Risk management committee and corporate litigation: evidence from China

In this paper, we examine the association between the existence and the characteristics of a risk management committee (hereafter, RMC) and corporate litigation in China. Corporate litigation generally stems from the divergence between the interests of shareholders and managers, i.e. due to agency problems (Baker & Griffith, 2007; Bhagat, Brickley, & Coles, 1987, Bhagat, Brickley, & Coles, 1994). In addition, other stakeholders, including customers, suppliers, regulators, employees and activist groups, can sue firms for various reasons. Corporate litigation imposes substantial costs on firms, such as settlement payments, legal expenses, managerial distraction and reputational damage, which divert executive attention away from strategic decision-making despite partial financial protection from liability insurance (Adhikari, Agrawal, & Malm, 2019; Wu, Peng, Shan, & Zhang, 2020). Extant empirical evidence suggests that corporate litigation reduces firm investment, increases financial leverage (Malm & Krolikowski, 2017; McTier & Wald, 2011), increases the cost of capital, particularly for lawsuits involving the violations of GAAP (Arena & Ferris, 2017; Chava, Cheng, Huang, & Lobo, 2010), results in changes in top management (Ferris, Jandik, Lawless, & Makhija, 2007), creates governance weaknesses (Chapple & Tan, 2012) and decreases firm performance (Wu et al., 2020).

Given the negative consequences and the high costs of litigation, it is important to understand the factors that can mitigate litigation risk. In this space, the previous studies have found that women in top management (Adhikari et al., 2019) and older CEOs (Malm, Adhikari, Krolikowski, & Sah, 2021) reduce litigation risk. Krishnan, Liu, and Shi (2022) found that audit committee members with financial and industry expertise affect litigation risk. However, despite this growing literature, there is limited evidence on whether specialized board sub-committees, such as RMCs, mitigate litigation risk, a key legal consequence of ineffective risk oversight. We address this paucity in the literature. Our focus on China offers a unique perspective by exploring how the institutional context shapes the effectiveness of firm-level governance mechanisms. While governance scholars recognize the importance of institutions, “how they matter remains a hotly contested question” (Jackson & Deeg, 2008, p. 540). Therefore, the interplay between governance mechanisms, the institutional context and firm-level outcomes demands further attention (Zattoni, Dedoulis, Leventis, & Van Ees, 2020).

Although the board of directors shoulders the responsibility for risk management, there is an ongoing debate about whether having a standalone RMC can be an effective mechanism for controlling a firm's risk (Brown, Steen, & Foreman, 2009; Moore & Brauneis, 2008). An RMC is an important governance committee designed “…to oversee an organization's risk management framework including the process for identifying, assessing, and responding to all current and future risks that threaten an organization's existence, including financial, credit, operational, market, and compliance risks” (Hines, Masli, Mauldin, & Peters, 2015, p. 62). RMCs enhance board risk governance by offering dedicated resources to consistently assess the firm's risk appetite and validate the effectiveness of internal controls related to risk management. More specifically, RMCs ensure that risk management aligns with the firm's strategy by considering major risks, evaluating management assumptions and providing the board with insights on risk management by developing a risk matrix that recommends the risk tolerance levels (Deloitte, 2014) ^[1]. While risk management traditionally has been one of the responsibilities of the audit committee (Daly & Bocchino, 2006), the risk management committee is better suited to perform specific evaluations of the unknown risks facing the organization because of the attributes specified above (Hines & Peters, 2015; Ng, Chong, & Ismail, 2013). Furthermore, the focus on RMCs rather than audit committees is justified, as corporate litigation can arise from both accounting-related and non-accounting-related violations, such as antitrust violations, product liability, environmental breaches and securities fraud (Krishnan et al., 2022).

We use both agency theory and resource dependence theory (hereafter, RDT) to explain the relationship between the existence of RMCs and corporate litigation. According to agency theory, conflicts of interest between management and shareholders can lead to inefficiencies and poor firm-level outcomes (Fama & Jensen, 1983). To mitigate these issues, boards establish sub-committees like the RMC to enhance governance and risk oversight (Daily, Dalton, & Cannella, 2003; Kiel & Nicholson, 2003; Nicholson & Kiel, 2007). Empirical research has generally found a favorable effect of RMCs on various organizational outcomes (Jia & Bradbury, 2021; Malik, Nowland, & Buckby, 2021; Tao & Hutchinson, 2013). RMCs exclusively monitor risk management (Karamanou & Vafeas, 2005; Krishnamurti & Velayutham, 2018; Pirson & Turnbull, 2011) by designing risk policies and controls, which effectively oversee management's risk-related activities. The RMC-induced reduction in information asymmetry through risk disclosures (Al-Hadi, Hasan, & Habib, 2016; Jia & Bradbury, 2021) combined with effective monitoring of risk fosters stakeholder trust, reduces uncertainties and enhances risk management, ultimately decreasing the risk of litigation.

According to RDT, the board of directors plays a key role in shaping and executing firm strategy by connecting the firm to its external environment, which provides access to critical information and resources that reduce uncertainty in decision-making (Salancik & Pfeffer, 1978). We extend this argument to the RMC as a sub-committee on the board. Accordingly, RMC human capital not only monitors managers' risk behavior but also provides access to resources and offers advice and guidance (Hillman & Dalziel, 2003) by improving the level and quality of risk-related disclosures (Ong & Wan, 2008; Salancik & Pfeffer, 1978). Thus, RMCs can mitigate corporate litigation by improving risk governance, reducing information asymmetry, increasing management's awareness of compliance with regulations and reducing agency costs.

In China, a majority of listed firms are state-owned enterprises (SOEs), with the state, its ministries and local governments owning about 30% of the shares on average. Additionally, another 30% of shares are owned by legal entities, which are largely controlled by the state (Firth, Rui, & Wu, 2011). SOEs dominate China’s economic landscape of China and shoulder political, social and economic goals, enabling us to explore how firm-level governance mechanisms, such as RMCs, interact with institutional factors in shaping litigation outcomes. Prior research suggests that evidence on litigation risk from the USA cannot be directly applied to China due to key differences in their legal systems (Wu et al., 2020). The impact of RMCs on litigation risk may also vary between SOEs and non-SOEs, as the controlling state owner often prioritizes political agendas and public benefits over economic performance (Jin, Xu, Xin, & Adhikari, 2022). Although prior research posits that SOEs are inefficient (Allen, Qian, & Qian, 2005), their equity participation significantly enhances the risk-taking capacity of private firms by improving access to resources and alleviating constraints imposed by resource scarcity and underdeveloped legal frameworks (Jin, Yu, Yuan, & Zang, 2024). Thus, the role of RMCs in SOEs requires careful examination.

