The Independence of Judges and Corporate Social Responsibility

Abstract

Limited research has focused on the influence of judge independence on firms' corporate social responsibility (CSR), despite extensive literature examining the impact of the legal environment on CSR. To address this gap, we analyze the staggered adoption of judicial delocalization reform in China. This reform aimed to enhance local judges' independence and our analysis shows that firms have exhibited higher CSR scores since its implementation. Our channel analysis reveals an increase in lawsuits and monetary claims against firms due to the reform, indicating that firms engage in CSR activities to mitigate the negative effects of legal proceedings. The impact of the reform on CSR is stronger for high-litigation-risk firms and weaker for politically connected firms. Moreover, the positive relationship between the reform and CSR is more pronounced in regions with higher levels of government interference with judges before the reform. In summary, our findings highlight the important role of a supportive legal environment in fostering CSR, emphasizing the influence of judge independence.

Introduction

Numerous studies have highlighted the significant influence of institutional conditions on corporate social responsibility (CSR). These conditions encompass factors such as national culture (Cai et al., 2016; Peng & Lin, 2009), the political environment (Di Giuli & Kostovetsky, 2014), political intervention by the state (Li & Zhang, 2010), and social capital (Jha & Cox, 2015). Additionally, prior research has examined how legal origin affects CSR activities, contributing to the development of the institutional theory of CSR, which asserts that a firm's treatment of stakeholders is shaped by the institutions within which it operates (Campbell, 2007; Demirbag et al., 2017; Ioannou & Serafeim, 2012; Liang & Renneboog, 2017).

However, existing studies have largely overlooked the influence of ex post settling through judicial mechanisms on CSR behavior. Specifically, there is a gap in evidence regarding whether the attributes of judges impact CSR. Judge independence, as one of the key external factors, holds the potential to shape the institutional environment. It refers to the ability of judges to make impartial decisions based on factual evidence and the law, free from improper pressures and influences from any source. Judge independence is considered a fundamental condition for achieving the rule of law, as it ensures that all parties are equal under the law, irrespective of their political influence or status (Buchanan, 1974; Hayek, 2020). Existing literature provides evidence of how judge independence affects various corporate activities, such as fixed asset investment (Zhao & Zhang, 2022), external financing (Liu et al., 2022), and the number of newly established firms (Conti & Valentini, 2018). However, the understanding of whether and how judge independence influences CSR remains unclear and limited.

In the context of an independent and impartial judicial system, firms may be inclined to increase their engagement in CSR activities when faced with heightened litigation pressure. Judge independence serves to reduce the potential for local government interference in judicial processes, thereby curbing local protectionism and bolstering the fairness of the legal system. Consequently, increased judge independence amplifies litigation pressure on firms. Given that poor CSR performance presents a significant source of litigation risk for firms, heightened CSR investment can directly mitigate this risk (Barnett et al., 2018; Wang, 2012). As a result, firms tend to intensify their CSR activities in response to increased litigation pressure. Based on these arguments, we posit that judge independence facilitates higher levels of CSR engagement.

Judge independence in China has undergone significant developments during our sample period, presenting an ideal research context for our study. The Chinese government implemented gradual reforms in judicial delocalization from 2014 onwards, aiming to enhance the autonomy of judges and reduce local government interference. This reform involved transferring the fiscal and personnel management of Intermediate and Basic People's Courts from the jurisdiction of prefecture and county governments to provincial governments.¹ As a result, the reform has reduced the reliance of local judges on local government support, thereby bolstering judge independence. Consequently, the judicial delocalization reform provides a unique opportunity to examine the impact of establishing judge independence on firms' CSR practices.

To investigate the impact of the judicial delocalization reform on CSR, we employ a staggered adoption approach. Our dataset comprises 25,474 firm-year observations spanning from 2010 to 2020. Our empirical findings reveal a significant and positive association between the implementation of the judicial reform and CSR. Specifically, the judicial reform leads to a substantial increase in CSR, with an estimated effect size of 0.669, corresponding to 2.89% of the sample mean.

Importantly, our results demonstrate that the implementation of the judicial reform is accompanied by a notable rise in the number of lawsuits targeting listed firms. Moreover, we observe larger monetary claims associated with these lawsuits against listed firms after the reform compared to the pre-reform period. These findings provide compelling evidence in support of our central argument, which posits that firms are motivated to engage in a higher level of CSR activities as a protective measure against legal actions.

