasingly difficult for stakeholders to discern critical climate-related risks. This strategic ambiguity further enables the suppression of disclosures, aligning with managerial objectives. Similarly, in firms with weaker external monitoring, the absence of robust oversight mechanisms allows managerial discretion to operate unchecked, exacerbating the climate risk disclosure. Without adequate external checks, co-opted directors can more easily prioritise managerial interests over transparency. Finally, in environmentally sensitive industries, where firms face heightened scrutiny from regulators and stakeholders, the stakes of disclosing climate risks are particularly high. We argue that co-opted directors in these firms may act to minimise such disclosures to avoid reputational damage or regulatory consequences, further reflecting the contingent nature of their influence. These findings highlight the importance of understanding how organisational characteristics and external pressures interact with governance dynamics to shape climate risk disclosure practices, underscoring the need for targeted reforms to mitigate the negative effects of co-opted directors in these contexts.

This study makes several key contributions to the literature. Firstly, it extends the stakeholder-agency framework by revealing how co-opted directors exploit regulatory discretion to exacerbate agency conflicts. Prior studies (e.g. Cassell et al., 2018; Zaman et al., 2021a, 2021b) have demonstrated the adverse effects of co-opted boards on governance outcomes, but our research broadens this understanding by situating these dynamics within the context of mandatory climate disclosures. Unlike previous studies that focus on voluntary reporting (e.g. Ben-Amar et al., 2017; Liao et al., 2015; Lewis et al., 2014; Schneider et al., 2018), our findings demonstrate that even in regulated environments, governance weaknesses such as board co-option can undermine disclosure integrity. Secondly, our research distinguishes between transition and physical risks, offering a nuanced perspective on corporate climate disclosures − a dimension that has been largely overlooked in prior studies (Bolton and Kacperczyk, 2021; Lewis et al., 2014). Transition risks due to their immediacy and regulatory sensitivity are more likely to be strategically obscured by co-opted boards compared to physical risks – which are perceived as long-term and less immediately impactful. This differentiation advances the literature by highlighting the heterogeneity in climate risk disclosure practices. Thirdly, we extend and add depth to recent studies, such as Ghafoor et al. (2023), by exploring the mechanisms through which co-opted boards influence climate risk disclosure. While Ghafoor et al. (2023) focused on the role of co-opted boards in increasing exposure to climate risks, our study extends this line of inquiry by examining the association between co-opted directors and climate risk disclosure. Our findings reveal how co-opted directors actively suppress disclosure, particularly transition risks, which are closely tied to regulatory and geopolitical developments. This suppression is exacerbated in contexts with weak external monitoring, where the absence of robust oversight mechanisms allows managerial discretion to operate unchecked. By exploring mechanisms such as managerial risk-taking, linguistic obfuscation and the role of external monitoring, our research adds significant depth to the discourse on governance and climate accountability, highlighting the nuanced ways in which co-opted directors influence transparency of non-financial disclosure.

Our findings have direct implications for CG practices. The discretionary nature of the 2010 SEC guidelines underscores the urgent need for firms to strengthen board independence to prevent co-opted directors from exploiting regulatory gaps. Specifically, limiting CEO influence over board appointments is a critical step towards mitigating managerial capture and ensuring that directors serve the interests of stakeholders (Zaman, 2024). Furthermore, unlike studies that overlook variations in climate risk disclosure (Farooq, Azantouti, and Zaman, 2024; Tang and Demeritt, 2018; Lewis et al., 2014; Schneider et al., 2018), our findings highlight the heterogeneity in disclosure practices. Transparent reporting of climate risks, particularly transition risks, is vital to align with regulatory expectations, maintain investor confidence and support long-term resilience. Failure to address governance shortcomings risks reputational damage and increased regulatory scrutiny. Practical measures, such as enhancing external monitoring and adopting stricter internal governance policies, are essential to mitigating these risks.

Our findings also have significant implications for policymakers, particularly concerning the limitations of the 2010 SEC guidelines. While these guidelines aimed to enhance transparency, the discretion granted in determining materiality has allowed managerial interests to undermine this objective (Peters and Romi, 2013). Policymakers should consider revising disclosure frameworks to reduce variability in reporting practices and strengthen enforcement mechanisms. Introducing stricter materiality thresholds, mandatory third-party audits of climate risk disclosures and enhanced penalties for non-compliance could help close this regulatory gap. In addition, regulatory measures to limit board co-option − building on prior research indicating that CEO influence in board selection remains prevalent (Carcello et al., 2011) − could include mandating a minimum percentage of truly independent directors and reducing CEO authority over board appointments. Such reforms would help align governance practices more effectively with stakeholder interests, ensuring greater accountability and impartiality in decision-making particularly related to climate risk disclosure. Extending this argument, we propose that such reforms be integrated into global climate disclosure standards, including the SEC’s upcoming 2024 climate disclosure framework, IFRS S1, and IFRS S2, to promote consistency and comparability across jurisdictions. Strengthening governance and regulatory frameworks in this manner will be critical to ensuring that climate risk disclosures meet their intended objectives of enhancing transparency, accountability and resilience in corporate practices.

While our study offers valuable insights, it has limitations. Firstly, the data set is restricted to US companies, which may limit the generalisability of the findings to other regulatory environments. Future research should examine how governance and disclosure practices vary across jurisdictions with different cultural and regulatory contexts. Secondly, while we categorise climate risks into transition and physical types, future research could delve deeper into subcategories − such as regulatory-driven versus market-driven transition risks − to uncover additional nuances. Finally, our analysis ends in 2018, and future studies could explore how subsequent regulatory developments, such as the IFRS standards and the COVID-19 pandemic, have influenced climate risk disclosures.

The authors would like to acknowledge the financial support received by the Edith Cowan University Research Grant (No. G1006499) and the United Arab Emirates University 2023 Research Grant (12B039).