Using a large sample of listed Chinese firms from 2010 to 2020, we find that the existence of a standalone RMC reduces the number of lawsuits filed against firms. In terms of economic significance, the existence of a standalone RMC reduces the probability of litigation by approximately 24%, which saves the average firm in our sample approximately RMB 2.49 million in litigation costs. We further find that larger RMCs and RMCs with greater human capital are associated with fewer lawsuits. We further find that information asymmetry proxied by analyst forecast dispersion, mediates the negative association between RMCs and the likelihood of litigation. More importantly, we find that the effectiveness of RMCs depends on the institutional context, as our cross-sectional results show that the significant negative association between RMCs and litigation risk is not present in SOEs. To validate the parallel trends assumption and assess the dynamic effects of establishing a voluntary RMC on corporate litigation, we perform a dynamic DID test. To address endogeneity stemming from unobservable heterogeneity, simultaneity and causality effects, we apply the propensity score matching and the generalized method of moments (GMM) technique. Our results remain consistent.

Our study makes several contributions to the literature. First, it extends the limited literature on the role of RMCs in risk oversight and governance. The existing studies primarily focus on RMCs' role in enhancing disclosure quality, financial performance and other financing outcomes (Al-Hadi et al., 2016; Jia & Li, 2022; Jia & Bradbury, 2020; Malik et al., 2021; Ng et al., 2013). However, there is no empirical evidence on how RMCs mitigate litigation risk. In this study, we examine whether the presence and characteristics of RMCs reduce corporate litigation risk, thereby extending the literature beyond informational and financial outcomes to a critical legal and governance dimension. Second, we extend existing literature such as that of de Villiers, Jia, and Li (2022) and Al-Hadi et al. (2016), by demonstrating that RMCs enhance their monitoring role by addressing information asymmetry and leveraging resource dependence. Thus, our study provides insights into how RMCs function to reduce litigation risk, offering an integrated explanation that draws on information asymmetry, monitoring mechanisms and RDT. Third, we extend the corporate governance literature explaining the variations of litigation risk. Prior research mainly focused on governance characteristics such as female directors (Zheng & Wang, 2024) and CEO characteristics (Ting & Lin, 2022; Zhang, Peng, Shan, & Chen, 2023) while we focus on the monitoring role of RMCs. Fourth, we add to the debate over whether some Western governance models can be directly applied in emerging economies (Fiss, Kennedy, & Davis, 2012; Wu et al., 2020). We argue that due to the principal–principal (hereafter, P–P) agency conflicts, political appointments to boards and the lower likelihood of litigation, SOEs leave little room for RMCs to effectively perform their monitoring role, reducing them to mere box-ticking governance mechanisms. Accordingly, our research contributes to the corporate governance literature by demonstrating that the diffusion of Anglo-Saxon governance models into state-dominated institutional environments creates decoupling issues, where these models exist in form but fail to effectively improve governance. Finally, our study offers practical implications for policymakers and regulators in emerging economies by highlighting the substitutive role of state involvement and the challenges of implementing Western governance models in state-dominated environments.

The remainder of the paper proceeds as follows. Section two reviews the literature on RMCs and corporate litigation and develops the hypotheses. Section three describes our research methodology. Section four reports the empirical results. Section five concludes the study.

RMC oversees a broad range of risks, including financial, operational and environmental risks. Risk oversight involves evaluating the company's risk management tools, rules and culture; regularly identifying significant or emerging risks; and providing feedback to minimize their potential impact. The role of the RMC includes risk oversight, setting risk appetite and tolerance levels, reviewing risk policies, reporting and communicating risk-related information, and enhancing compliance and regulatory adherence (Al-Hadi et al., 2018; Bugalla, Kallman, Lindo, & Narvaez, 2012; Deloitte, 2014). RMCs enhance risk awareness and encourage actions to minimize risks by communicating policy violations, internal control deficiencies and compliance issues (Malik et al., 2021). By performing these functions, RMCs help firms keep their risk profile within acceptable levels and decrease the risk of information asymmetry (Bugalla et al., 2012), hence protecting firm reputation (Magee, Schilling, & Sheedy, 2019). As a sub-committee within the board, the RMC has a fiduciary duty to monitor managerial behavior to ensure that appropriate risks are undertaken in the best interest of shareholders.

Extant research on RMCs has mainly focused on firm-level financial and disclosure outcomes. For example, Jia and Bradbury (2020) find that voluntarily formed RMCs are associated with improved firm performance. Further extending this line of research, Jia and Bradbury (2021) find that RMCs increase firm value, while Wu, Kweh, Lu, and Azizan (2016) find that they increase firm efficiency. Furthermore, firms with RMCs increase risk disclosures (Al-Hadi et al., 2016; Jia, Li, & Munro, 2019) with better readability (Jia & Li, 2022), which reduces information asymmetry, as a result, allows firms to enjoy capital market benefits such as better credit ratings (Ames, Hines, & Sankara, 2018), lower financial constraints (Malik et al., 2021) and a lower cost of equity capital (Al-Hadi et al., 2018). Others find that RMCs reduce information asymmetry, financial constraint risk and underwriting risk (Malik et al., 2021; Tao & Hutchinson, 2013). While these studies provide important insights into the informational and financing benefits of RMCs, they largely overlook legal risk outcomes, particularly firms' exposure to litigation, which represents a critical consequence of ineffective risk oversight. Given that litigation risk arises from failures in monitoring, disclosure and compliance, it provides a direct and economically meaningful setting to assess the effectiveness of RMCs. Our study addresses this gap by examining whether the presence of RMCs reduces litigation risk, thereby extending the RMC literature beyond financial and disclosure outcomes to include a legal and governance consequence. In addition, we extend the corporate governance and RMC literature by examining how effective committee characteristics such as size, independence and human capital contribute to positive firm-level outcomes (Jia & Bradbury, 2020).

Proponents of a dedicated RMC further argue that the traditional model of the combined audit committee and RMC overlooks the effective monitoring of risk-taking in other areas due to its primary focus on financial reporting processes and audit oversight (Bates & Leclerc, 2009; Hines et al., 2015; Iselin, 2020; Malik et al., 2021). Time constraints and fatigue are more prevalent in combined committees, which can reduce the committee members' motivation and capacity to conduct thorough risk reviews (Subramaniam, McManus, & Zhang, 2009). Opponents of forming a standalone RMC, however, argue that having a standalone RMC adds another layer of bureaucracy and compromises efficiency (Bates & Leclerc, 2009). Eggleston and Ware (2009, p. 55) point out that the benefits of having a separate RMC are contingent upon how well the RMC is designed and how effectively its functions are executed, otherwise, it can do more harm than good. Our study contributes to this debate by empirically examining whether the establishment of a standalone RMC enhances firm outcomes, specifically focusing on litigation risk.