Additionally, we explore how the influence of judge independence on CSR varies across different contexts. Our analysis reveals that the impact of judge independence on CSR is particularly pronounced among firms facing elevated litigation risk and those headquartered in provinces characterized by more prevalent local government interference with judges. Interestingly, we also find that the positive relationship between judge independence and CSR is attenuated for politically connected firms. This could be attributed to the presence of already elevated CSR practices within those firms, as suggested by prior research (Chen et al., 2018; Li et al., 2015).

Our study contributes to the existing literature in two main ways. Firstly, we enhance the understanding of the role of institutional conditions in shaping socially responsible corporate behavior, building upon prior work such as Ioannou and Serafeim (2012). While previous research based on cross-country data has demonstrated that firms are more likely to engage in social responsibility when strong laws and regulations are in place to promote CSR activities, these studies are susceptible to potential issues such as omitted variables and measurement biases (Demirbag et al., 2017; Liang & Renneboog, 2017). Crucially, the impact of ex post settlement through judicial mechanisms on CSR behavior has been overlooked. By leveraging an exogenous judicial delocalization reform in China, we examine the relationship between increased judge independence and heightened litigation pressure on firms, thereby influencing their CSR performance. Our empirical findings provide evidence supporting a positive association between judge independence and firms' CSR performance. To the best of our knowledge, this paper represents the first attempt to investigate the impact of heightened judge independence on corporate social responsibility.

Secondly, our study contributes to the growing body of literature on the motivations underlying CSR. Existing research has identified various drivers of CSR, including value enhancement (Freund et al., 2023; Lins et al., 2017), risk management (Miras‐Rodríguez et al., 2015; Barnett et al., 2018), and altruism (Bénabou & Tirole, 2010; Xue et al., 2022). In our study, we explore a less-explored motivation rooted in risk management. We propose that firms are more inclined to engage in CSR activities as a means of protecting themselves from litigation risks when judge independence increases and leads to heightened litigation pressure. This study represents the first endeavor to examine the impact of judge independence on CSR based on risk management motives.

The remainder of the paper is organized as follows. Section "Institutional Background and Hypothesis Development" develops testable hypotheses. Section "Data and Sample Construction" discusses the data and sample construction. Section "Empirical Results" presents the empirical results. Finally, Section "Discussion and Conclusion" concludes the paper.

Institutional Background and Hypothesis Development

Institutional Background

The existing literature has predominantly focused on examining the independence of judges within the framework of democratic political systems (Helmke & Rosenbluth, 2009; La Porta et al., 2008). However, the applicability of these findings to emerging markets, particularly in authoritarian contexts like China, may be limited. Unlike the United States, where federal judges enjoy lifetime tenure and have protection against salary reductions to enhance their ability to resist potential political pressures, China's political structure lacks a separation of powers (Michelson, 2007). The absence of a checks-and-balances system in China's unified government, led by a single-party regime, undermines judicial independence and weakens the rule of law (Lei & Li, 2022). Prior to the judicial reform examined in this study, the Chinese local government often exerted influence over judges, guiding their decisions on politically sensitive disputes based on personal preferences rather than the law (Li, 2016). This political interference historically resulted in widespread judicial favoritism at the local level in China.

China's court system is structured into four tiers: the Basic People's Court at the county level, Intermediate People's Court at the prefecture level, Higher People's Court at the province level, and the Supreme People's Court at the national level. Before the implementation of the judicial reform, the management of the Higher, Intermediate, and Basic Courts was under the control of the respective governments at the province, prefecture, and county levels. Most first-instance hearings for cases took place in Basic and Intermediate People's Courts. Consequently, local firms or state-owned enterprises (SOEs) enjoyed significant advantages over non-local plaintiffs due to the influence exerted by local governments and the application of the principle of place of defendants, which directed lawsuits to be processed in the courts located in the defendants' cities (Firth et al., 2011).

Prior to the implementation of the judicial reforms, an independent judiciary was non-existent in China, as local judges inevitably faced political interference from local governments. Prefecture and county governments held control over decisions regarding judges' personnel and salaries in the Intermediate and Basic People's Courts. Consequently, local judges lacked the means to resist local government interference, leading to prevalent local judicial protectionism throughout China (Lei & Li, 2022; Xu, 2011). Local politicians in prefectures and counties frequently intervened in case decisions, resulting in outcomes that favored local litigants (Peerenboom, 2002). These protectionist interventions served the interests of local politicians seeking to increase tax income, create job opportunities, and engage in rent-seeking activities.²

The lack of judicial independence, coupled with corruption, has resulted in inefficiencies within China's judicial system. Judicial efficiency, referring to the speed with which cases are resolved, has been significantly compromised (Shah et al., 2017). Judicial systems lacking independence exhibit inefficiencies throughout various stages of the trial process in China. For example, judges may be hesitant to accept cases involving local firms or SOEs. Even when such cases are accepted, judges frequently prolong the duration of hearings, making litigation unattractive for plaintiffs. As a result, due to the inherent inefficiencies in China's judicial process, the cost of litigation remains high, while the potential compensation, discounted by the probability of success, is often low. These factors collectively contribute to a significantly low prosecution rate in China (Huang, 2018).

Since 2014, the Chinese government has gradually implemented the judicial delocalization reform to enhance the independence of judges in ruling on cases. This reform involved recentralizing the fiscal and personnel management of Intermediate and Basic People's Courts, transferring it from the control of prefecture and county governments to provincial governments. As a result of this reform, judges have become independent from local politicians in terms of monetary benefits and career promotions. This shift means that local governments can no longer instruct judges to favor local firms. Recent studies have demonstrated that the judicial delocalization reform effectively enhances judicial autonomy by transferring authority over local court personnel and finances to the provincial level (Wang, 2021). It has also contributed to increased efficiency by enabling the selection of better-trained judges and enhancing judicial independence from local politicians (Li & Ponticelli, 2022; Wang, 2021).

In conclusion, while existing literature primarily focuses on the independence of judges within democratic political systems, its findings may have limited applicability to authoritarian systems like China. China's political structure, lacking a separation of powers, has historically hindered the independence of judges and resulted in a lack of judicial efficiency. However, the implementation of the judicial delocalization reform since 2014 has aimed to strengthen the independence of judges and improve the efficiency of the judicial system. This reform has shifted control over personnel and finances to the provincial level, reducing the influence of local politicians and promoting judicial autonomy. The subsequent improvements in judicial efficiency and independence have significant implications for understanding the relationship between judge independence and corporate social responsibility in the Chinese context.

Hypothesis Development

Our hypothesis posits that increased judge independence heightens litigation pressure on firms, prompting them to engage more in CSR activities as a protective measure against litigation. Judge independence refers to the capacity of judges to deliver impartial judgments based solely on the facts and law, free from any undue pressure or influence. It shields legal proceedings from local government interference, including attempts to exert pressure on judges or directives to rule in favor of local firms (Xu, 2020). Consequently, judge independence effectively eradicates judicial local protectionism, enhancing the fairness of the legal system. For example, Lei and Li (2022) found that judge independence lowers the success rates of local firms by reducing the dependency of local judges on financial support from local politicians. Liu et al. (2022) further report a significant decline in the success rate of local defendants against non-local plaintiffs following increased judge independence, attributed to improved judicial decision-making quality, as evidenced by lower appeal rates and more nuanced reasoning. Additionally, Zhao and Zhang (2022) discovered that judge independence reduces the cost of bribing local courts, indicating reduced reliance on bribery by firms to win lawsuits. Overall, judge independence not only advances fairness and justice within the legal system but also increases the probability of potential plaintiffs succeeding and reduces litigation-related costs, encouraging potential plaintiffs to utilize the legal system to resolve disputes with listed firms.

Firms are more likely to engage in CSR activities when faced with greater litigation pressure. CSR serves as a risk management tool to protect firms from litigation (Miras‐Rodríguez et al., 2015; Barnett et al., 2018; Xue et al., 2022). While poor CSR performance represents a significant source of litigation risk for firms, increased investment in CSR can directly lower a firm's litigation risk (Wang, 2012). Firms demonstrating superior CSR performance are more likely to comply with various rules, laws, and regulations and are less inclined to engage in actions that harm stakeholder relationships. Waddock and Graves (1997) suggest that sound management practices, manifested through CSR activities, reduce corporate risk exposure, which could otherwise provide grounds for litigation. Moreover, stakeholders consider a firm's history of social actions when deciding how to respond to its current actions (Barnett, 2007). A negative event occurring in a firm with a strong commitment to social responsibility may be perceived as a result of misfortune rather than indicative of poor management or attitude (Barnett, 2014; Minor & Morgan, 2011). Consequently, potential plaintiffs may be less likely to sue firms with a stronger CSR commitment. If judge independence incentivizes potential plaintiffs to initiate proceedings against listed firms in China, and if CSR activities shield firms from litigation, then we anticipate a positive association between judge independence and CSR. Thus, we formulate our hypothesis as follows:

Hypothesis 1:

Judge independence exerts a positive effect on CSR.