Research on corporate litigation has primarily focused on shareholder litigation, such as securities class actions and derivative lawsuits, mainly in the USA (Wu et al., 2020). Arena and Ferris (2017) argue that corporate laws in civil law countries like China do not facilitate class action litigation as readily as US corporate law ^[2]. Additionally, Dash and Raithatha (2018) suggest that securities class action lawsuit procedures are generally unsuitable for most emerging economies due to their unique regulatory frameworks and varying levels of development. Our research includes various types of lawsuits such as criminal proceedings, arbitration, civil litigation and administrative litigation. We investigate lawsuits filed by stakeholders for issues such as contract breaches and regulatory violations. For example, the case between Alibaba and JD.com in China highlights corporate litigation between competitors, where Alibaba was found to have “abused its market dominance” by the Beijing High People's Court. Conversely, firms themselves may act as plaintiffs, suing for contract breaches or other legal violations or initiating legal action against other firms ^[3]. Tao (2012) notes that corporate litigation in China mainly involves civil litigation and can be classified into five categories: contract disputes, insurance disputes, loan and labor disputes, land and intellectual property disputes and shareholder disputes. For many listed companies in China, litigation expenses have surpassed the previous year's net profit and even operating income, significantly threatening their continued operations (Wu et al., 2020). This highlights that litigation has increasingly become a critical risk factor for Chinese firms, especially those involved as defendants (Bhagat, Bizjak, & Coles, 1998; Wu et al., 2020).

A few studies have examined the governance-related determinants of litigation risk in China. For example, Zheng and Wang (2024) argue that women directors are more risk-averse and as a result, firms with gender-diverse boards are less likely to be sued. Zhang et al. (2023) posit that CEO social capital reduces litigation losses because CEOs with high social capital have an incentive to reduce the risk of litigation due to reputation concerns. They further find that the CEO effect is not present in SOEs. Ting and Lin (2022) find that CEOs with foreign experience are associated with fewer lawsuits in China. Evidence also exists to suggest that geographical proximity to local government agencies (Lu, Sun, Tian, & Zhao, 2024), bankruptcy judicial system reforms (Lai & Hu, 2024), the existence of judicial local protection (Li, 2024) and stock market liberalization (Xiong, Deng, & Xiao, 2021) affect litigation risk in China.

Agency theory posits that conflicts of interest between management and shareholders can lead to inefficiencies and poor performance (Fama & Jensen, 1983; Jensen & Meckling, 1976). Internal governance mechanisms, such as the board of directors, monitor managers and guide strategic direction (Hillman & Dalziel, 2003), ensuring the alignment of interests between shareholders and managers (Fama & Jensen, 1983). To enhance their monitoring and stewardship functions, boards establish sub-committees tasked with specific responsibilities (Daily et al., 2003; Kiel & Nicholson, 2003; Nicholson & Kiel, 2007). Regulators, policymakers and other stakeholders have increasingly focused on improving the governance, management and reporting of financial and business risks (Brown et al., 2009). Prior research (Jia & Bradbury, 2021; Jia et al., 2019; Malik et al., 2021) has emphasized the importance of having a standalone RMC. As the RMC is mainly responsible for designing and monitoring risk management policies and frameworks, we expect that the RMC will reduce the likelihood of corporate litigation.

The function of the RMC is to oversee, advise on and enhance risk management and risk information disclosure to ensure enough awareness of all types of risks (Hines et al., 2015; Jia et al., 2019). From an agency theory perspective, forming a separate RMC strengthens firm governance as the RMC acts on behalf of shareholders and monitors management (Hillman & Dalziel, 2003). Moreover, a standalone RMC exclusively focuses on risks and thus has the time and capacity (Karamanou & Vafeas, 2005; Krishnamurti & Velayutham, 2018; Pirson & Turnbull, 2011) to better monitor managers' risk-taking actions that could trigger litigation. In addition, the existence of an RMC relieves the board of the responsibility for monitoring on the firm's exposure to risks (Bates & Leclerc, 2009), thereby, ensuring the quality of a firm's risk management (Cai, Cui, & Jo, 2016; Ding & Wei, 2023). The RMC is likely to ensure that a firm has a comprehensive risk management framework in place and that it is being effectively implemented by providing recommendations to the board on risk appetite and tolerance levels (Aebi, Sabato, & Schmid, 2012; Brown et al., 2009; Florio & Leoni, 2017; Iselin, 2020; Magee et al., 2019). We, thus, argue from an agency perspective that when firms establish RMCs, corporate risk governance improves due to the presence of a standalone body that is well-designed and structured, which can reduce both financial and non-financial risks and consequently the risk of litigation (Brancato, Tonello, Hexter, & Newman, 2009; Kirkpatrick, 2009). In conclusion, the dedicated RMC's exclusive focus on risk management can enhance risk oversight and accountability, and improve risk communication, which results in a lower likelihood of litigation by stakeholders. Therefore, drawing insights from agency theory and its monitoring perspective, we develop the following hypothesis.

Drawing from the RDT, prior research (Hillman, Cannella, & Paetzold, 2000; Hillman & Dalziel, 2003) argues that larger boards are better equipped to access human capital, knowledge and external networks, which facilitate effective monitoring and provide valuable advice and counsel to senior management. We extend this resource-based perspective from boards to their sub-committee, the RMCs, which plays a specialized governance role by focusing on risk oversight. Like larger boards, larger RMCs are expected to bring together diverse experts and allocate responsibilities more effectively, ensuring that each member has sufficient time and expertise to monitor the firm's risks. As a result, larger RMCs are better equipped to design, implement and monitor robust risk management strategies that protect the interests of shareholders and other stakeholders, thereby reducing the likelihood of litigation. Additionally, the size of the RMC signals the board's commitment to investing resources in risk management oversight (Hines et al., 2015). Extant empirical evidence supports this argument, showing that larger RMCs enhance environmental performance (de Villiers et al., 2022) and strengthen risk governance (Magee et al., 2019; Malik et al., 2021). The larger the RMC, the stronger the firm's risk governance and the lower the level of information asymmetry (Malik et al., 2021), which further mitigates litigation risk. Therefore, we develop the following hypothesis.

Drawing on agency theory and the reputation hypothesis, we argue that independent directors affect the likelihood of corporate litigation. First, we argue that independent directors' monitoring role impacts corporate litigation through effective oversight and through their role in reducing information asymmetry (Kassinis & Vafeas, 2002). Practitioners and regulators emphasize that the independence of RMCs is just as important as the independence of the board (Hines et al., 2015). Increased information disclosure and transparency in firms with highly independent RMCs result in a lower likelihood of litigation because these firms are better able to identify and mitigate potential risks. Second, independent directors on the RMC are motivated to minimize the likelihood of corporate litigation due to their reputational concerns. Fich and Shivdasani (2007) find that independent directors face significant reputational harm if they are linked to a securities class action lawsuit. Similarly, Masulis and Mobbs (2014) reveal that firms with independent directors who have strong reputational incentives are less likely to become targets of class action lawsuits. While both studies focus on securities lawsuits, their arguments extend to other types of litigation, indicating that independent directors can suffer reputational damage from any legal action against the firms they serve. Thus, independent RMCs are likely to reduce the likelihood of litigation.