An alternative hypothesis suggests that judge independence may not significantly influence CSR. It is plausible that increased judge independence could improve judicial efficiency, but enhanced efficiency may result from a combination of factors such as reduced judicial corruption, in addition to increased independence (Peerenboom, 2006; Voigt, 2016). Furthermore, even in the absence of political interference, Chinese judges could potentially misuse their legal power for personal gain, thus impeding optimal judicial efficiency (Gong, 2004). These factors imply that an increase in judge independence may not automatically translate into improved efficiency within China's judicial system. Consequently, the reliance of the public on the judicial system for resolving disputes involving public firms might remain largely unchanged. Given this perspective, firms may not perceive a compelling need to adjust their CSR activities accordingly. Therefore, the impact of judge independence on firms' CSR remains an empirical question that necessitates further investigation.

If hypothesis 1 is correct, firms with higher litigation risk may experience a larger increase in litigation pressure following enhanced judge independence, leading them to engage in more CSR activities as a protective measure. We expect the positive relationship between judge independence and CSR to be stronger in firms with higher litigation risk. Thus, we formulate hypothesis 2a as follows:

Hypothesis 2a:

The positive effect of judge independence on CSR is stronger among firms with higher litigation risk.

Chen et al. (2018) argue that SOEs, as government-controlled shareholders, allocate more resources to CSR activities than non-SOEs. Additionally, Li et al. (2015) find that non-SOEs with political connections also engage in more CSR activities. Politically connected non-SOEs have strategic incentives to engage in CSR activities to maintain government support and political benefits. Prior literature consistently documents that politically connected firms, including both SOEs and politically connected non-SOEs, engage in more CSR activities than non-politically connected firms. Considering that poor CSR performance is a significant source of litigation risk (Barnett et al., 2018; Wang, 2012) and that judge independence may have a larger impact on the litigation risk of firms with less CSR engagement, we expect non-politically connected firms to increase their CSR engagement more following increased judge independence. Therefore, if hypothesis 1 is correct, we hypothesize that the positive effect of judge independence on CSR will be weaker among politically connected firms compared to non-politically connected firms. We formulate hypothesis 2b as follows:

Hypothesis 2b:

The positive effect of judge independence on CSR is weaker among politically connected firms.

As the promotion of local government officials often relies on local economic development, aspiring local officials may interfere with the judicial process to support local interests by controlling the personnel and finances of local judges (Chen, 2003; Li & Zhou, 2005). In provinces with significant political interference, lax and opaque law enforcement undermines the credibility of justice and the authority of the law. If hypothesis 1 is correct, we predict that the effect of judge independence on CSR will be more pronounced in provinces with higher levels of local government intervention. We formulate hypothesis 2c as follows:

Hypothesis 2c:

The positive effect of judge independence on CSR is stronger in provinces with higher levels of local government intervention.

Data and Sample Construction

Data and Sample

The Chinese government implemented the judicial delocalization reform between 2014 and 2016, requiring different provinces to stagger the start of adoption.³ Our analytic sample begins in 2010, four years prior to the first province enacting the reform, and ends in 2020, four years after the last group of provinces began the reform.⁴ However, it is important to note that the judicial delocalization reform may have had different years of commencement within cities of the same province. Therefore, relying solely on the schedule for the roll-out of the reform at the provincial level as a treatment variable may not accurately identify the effect of the judicial reform. To address this, we follow the approach of Liu et al. (2022) and Zhao and Zhang (2022) by manually collecting information on the schedule for the roll-out of the reform at the city level from the Yearbook of Judicial Reforms in China. This yearbook provides a chapter for each province, which includes information on the schedule for the roll-out of the reform in different cities within the provincial jurisdiction. This detailed city-level information enables us to capture the variation in the timing of the reform and improve the accuracy of our identification strategy.

We collected financial performance and corporate governance data from the Chinese Security Market and Accounting Research Database (CSMAR). The biographical information and work experience of managers were obtained from the China Listed Firm's Corporate Governance Research Database in CSMAR, which provides comprehensive details on executives such as the CEO, chairman, board members, and senior executives. Following the approach of Liu et al. (2016), we identified political connections by analyzing whether a firm's CEO or chairman had a background in government or military service, or had served as a deputy of the Provincial/National People's Congress or the Chinese People's Political Consultative Conference. The data on lawsuit information of listed firms in China were obtained from the Chinese Research Data Services (CNRDS). Our sample includes various types of lawsuits, such as civil litigation, criminal proceedings, administrative litigation, and other unclassified cases.⁵ To ensure comparability, we excluded firms listed in the financial industry due to differences in accounting standards. After excluding observations with missing values, our final sample consists of 25,474 firm-year observations. To mitigate the impact of outliers, we winsorized all continuous variables at the 1% and 99% levels.