Prior research on board independence in China has also documented that the presence of a larger percentage of independent board members increases firm profitability (Jiang, Yue, & Zhao, 2009; Liu, Miletkov, Wei, & Yang, 2015) and increases the probability of replacing poorly performing CEOs (Conyon & He, 2011). Using independent-director rankings as a proxy for board hierarchy, Zhu, Ye, Tucker, and Chan (2016) documented a positive association between independent-director rankings and firm value. Importantly, the authors find that the likelihood of an independent director voting against management significantly increases with that director's ranking within the board. Evidence also suggests that independent directors in China can mitigate the tunneling problem (Liu, Wang, & Wu, 2016) that stems from the P–P agency conflict. If independent directors are effective monitors from an agency theory perspective in the context of China, then we can predict the following:

Hillman and Dalziel (2003) introduced the concept of board human capital, which can be developed through training, education and experience. Following prior research (De Maere, Jorissen, & Uhlaner, 2014), we identified directors' educational backgrounds, qualifications and experience as key factors that form the foundation of RMC human capital. Specifically, directors' education (Dalziel, Gentry, & Bowerman, 2011) and industry experience (Chen, 2014) are known to influence strategic decision-making. Knowledgeable and experienced directors are better equipped to identify, evaluate and manage potential risks proactively. They can help implement sound governance practices by improving transparency and reducing information asymmetry, which is a known contributor to litigation (Healy & Palepu, 2001). As such, human capital within the RMC can serve as a key safeguard against legal disputes by enhancing compliance and ensuring timely risk disclosures.

Directors with accounting expertise are especially valuable in this context. Accounting professionals excel at assessing events related to legal risks, such as lawsuits or warranties (Krishnan & Visvanathan, 2008). Additionally, studies suggest that directors with accounting backgrounds demonstrate strong cognitive ability (Godart, Maddux, Shipilov, & Galinsky, 2015; Von Meyerinck, Oesch, & Schmid, 2016), thereby improving reporting timeliness (Alkebsee, Habib, Huang, & Tian, 2022) and reducing the likelihood of litigation (Krishnan et al., 2022). Similarly, RMCs with financial expertise are more effective at disclosing market risks (Al-Hadi et al., 2016) and addressing financial constraints (Malik et al., 2021). These qualities are essential to risk governance and help prevent legal liabilities by minimizing misreporting and non-compliance. The presence of academic directors further strengthens the RMC's ability to monitor management and provide strategic advice. Academics possess the capability to process large amounts of complex information (Audretsch & Lehmann, 2006; White, Woidtke, Black, & Schweitzer, 2014). They also enhance the advisory role of boards in complex firms, often suggesting innovative solutions (Anderson, Reeb, Upadhyay, & Zhao, 2011). Additionally, academic directors can facilitate knowledge transfer and improve firms' access to external resources through their extensive professional and social networks (Güner, Malmendier, & Tate, 2008; Lynall, Golden, & Hillman, 2003). Expert advice and monitoring from legal experts are also crucial in deterring litigation risk. Directors with legal expertise are more attuned to legal risks (Krishnan, Wen, & Zhao, 2011) and can detect potential issues at an early stage. Their specialized knowledge and resources enable them to effectively manage legal risks and appropriately handle litigation. Moreover, legal experts play a crucial role in advising and guiding management on the strategic application of legal tactics in business operations (Liu & Sun, 2021). Accordingly, drawing on RDT and agency theory, we propose the following:

Increased levels of disclosure reduce information asymmetry, which eventually mitigates agency frictions between managers and external parties (Karamanou & Vafeas, 2005). Empirically, risk disclosures have been found to reduce information asymmetry and agency costs (Campbell, Chen, Dhaliwal, Lu, & Steele, 2014; Deumes & Knechel, 2008; Elshandidy, Neri, & Guo, 2018) and the existence of a separate RMC leads to high-quality risk information disclosure (Buckby, Gallery, & Ma, 2015; Jia et al., 2019). Companies with a dedicated risk committee tend to make more comprehensive market risk disclosures in the Gulf Cooperation Council (GCC) countries (Al-Hadi et al., 2016).

An RMC that prioritizes comprehensive risk disclosure signals to the market that the firm has robust mechanisms in place to identify and manage risks. This proactive approach reduces the probability of unexpected adverse events that could trigger legal action. By reducing information asymmetry, risk disclosures ensure that all stakeholders, including investors and regulatory bodies, have access to accurate and timely information about the firm's risk exposure (Easley & O'hara, 2004). Disclosure on risk management is expected to help readers to “fully assess public companies and their associated risks” (Buckby et al., 2015, p. 813). This transparency regarding risks decreases the uncertainty and the potential for disputes over undisclosed risks, thereby lowering the risk of litigation. Transparent risk disclosures help lower the cost of capital by building trust among investors and stakeholders (Solomon, Solomon, Norton, & Joseph, 2000). When stakeholders perceive that a firm is managing its risks effectively and disclosing pertinent information, they are less likely to pursue litigation as a means of addressing grievances. Accordingly, we argue that the RMC has a favorable effect on reducing information asymmetry between managers and shareholders and other stakeholders, which is expected to mitigate the probability of litigation. We therefore hypothesize as follows:

Given the significant involvement of SOEs in China, it is crucial to investigate how RMCs function within SOEs and their impact on corporate litigation. In the Chinese context, the Anglo-American principal–agent model – which focuses on conflicts between managers and shareholders – may not fully explain litigation risks in Chinese SOEs. Instead, P–P agency conflicts between the state (as a controlling shareholder) and minority shareholders take precedence (Sun, Yuan, Cao, & Wang, 2017). Unlike non-SOEs, where internal governance tools such as RMCs are essential for risk oversight and mitigating litigation risk (Al-Hadi et al., 2016; Malik et al., 2021), SOEs rely more on state influence and external protections to manage risks due to their access to policy insights, government backing and essential resources (Zhou, Gao, & Zhao, 2017). We argue that this state influence in SOEs dilutes the role of internal governance mechanisms such as RMCs. In SOEs, P-P conflicts arise when the state, as a controlling shareholder, expropriates resources from minority shareholders (i.e. tunneling) to pursue political goals rather than shareholder interests (Gan, Guo, & Xu, 2018; Li, 1997; Lin, Cai, & Li, 1998). This focus on political objectives over economic performance reduces the priority given to risk management and oversight activities. As a result, RMCs in SOEs face resource constraints and are less effective at fulfilling their monitoring roles. Additionally, political appointments to SOE boards – often driven by nepotism and cronyism – further weaken governance. SOEs tend to appoint directors with political affiliations rather than those with relevant qualifications and expertise (Fan, Wong, & Zhang, 2007; Jiang & Kim, 2015, 2020). This lack of qualified directors limits the RMC's ability to monitor effectively and provide strategic counsel (Lin, Lu, Zhang, & Zheng, 2020). The state-appointed directors' independence is also questionable, as their incentives are often aligned with the state's political and social goals (Park, Li, & Tse, 2006). Consequently, SOE managers prioritize state interests over economic outcomes, weakening the governance improvements that RMCs might offer.