Measuring CSR

According to prior research (Xu & Ma, 2021; Zhu et al., 2022), we obtained the data on Chinese firms' CSR scores from Hexun Information Technology Limited Company.⁶ Hexun serves as a professional assessment organization for evaluating the social responsibility system of listed firms in China. The company collects information from firms' CSR and annual reports, making it one of the most widely used sources of CSR scores in the literature on Chinese listed firms. Hexun's CSR score is calculated based on the assessment of five stakeholder categories, namely environmental responsibility, employee responsibility, shareholder equity responsibility, supplier and customer rights, and social responsibility. Hexun has assigned weights of 20%, 15%, 30%, 15%, and 20% to these respective categories. We calculate the overall CSR score by summing up the scores from these five stakeholder categories. It is important to note that the CSR score exhibits a positive association with CSR activities. Figure 1 presents the average CSR score over time, illustrating significant variations in the CSR score throughout the observed period.

Fig. 1

Average CSR Score Over Time

Control Variables

We control for a set of variables, including the ratio of the book value of long and short-term debt to the book value of assets (Lev), the ratio of the market value of assets to the book value of assets (MB), the ratio of cash to total assets (Cash), natural logarithm of total assets (Size), the ratio of net income to total assets (ROA), an indicator equal to one if a firm’s controlling shareholder is the government or its agent (SOE), natural logarithm of the number of years since initial public offering (Age), the ratio of intangible assets to total assets (Intangible), the number of directors on the board (BoardSize), the proportion of independent directors on the board (IndDirect), the shareholding held by the largest shareholder (Top1), annual GDP growth rate (GDP growth), natural logarithm of GDP per capita (GDP per capita), and marketization index (Market).

Debt has been found to potentially limit a company's ability to allocate resources towards CSR initiatives (Freund et al., 2023). Therefore, we anticipate a negative coefficient for the control variable Lev in the CSR regression. On the other hand, research suggests that firms with higher value, greater cash reserves, larger size, and higher profitability are more likely to invest in CSR activities (Chen et al., 2020; Hong et al., 2012; Lins et al., 2017). Consequently, we expect the control variables MB, Cash, Size, and ROA to have positive coefficients. Furthermore, Chinese SOEs often engage in CSR to fulfill political and social objectives set by the government (Chen et al., 2018). Additionally, firms utilize CSR practices to leverage the potential of their intangible assets (Shen et al., 2020). Managers also employ effective corporate governance mechanisms that align with CSR practices to mitigate conflicts among stakeholders (Jo & Harjoto, 2012). Hence, we anticipate positive coefficients for the variables SOE, Intangible, BoardSize, and IndDirect in our regression model. To account for the influence of firm age on CSR activities, we include it as a control variable. Prior research indicates a positive association between firm age and CSR engagement (Freund et al., 2023). Moreover, we consider the impact of the shareholding of the largest shareholder and regional economic development on CSR. Li and Zhang (2010) find that the shareholding of the largest shareholder is positively associated with CSR in SOEs, while it is negatively associated with CSR in non-SOEs. They also suggest that the relationship between regional economic development and CSR is negative in SOEs and less significant in non-SOEs. Therefore, the impact of these variables on CSR is uncertain. In our study, we measure regional economic development using the natural logarithm of GDP per capita, GDP growth rate, and the marketization index at the province level.⁷

Summary Statistics

Panel A of Table 1 presents summary statistics for the key variables used in our analysis. The variable Justice is a dummy variable that takes a value of one if a firm is headquartered in a city where the judicial delocalization reform was implemented in a given year, and zero otherwise. The average value of Justice is 0.331, indicating that 33.1% of the firm-year observations in our sample are exposed to the judicial delocalization reform. It is worth noting that our sample period spans from 2010 to 2020, which includes a shorter pre-reform period and a longer post-reform period compared to the study by Zhao and Zhang (2022). Consequently, the mean value of the treatment variable in our paper is larger than that in Zhao and Zhang (2022), which had a sample period from 2009 to 2018 with a mean value of 0.187 for the treatment variable.

Table 1 Summary statistics

The mean value of the CSR variable is 23.173, with a median value of 21.160, and a standard deviation of 15.363. Panel B of Table 1 presents the univariate analysis of CSR before and after the adoption of the judicial delocalization reform. The mean CSR value before the judicial reform is 19.403, while after the reform, it increases to 26.472. The statistical analysis indicates a significant increase of 7.069 (significant at the 1% level) in CSR following the adoption of the judicial reform. Thus, the results of the univariate analysis provide preliminary evidence for the positive impact of the judicial reform on CSR.