Furthermore, SOEs are less likely to face litigation from stakeholders due to the selective enforcement of the law against SOEs which provides disincentives for minority shareholders and other stakeholders to initiate litigation actions (Chen, Jiang, Liang, & Wang, 2011; Firth et al., 2011). Prior research demonstrates that selective law enforcement in China results in more lenient penalties for SOEs due to their political significance, granting them privileges from supervision and judiciary agencies, which lead to more favorable trial outcomes and better appeal results compared to non-SOEs (Chen et al., 2011). Firth et al. (2011) also highlight that the state has significant influence over the legal system, suggesting that courts may often rule in favor of SOEs. As a result, plaintiffs might react less negatively to lawsuits against SOEs (as defendants) or may be less likely to file lawsuits against them altogether. In conclusion, SOEs tend to have a lower likelihood of facing corporate litigation than their non-SOE counterparts.

Figure 1 presents the conceptual framework, illustrating the relationship between RMCs and litigation risk, along with the mediating role of information asymmetry and the moderating role of the SOE institutional context.

We begin with an initial sample of 34,332 firm-year observations over the period 2010 to 2020. Our sample period started in 2010, when regulators and practitioners started encouraging Chinese public firms to establish a standalone and/or integrated risk committee after the 2008 financial crisis. We remove observations with missing data on corporate litigation (6,501 observations) ^[4], RMC (2,065 observations), and control variables (4,028 observations). After using these filters, we end up with a final sample of 21,738 firm-year observations. Panel A of Table 1 shows the sample selection procedure.

We manually collected RMC data from the China Stock Market and Accounting Research (CSMAR) database. First, we retrieved the “Committee members” dataset from the corporate governance sub-database within CSMAR. This dataset contains comprehensive information about all committee members, including their names, IDs, committee affiliations and other details. Simultaneously, we acquired additional director information from the CSMAR. Next, we extracted the names of RMC members from the “RMC members” file and compiled all relevant data for each member. Finally, we calculated the average and counted the number of members. A firm is assigned to have a standalone RMC if the committee's name has the word “Risk committee or risk management committee.” We were careful in excluding all observations with the “Audit and Risk committee” or “any other combined committee with RMC”. Corporate litigation, financial and corporate governance data are derived from the CSMAR database. We winsorize the continuous variables at the top and bottom 1% of their respective distributions. Panel B of Table 1 reports the distribution of our sample across the industry. The Manufacturing industry represents the majority of our sample (64.9%), followed by the Information transmission, software and information technology services (7.6%) and the wholesale and retail business industries (5.3%).

Our dependent variable is corporate litigation (LAWSUIT) ^[5]. Unlike other studies that focus on security class-action lawsuits (Arena, 2018; Freund, Nguyen, & Phan, 2023; Hassan, Houston, & Karim, 2021), we examine various civil lawsuits and criminal proceedings where firms were named as defendants. Because we are motivated by the notion that defendant firms are likely to face a stronger negative market reaction from stockholders than plaintiff firms, who are generally perceived as victims (Liu, Si, & Miao, 2022), we excluded plaintiff firms from our sample. We then counted the number of lawsuits for each defendant firm during the sample period. Following prior studies on corporate litigation (e.g. Liu, Aharony, Richardson, & Yawson, 2016; Qin, Yang, He, & Sun, 2021; Xiong et al., 2021), we measure LAWSUIT as the total number of lawsuits filed against a firm. Additionally, we used SUIT_COST, defined as the natural logarithm of one plus the litigation costs incurred by the defendant firm, as an additional proxy for corporate litigation.

RMC is a dummy variable coded one if a firm forms a standalone RMC, and zero otherwise. This proxy enables us to employ a large sample and compare firms with and without a standalone RMC and its association with corporate litigation. We also use several RMC characteristics to test whether such characteristics are related to corporate litigation: RMCSIZE is the total number of directors on the RMC, RMCIND is the proportion of independent directors on the RMC, RMC_HC is the human capital aspect of the RMC index, measured as the proportion of directors on the RMC (1) who are working as a teacher in university, college or institutions, (2) who has previous work experience in risk management, audit, legal department, (3) who has financial expertise. RMC_LEGEX is a dummy variable coded 1 if there is at least one member on the RMC who has legal expertise and 0 otherwise. Although this is a component of RMC human capital, we measure it as a standalone variable because of its categorical nature.

Following prior studies, we controlled for factors that were found to influence corporate litigation. We include the existence of audit committees to control for the fact that audit committees may also oversee risk management functions (Krishnan et al., 2022) and hence can be related to corporate litigation. Therefore, we include a dummy variable (AC) coded 1 if a firm has an audit committee, and 0 otherwise. We also control for several proxies of corporate governance that influence corporate litigation (Haslem, 2005; Kassinis & Vafeas, 2002; Krishnan et al., 2022): board independence (BIND), measured as the proportion of independent directors on the board; CEO duality (DUAL), coded 1 if the CEO is also the chairman of the board, and 0 otherwise; and board size (BSIZE), measured as the number of directors on the board. Additionally, we control for several financial variables related to corporate litigation. Larger firms are more likely to be sued than smaller firms (Dechow, Ge, Larson, & Sloan, 2011; Gande & Lewis, 2009). Thus, we include firm size (SIZE) defined as the natural log of total assets. The literature documents that firms with good performance are less likely to be sued (Crutchley, Jensen, & Marshall, 2007). Thus, we include book-to-market ratio (BTM) defined as total book value scaled by total market value with a high BTM proxying for fewer growth opportunities, less risk and hence lower risk of litigation; firm profitability (ROA), measured as net profit scaled by the total assets; and LOSS, measured as a dummy variable coded 1 if a firm incurred a loss, and 0 otherwise. Firms with higher leverage are more prone to lawsuits (Ashbaugh-Skaife, Collins, & LaFond, 2006); therefore, we include firm leverage (DEBT) measured as total debts scaled by total assets. Prior studies have documented that financial constraints and audit quality are related to corporate litigation (Mezzanotti, 2021; Xiong et al., 2021); thus, we include financial constraints (FCS) coded as 1 if a firm is financially constrained, and zero otherwise ^[6] and BIG4 defined as a dummy variable coded 1 if a firm is audited by a Big Four audit firm, and 0 otherwise.

To test the relationship between the existence of RMC and its characteristics and corporate litigation, we use the following Poisson regression model:

The Poisson model is appropriate because our dependent variable is a count variable that takes non-negative integer values and exhibits a skewed distribution. Accordingly, the Poisson specification allows us to estimate the expected number of litigation events as a function of the existence of the RMC and its characteristics. A negative and significant coefficient on β₁ will support H1. We replaced RMC with RMCSIZE, RMCIND, RMC_HC and RMC_LEGEX to test H2a-H2c. In all the regression specifications, we include Industry and Year fixed effects and cluster standard error at the firm level.