Empirical Results

Baseline Regressions

To examine the effect of the independence of judges on CSR, we design the following staggered difference-in-difference (DID) test by employing the staggered adoption of the judicial delocalization reform as a quasi-natural experiment.

$$CSR_{{i,c,p,t}} ~ = ~\alpha + \beta Justice_{{c,p,t}} + \lambda ^{\prime} X_{{i,c,p,t - 1}} + \nu _{i} ~ + \nu _{t} + \varepsilon _{{i,c,p,t}}$$

(1)

where i, c, p, and t index firm, city, province, and year, respectively. Justice is set to one if a firm is headquartered in a city c of province p that has adopted judicial delocalization reform in year t, and to zero otherwise. The dependent variable is CSR. ${v}_{\text{i}}$ represents firm fixed effects, and ${v}_{\text{t}}$ represents year fixed effects. Following prior literature (e.g., Freund et al., 2023), we control for a set of prior year variables as shown in Section "Control Variables" in ${X}_{\text{i,c,p,t-1}}$.

Table 2 presents the results of the baseline regressions. In Column 1, we include firm and year fixed effects as controls. In Column 2, we further control for additional variables, including cash holdings, book leverage, market-to-book ratio, return on assets, firm size, a dummy variable indicating whether a firm is an SOE, the logarithm of firm age, intangible assets, the number of directors on the board, the proportion of independent directors on the board, and the shareholding of the largest shareholder. Column 3 includes additional control variables, namely GDP growth, GDP per capita, and Market.

Table 2 Baseline regressions: judicial reform and CSR

The coefficients of the Justice variable are positive and statistically significant at the 5% level in all regressions. This indicates that the adoption of the judicial delocalization reform has a positive effect on CSR. In terms of economic significance, the reform is associated with an increase in CSR ranging from 0.661 to 0.669. This increase represents approximately 2.85% to 2.89% of the sample mean. These findings provide evidence that greater judge independence leads to higher levels of CSR, consistent with the expectations outlined in Hypothesis 1.

Channel Analysis

Lawsuits

Potential plaintiffs have increasingly relied on courts to resolve their disputes with listed firms following the judicial reform due to the increased fairness and reduced litigation-related costs associated with judge independence. Consequently, we anticipate a positive relationship between the judicial reform and corporate lawsuits. In this section, we aim to test whether corporate lawsuits serve as a channel through which judge independence affects CSR.

To analyze firms' lawsuit case data, we extract information from Chinese Research Data Services. Following the approach of Jia et al. (2019), we define the variable Number as the total number of times a firm is sued as a defendant in a given year. Additionally, we calculate Amount as the total monetary claim against a defendant divided by the book value of the defendant firm's assets in the same year.

We conduct lawsuit regressions and report the results in Table 3, Columns 1 to 2. The coefficients of the Justice variable are positive and statistically significant in all models, indicating that judge independence leads to an increase in the number of lawsuits against firms and the amount of monetary claims made against defendants in lawsuits. In our analytic sample, the mean values of Number and Amount are 1.19 and 0.02, respectively. The adoption of the judicial delocalization reform results in an increase in Number by 0.940, which corresponds to 78% of its sample mean. Similarly, the adoption of the reform leads to an increase in Amount by 0.011, which represents 55% of its sample mean.

Table 3 Judicial reform and corporate lawsuits

We further examine the type of dispute to determine whether the increase in lawsuits is a result of reduced external political pressure on the judicial system. We expect the positive impact of judge independence on corporate litigation pressure to be more pronounced in cases that are susceptible to government intervention. Xu (2020) finds that loan- and contract-related disputes exhibit less political interference in judicial proceedings compared to other types of disputes. This is because judges have limited leeway to manipulate decisions in loan- and contract-related cases, as they must adhere to the agreements and contractual obligations established between the parties involved. Conversely, other types of lawsuits offer greater opportunities for political manipulation by local governments.