Table 2 shows the descriptive statistics for the observations with and without a standalone RMC. The average (median) standalone RMC is 3.4% (0), suggesting that less than 4% of our sample firms voluntarily form a standalone RMC. Malik et al. (2021), using data from US firms spanning the period 2005–2017, report an average RMC rate of 2%, which is close to that reported in our study. The average (median) number of lawsuits is 0.506 (0.00). Our lawsuit statistics are consistent with those reported by Ting and Lin (2022), who used Chinese data over the period 2008–2018 and reported an average (median) number of lawsuits of 0.462 (zero). Approximately 98.3% of our sample firms have an audit committee. The average RMC size is about 6 members. Around 32.8% of RMC members are independent. The proportion of RMC human capital (RMC_HC) is around 27.3% and about 64.4% of the RMC members have legal expertise. The average board size is around nine members, with 37.5% of the board members being independent. Around 20.1% of our sample firms have a CEO who is also the chairman of the board. The average firm size is 22.34 with a BTM value of 0.616, an ROA of 4.7% and 9.5% of the sample firms reporting a loss in their financial statements. The mean leverage ratio is 0.45. Almost 53.4% of our sample firms are financially constrained. Only 3% of our sample firms were audited by one of the Big Four audit firms, which can be attributed to the dominance of the Chinese audit market by the top 10 audit firms. Around 34.4% of the observations are SOEs. Finally, the average risk disclosure (RD), our proxy for information asymmetry (ASY), is 0.010.

Table 3 shows the Pearson correlation analysis. The correlation coefficient between RMC and LAWSUIT is negative and significant (correlation −0.019, p < 0.01). The correlations between the LAWSUIT and the control variables are consistent with prior literature. For instance, firms with good performance and more risk-related disclosures are likely to have fewer lawsuits (Rogers & Van Buskirk, 2009; Wu et al., 2020), whereas firms with financial constraints are more likely to be sued (Arena, 2018). Because none of the correlation coefficients between the independent variables exceed 0.60, multicollinearity in our model is not a concern (Belsley, 1991). In addition, the highest variance inflation factor (VIF) is 2.06, which is well below the commonly used threshold of 10.00.

Column 1 of Table 4 presents the Poisson regression results for Eq. (1). The coefficient on RMC is negative and significant (coefficient = −0.120, p < 0.05), implying that firms that voluntarily establish a standalone RMC are less likely to be sued by firms or individuals, thereby supporting H1. Given that our independent variable is binary, we calculate the percent decrease in corporate litigation as follows: −0.120 (the coefficient on RMC)/0.506 (mean of LAWSUIT) = −23.7%. Further, we perform a panel Poisson regression with firm fixed effects (FFE) to reduce the effects of time-invariant firm characteristics. Column (2) reports the result of the FFE model. The coefficient on RMC remains negative and significant (coefficient = −0.208, p < 0.01). Theoretically, these findings support the agency theory view that firms' boards that voluntarily set a separate RMC improve risk-governance mechanisms, reduce agency costs and decrease information asymmetry related to risk, which in turn, mitigates corporate litigation. Our results are consistent with extant studies that document a beneficial role of the existence of an RMC (Al-Hadi et al., 2016, 2018; Hines et al., 2015; Jia & Li, 2022; Malik et al., 2021). Regarding control variables, we find that the number of lawsuits increases for firms with poor performance (LOSS) and high leverage (DEBT). The number of lawsuits decreases for large firms (SIZE) and firms with high BTM and good performance (high ROA).

We additionally run our main regression model using SUIT_COST as an additional litigation variable. Column 3 of Table 4 reports the results of this analysis. The coefficient on RMC is negative and highly significant (coefficient = −0.644, p < 0.05) ^[7]. These findings indicate that our baseline results are robust and consistent across different proxies for corporate litigation.

Panel A of Table 5 shows the results of the association between RMC characteristics and corporate litigation: testing hypotheses H2a-H2c. In Column (1), the coefficient of RMCSIZE is negative and significant (coefficient = −0.203, p < 0.01), suggesting that the larger the RMC, the less likely the firm is to be subject to litigation. This therefore supports H2a. The coefficient on RMCIND is negative and significant (coefficient = −0.553, p < 0.05), implying that the probability of litigation decreases as the proportion of independent members on the RMC increases. H2b, therefore, is also supported. The coefficient of RMC_HC is positive and insignificant. Accordingly, H2c is not supported. Finally, the coefficient on RMC_LEGEX is also negative and significant (coefficient = −1.585, p < 0.05). In addition, we perform a panel Poisson regression with firm fixed effects. Column (2) of Table 5 presents these results. The coefficients of RMCSIZE, RMCIND and RMC_LEGEX remain negative and significant. This suggests that risk committee members with legal expertise can ensure a firm's legal compliance, thereby reducing the probability of litigation risk. For example, RMCs with legal expertise may help firms avoid lawsuits by either solving disputes before a lawsuit is filed or negotiating lawsuits filed against a firm and reaching lower-cost settlements. Our findings are generally consistent with RDT and agency theory, which propose that large, and qualified RMCs are likely to identify, evaluate and manage potential risks proactively, thereby reducing information asymmetry and hence the risk of litigation. In addition, we regress the RMC characteristics on SUIT_COST (Column 3). The results show that the impact of the RMC characteristics' impact remains robust and consistent.

In addition, Panel B of Table 5 reports the results for the characteristics of the RMC in a parsimonious specification with fewer control variables to check the robustness of the results given that we used 17 covariates for only 730 observations. Column (1) presents the baseline model without control variables, while Columns (2) and (3) incorporate alternative sets of firm-level and governance control variables. The coefficients for RMCSIZE, RMCIND, and RMC_LEGEX are all negative and statistically significant across all three models. Overall, the consistency of the main results across alternative specifications provides assurance that the findings are not sensitive to model specification or the inclusion of additional control variables.

To explore the mediating effect of information asymmetry (H3), we formulate the following regression specifications:

We use analyst forecast dispersion (ASY) as the mediating variable, which captures disagreement among analysts about a firm's future earnings and is a well-established proxy for information asymmetry (Barron & Stuerke, 1998). We estimate the mediation using a three-step regression system. First, we replicate Equation (1) using a smaller sample, as there are many missing values for ASY. Second, we regress ASY on RMC and all control variables (Eq. 2). Third, we regress LAWSUIT on RMC, ASY, and all control variables (Eq. 3). To test the significance of the indirect effect, we use a bootstrapped approach.

We present the empirical estimates of the mediation analysis in Table 6. As shown in Columns (2) and (3), RMC significantly reduces ASY (coefficient = −0.295, p < 0.01) and ASY, in turn, significantly increases the number of lawsuits (coefficient = 0.300, p < 0.01). The direct effect of RMC on LAWSUIT becomes insignificant once ASY is included (coefficient = −0.111, insignificant). These results support full mediation. The bootstrapped indirect effect of −0.295 (z = −4.16, p < 0.01) confirms that the mediated pathway is statistically robust. We address endogeneity concerns by using lagged analyst forecast dispersion as an instrument and re-run the mediation model. We present these results in Columns (4) and (5). We find that our instrument is valid as lagged dispersion (L.ASY) is a valid instrument because it is strongly correlated with current dispersion (z = 1.865, t = 25.20). The full mediation finding is therefore substantiated. Furthermore, we re-ran the mediation analysis after replacing current-period dispersion with lagged dispersion directly in the mediation model and the results support full mediation (results untabulated). Since full mediation assertion is consistent across all specifications, we suggest that information asymmetry mediates the association between RMC and LAWSUITS, thereby supporting H3.

To explore the moderating effects of SOE status on the relationship between the RMC and corporate litigation, we use both a split-sample approach and an interactive regression approach. Panel A of Table 7 reports the regression results for the standalone RMC while Panel B of Table 7 reports the regression results for the RMC characteristics. The coefficient of RMC is negative but insignificant for the SOE sub-sample in Column (1), while it is negative and significant for the non-SOE group (coefficient = −0.598, p < 0.01) in Column (2), suggesting that the RMC is more effective in reducing the risk of lawsuits for the non-SOE firms than for their SOE counterparts. For the interactive model, we find that the coefficient on the interactive variable RMC*SOE positive and significant (Column 3). Evaluating the combined effect of RMC and the interaction term reveals that the net effect of RMC for SOEs is close to zero (−0.598 + 0.570). Thus, while RMCs play an important role in reducing lawsuit risk among non-SOEs, such risk governance benefits largely disappear in SOEs. This finding suggests that state ownership compromises the risk governance effectiveness of the RMC. This supports H3.

In panel B of Table 7, we find that the coefficients of the RMC characteristics, namely RMCIND and RMC_LEGEXPER, are insignificant for SOEs but negative and significant for non-SOEs (coefficient = −2.951, and −1.852, p < 0.01, respectively). The coefficient of RMCSIZE is negative and significant for both SOEs and non-SOEs. Taken together, the findings support H4. Generally, our results show that the state ownership compromises the role of the RMC in mitigating litigation risk, i.e. the RMC is more likely to enhance risk management and reduce the risk of litigation for non-SOE firms than for SOEs.

We further examine whether overlapping membership between the RMC and the audit committee affects our main results. We include an additional variable capturing overlapping membership between the two committees (RMC_AC). Panel C of Table 7 reports the estimates for this analysis. The results show that the coefficient on RMC_AC is insignificant, while the coefficient on RMC remains negative and significant. These findings suggest that our main results are robust after controlling for overlapping committee membership.

To establish a causal relationship between establishing a voluntary RMC and the subsequent reduction in corporate litigation, we perform a dynamic DID test. This approach evaluates whether pre-treatment trends are similar between the treatment and control groups and tracks the temporal evolution of the policy (RMC formation in our setting) effect. To perform the dynamic DID analysis, we construct seven event-time indicators to capture trends around the year of the voluntary establishment of the RMC: BEFORE(−1), BEFORE(−2), BEFORE(−3), CURRENT, AFTER(+3), AFTER(+2), and AFTER(+1) for the six years around the establishment of the RMC event. BEFORE (−1) refers to one year before establishing an RMC and is coded as one if the firm-year observations belong to the year before establishing an RMC, and zero otherwise. BEFORE (−2) is coded as one if the firm-year observations belong to two years before establishing an RMC and zero otherwise. BEFORE (−3) is defined similarly. CURRENT refers to the year of the establishment of an RMC. AFTER (+1) refers to one year after establishing an RMC and is coded as one if the firm-year observations fall one year after establishing an RMC, and zero otherwise. AFTER (+2) and AFTER (+3) are defined analogously. To validate our baseline results (the parallel trend assumption), the coefficients on the AFTER indicators should be negative and significant, while the coefficient on the BEFORE indicators should be insignificant. Table 8 presents the results of the dynamic DID test.

The results reveal that the coefficients on BEFOREs (−1, and −2) are positive and significant, while the coefficients on BEFORE (−3) and CURRENT are insignificant. On the other hand, coefficients on AFTERs (+2, and +3) are negative and significant, suggesting that establishing a standalone RMC reduces corporate litigation lawsuits. However, the significant positive coefficients in some pre-treatment periods indicate the presence of pre-existing differences between the treatment and control groups. Therefore, the DID results should be interpreted with caution, as the parallel trend assumption may not be fully satisfied. Nevertheless, the negative and significant post-treatment coefficients provide evidence consistent with the view that standalone RMCs are associated with lower corporate litigation risk.

Our baseline results could be subject to endogeneity issues associated with unobservable heterogeneity, simultaneity and causality effects. According to Greene (2003), the efficiency of the model used to address endogeneity problems must be considered when selecting a suitable approach. We find that applying the GMM approach is a suitable and powerful method for dealing with endogeneity problems. The GMM approach addresses endogeneity problems that arise from simultaneity biases (Arellano & Bond, 1991; Arellano & Bover, 1995), i.e. when the dependent and independent variables may affect each other. In our case, firms that are sued may be more likely to establish a standalone RMC to mitigate corporate litigation.

The GMM approach addresses concerns related to unobservable heterogeneity concern by using instrumental variables. We used Arellano and Bond's (1991) autocorrelation tests: AR(1) and AR(2) to detect the dynamic specifications of LAWSUIT and RMC. Particularly, to check for autocorrelation in the differenced residuals of the GMM model, we use and report the AR(2) test, with the null hypothesis of “no serial correlation”. We are particularly interested in AR(2) because AR(1) usually rejects the null hypothesis (Roodman, 2009). The Sargan and Hansen tests were used to examine the exogeneity of instruments and the over-identifying restrictions in the two-step GMM process. The null hypothesis of the Sargan and Hansen tests is that the instruments are exogenous and valid. We are interested in the Hansen test because it is robust to heteroscedasticity and serial correlation compared with the Sargan test (see Arellano & Bond, 1991; Roodman, 2009). To perform the GMM analysis, we include the lag of LAWSUIT in our empirical model as an independent variable, and we use all control variables from Equation (1). As the residuals may be correlated across companies and over time, the standard errors can be biased. Therefore, following Petersen (2009) and Roodman (2009), we estimate clustered standard errors to produce more robust, reliable and unbiased coefficient estimates.