Drawing on the classification of cases by Firth et al. (2011), we categorize each case based on our analysis of the disputes mentioned in litigation announcements. We define Contract Number as the count of times a firm is sued as a defendant in contract- and loan-related disputes, and Other Number as the count of times a firm is sued as a defendant in other types of disputes. Additionally, Contract Amount represents the monetary claim against a defendant in contract- and loan-related disputes, while Other Amount represents the monetary claim in other types of disputes. These variables are derived by dividing the monetary claim by the book value of the defendant firm's assets. In Columns 3 and 4 of Table 3, the coefficient of Justice is statistically insignificant, suggesting that the implementation of the reform does not significantly increase contract- and loan-related lawsuits. However, in Columns 5 and 6 of Table 3, the coefficient of Justice is positive and statistically significant at the 5% level, indicating that the implementation of the judicial reform significantly increases other types of lawsuits. These results support the argument that the positive effect of judge independence on corporate lawsuits is stronger in case types that are vulnerable to political manipulation, demonstrating that judge independence reduces government intervention in judicial cases.

Other Possible Explanations

Cuervo-Cazurra (2006) has discovered that in the absence of judicial independence, firms often resort to bribery as a means to cultivate favorable relationships with local governments and mitigate potential encroachments on their interests. Additionally, when local judges' decisions are susceptible to interference from the local government, firms are more inclined to engage in bribery with local officials to secure case outcomes that align with their corporate interests. As a result, the lack of judge independence not only leads to increased corporate expenditures on bribes and administrative entertainment but may also displace spending on CSR initiatives.

Zhao and Zhang (2022) have demonstrated that judge independence reduces the necessity for firms to engage in bribery with local officials. Through the enhancement of judge independence, the judicial reform enables firms to resolve interest-based disputes in a fair manner, diminishing or eliminating the need for bribes. Consequently, we anticipate that judge independence alleviates the burden on firms to bribe local officials, allowing them to allocate more capital toward CSR activities. To proxy for corporate bribery expenditures, we follow the approach of Zhao and Zhang (2022) and use entertainment expenditures divided by sales. In Panel A of Table 4, Column 1, we find that the estimated coefficient of Justice is negative and significant, indicating that the judicial reform reduces corporate bribery expenditures. This result supports the argument that judge independence leads to more capital available for firms to invest in CSR activities.

Table 4 Other possible explanations and channel analysis

Furthermore, judicial independence can safeguard the rights and interests of shareholders and creditors and enhance investor confidence in capital markets (Boubakri & Ghouma, 2010; Laeven & Majnoni, 2005). Prior research provides empirical evidence that judicial independence increases firms' access to external finance. Zhao and Zhang (2022) show that judge independence significantly increases long-term loans obtained by firms from banks or other financial institutions. Wu et al. (2023) find that banks tended to grant short-term loans to firms prior to the judicial reform as a means to pressure indebted firms to repay their loans and mitigate default risk. They further find that enhanced judicial independence curtails opportunistic actions by firms and instills confidence in banks to extend more long-term loans. We argue that another explanation for this relationship is that judge independence leads to increased external funding for firms, enabling them to invest more in CSR activities. We measure Long loan as the natural logarithm of long-term bank loans plus one. In Column 2 of Panel A in Table 4, the coefficient of Justice is positive and significant, indicating that more external funds become available to firms following the judicial reform. This empirical result supports our argument that judge independence leads to greater external funding for CSR.

To address concerns that our findings may be driven by alternative explanations, we employ structural equation modeling for path analysis, following the methodology of prior literature (Callen et al., 2020). This approach allows us to compare the proportions of indirect effects from different mediating variables. In the path analysis models, we treat judge independence as having a direct effect on CSR and an indirect effect on CSR through various mediating variables. Additionally, path analysis enables quantification of the proportions of indirect effects from multiple mediating variables. We incorporate three mediating variables: Number and Amount representing corporate litigation, Bribe indicating bribery expenditures, and Long loan reflecting bank loans. These variables are included simultaneously to investigate their potential mediating effects. We regress CSR on the judicial reform adoption and the mediating variables, while also regressing the mediating variables on the judicial reform adoption. All regressions incorporate control variables and fixed effects, consistent with the baseline regression.

Panel B of Table 4 presents the results of the path analysis. In Column 1, we measure corporate litigation using Number, and in Column 2, we proxy for corporate litigation using Amount. The reported standardized path coefficients are denoted as p(.).⁸ The direct effect of judge independence on CSR is captured by the standardized path coefficient of Justice in the regression of CSR on the judicial reform adoption and the mediating variables. The indirect effect of judge independence on CSR is calculated as the product of the effect of the judicial reform adoption on the mediating variables and the effect of the mediating variables on CSR. To assess the significance of the indirect effect, we employ the Sobel (1982) test statistics.