Column 1 of Table 9 shows the regression results of the GMM model for the relationship between the existence of an RMC and corporate litigation. The coefficient on RMC is negative and significant (coefficient = −1.001, p < 0.05), which confirms our baseline results. Further, the coefficient on AR(1) is significant (coefficient = −13.51, p = 0.000), suggesting that the null hypothesis of no autocorrelation in the first difference is rejected, whereas AR(2) is insignificant (coefficient = 0.81, p = 0.414), indicating that the error terms in the level regressions are not correlated. Furthermore, the p-value of the Sargan test is statistically significant (423.99, p = 0.000), whereas the Hansen test is insignificant (235.91, p = 0.866), indicating that the instruments used in the GMM system are valid. Overall, our baseline findings supporting H1 are consistent and robust. Additionally, Column 2 of Table 9 reports the GMM results for the relationship between RMC characteristics and corporate litigation. The coefficient on RMCSIZE (coefficient = −0.072, p < 0.10), RMCINDP (coefficient = −1.500, p < 0.05) and RMC_LEGEXPER (coefficient = −1.295, p < 0.10) are negative and significant which confirms our baseline results. Moreover, the coefficient of AR(1) is significant (coefficient = −1.69, p = 0.090), suggesting that the null hypothesis of no autocorrelation in the first difference is rejected, while the coefficient of AR(2) is insignificant (coefficient = −0.44, p = 0.661), indicating that the error terms in the level regressions are not correlated. Moreover, the Sargan test's p-value is statistically significant (227.87, p = 0.000), and the Hansen test is insignificant (99.26, p = 0.982), indicating the validity of the instruments and the fitness of our model. In sum, our baseline findings supporting H2a-H2c are consistent and robust.

Our baseline results may be affected by endogeneity concerns stemming from differences in observable rather than unobservable variables (Rosenbaum & Rubin, 1985). Since RMCs in our sample are not randomly distributed across firms, the baseline results on the association between RMCs and corporate litigation could be biased. Furthermore, one may argue that our regression model may suffer from the functional-form misspecification issue. For instance, our empirical model assumes a linear relationship between the two variables, while in reality, the relationship could be non-linear. To address these potential concerns, we employ the PSM approach because it can handle issues such as limited overlap and nonlinear relationships between the covariates and the treatment assignment (Shipman, Swanquist, & Whited, 2017). Austin (2011) contends that, to implement the PSM, it is necessary to examine the distribution of the regression variables between the treatment and control (matched sample) groups. There should be no systematic differences in the chosen variables between the treatment and matched groups except for the outcome variable (LAWSUIT). We employ kernel matching with a caliper of 0.1 ^[8].

Panel A of Table 10 shows the distributions of the covariates. The statistics for most variables in the RMC treatment and control samples are insignificant, implying that the matching process was successfully conducted. That is, most covariates are evenly distributed across the treatment and control samples. However, a small residual imbalance remains for BIG4, suggesting that the matching procedure does not fully eliminate all observable differences. To further assess matching quality, Panel B presents standardized percentage bias diagnostics before and after matching. The results show a substantial reduction in covariate imbalance across all variables. In particular, the mean standardized bias decreases from 34.7% before matching to 4.2% after matching, while the median bias declines from 33.4% to 3.3%. Moreover, most post-matching biases fall well below the conventional 10% threshold, except for BIG4. Because BIG4 is a binary indicator with limited variance, its standardized bias is mechanically inflated. Rubin's B (13.9) and Rubin's R (1.31) remain within acceptable ranges, indicating satisfactory overall balance. Although BIG4 retained a relatively higher post-matching bias (11.1%), the overall diagnostics confirm a strong improvement in covariate balance. Panel C of Table 10 presents the results of the PSM regression using a Poisson model, with the coefficient on RMC being negative and significant (coefficient = −0.282, p < 0.01). This suggests that our main results are robust to endogeneity concerns arising from observable, rather than unobservable factors.

This study examines whether the voluntary establishment of a standalone RMC and RMC characteristics mitigate corporate litigation. Analyzing a sample of listed Chinese firms from 2010 to 2020, this study reveals that the existence of a standalone RMC reduces the number of lawsuits filed against firms. Also, the results show that various RMC characteristics are associated with a significant reduction in lawsuits. We further conclude that the impact of the RMC and its characteristics on reducing the probability of litigation is stronger in firms with higher levels of risk disclosure. More importantly, we suggest that the governance effects of the RMC in reducing litigation risk are weaker in the SOEs, suggesting that political influence over board and sub-committee appointments, coupled with the prioritization of political intentions rather than the maximization of firm values dilutes RMC effectiveness. Our baseline results remain consistent and robust to endogeneity concerns, which we address through the GMM and PSM approaches.

In this study, our findings are grounded in both agency theory and RDT. Through the lens of agency theory, we find that the existence of RMCs strengthens the monitoring of managerial risk disclosures, thus reducing information asymmetry and subsequently lowering litigation risk. Our study also considers the P–P conflicts in SOEs, where tunneling and political influence can undermine RMC effectiveness, challenging the suitability of global governance models in such unique institutional contexts. Overall, our findings underscore the integration of agency theory and RDT in explaining RMC effectiveness but suggest that governance models must be carefully adapted to fit institutional environments where political dynamics and P-P conflicts are significant.

This study contributes to the literature in several important ways. First, it contributes to the literature on corporate litigation. To our knowledge, this study is the first to examine the governance role of the RMC in deterring corporate litigation. We find that firms can reduce the number of lawsuits by having a well-designed and qualified standalone RMC. Our findings suggest that Chinese regulators should issue more detailed guidance on the formation, duties and responsibilities of RMCs. More importantly, our study highlights that the diffusion of Anglo-Saxon governance mechanisms into different institutional settings – where state ownership and influence are more pronounced – may not function effectively, as these mechanisms become decoupled from the institutional context. Thus, while we respond to calls for research examining the interplay between firm-level governance and institutional context, we believe that further research is needed to understand how state roles can be leveraged effectively to mitigate their negative influences. However, some of our findings can be generalized to other similar settings, such as emerging economies with strong state involvement or transition economies where governance frameworks are still evolving.

Second, this study offers important practical implications for boards of directors, regulators and other practitioners, urging them to develop guidelines and recommendations that adapt global governance models to align with specific institutional settings. Moreover, it has implications for managers, as the findings underline the significance of the structure and qualification of the RMC in mitigating corporate litigation. This study also has policy implications for firms and regulators. With the Chinese economy transitioning from a planned to a more market-based economy, the number of qualified foreign institutional investors (QFIIs), for example, in the domestic A-share markets, increased significantly, from only one QFII in July 2003, to 283 QFIIs in July 2017 (Li, Wang, & Wu, 2021). Thus, focusing on the role of RMCs in corporate litigation in China bears important implications for foreign investors, who would like to see their investee firms form a standalone RMC to manage potential litigation risk. Finally, our study supports the latest regulations and guidelines ^[9] for establishing an RMC and its formation. Despite documenting the beneficial effects of an RMC, we caution against regulations mandating an RMC for non-financial firms in China. We encourage further research to better understand the cost–benefit trade-offs of mandating a standalone RMC in the context of China and other similar economies that are transitioning from a planned to a market-based economy while remaining dominated by state ownership in listed companies.

The supplementary material for this article can be found online.