The results in Panel B of Table 4 indicate that both corporate litigation and bank loans are positively related to the adoption of judicial reform and have a positive impact on CSR. Conversely, bribery expenditures are negatively associated with the adoption of judicial reform and exert a negative effect on CSR. All path coefficients are statistically significant. Notably, in Column 1, the indirect effect of Number accounts for 62.8% (0.00076/0.00121) of the total indirect effects through all three mediating variables. Similarly, in Column 2, the indirect effect of Amount represents 68.8% (0.00104/0.00151) of the total indirect effects through all three mediating variables. These findings suggest that corporate litigation has a more substantial impact on the relationship between judge independence and CSR, which mitigates concerns related to alternative explanations.

Robustness Tests

Control for Specialized Courts

China has implemented specialized tribunals and established new specialized courts in certain pilot cities as a means to restrict local government interference in specialized cases. These specialized institutions operate under the guidance of highly skilled and newly appointed judges, deviating from the traditional civil courts. By possessing specialized education and training, these judges are better equipped to defend themselves against interventions by local politicians. Li and Ponticelli (2022) find that the introduction of specialized courts leads to reduced case duration and increased judicial independence. To ensure the robustness of our findings and account for the effects of specialization reforms, we conduct additional tests by controlling for the introduction of specialized courts.

In our analysis, we incorporate controls for specialized bankruptcy courts and specialized intellectual property courts in the baseline regression. Data on the establishment of specialized courts are manually collected from sources such as the People's Court website and news reports.⁹ We create the variables Bankruptcy and Intellectual Property as dummies. If a bankruptcy tribunal/court or an intellectual property tribunal/court has been established in the city where a firm is headquartered in a given year, the respective dummy variable is set to one; otherwise, it is set to zero. The results presented in Column 1 of Table 5 demonstrate that the estimated coefficient of Justice remains positive and statistically significant at the 5% level even after controlling for the introduction of specialized courts. This finding indicates that our main result regarding the impact of judge independence on CSR is robust. Notably, the coefficient of Intellectual Property is negative and significant. This may be attributed to the fact that intellectual property courts strengthen legal protection for intellectual property assets and create barriers to market entry for competitors (Branco & Rodrigues, 2006). Firms often engage in CSR activities to establish competitive advantages and differentiate themselves from rivals (Campbell, 2007; Dupire & M'Zali, 2018). Consequently, firms are less likely to prioritize CSR in an environment with reduced competitive pressures.

Table 5 Robustness tests

Control for the Rotation of Presidents of Higher People’s Courts (HPC)

Local firms often exploit the extended tenure of High People's Court (HPC) presidents and their geographical proximity to establish favorable relationships, known as "guanxi" in Chinese, with these presidents. These relationships can influence HPC presidents to abuse their authority and make judicial decisions that align with the interests of local firms.¹⁰ Such corrupt practices result in unfair trial outcomes and discourage people from resorting to the courts, contributing to the inefficiency of the judicial system in China (Firth et al., 2011). In an effort to curb judicial corruption and limit the tenure of HPC presidents within the same province, the Chinese central government initiated a rotation reform in 2013. Under this reform, HPC presidents are transferred from one province to another to serve, while those in provinces unaffected by the rotation reform continue in their positions.

In our analysis, we construct the variable Rotation as a dummy variable that takes a value of one if the province where a firm is headquartered has adopted the rotation reform in a given year, and zero otherwise. After incorporating this control variable, the estimated coefficient of Justice in Column 2 of Table 5 remains positive and statistically significant. This suggests that the positive effect of judge independence on CSR persists even after accounting for the impact of the rotation reform.

Control for Circuit Court

To enhance the independence of China's judicial system, the Chinese government has implemented circuit court reform since 2015. This reform involves the establishment of circuit courts of the Supreme People's Court in locations outside of Beijing, with these courts possessing the same jurisdiction as the Supreme People's Court. The key characteristic of circuit courts is their independence from local governments, as they operate under the direct authority of the Supreme People's Court. Lai et al. (2023) have found that the introduction of circuit courts reduces local judicial protectionism and strengthens legal enforcement.

In our analysis, we construct the variable Circuit Court as an indicator that takes a value of one if the province where a firm is headquartered has adopted the circuit court reform in a given year, and zero otherwise. After incorporating this control variable, the estimated coefficient of Justice in Column 3 of Table 5 remains positive and statistically significant. This result suggests that the positive relationship between judge independence and CSR persists even after accounting for the introduction of circuit courts.

Control for Anticorruption Campaign

According to prior studies (Hossain & Kryzanowski, 2021; Kong et al., 2021; Ucar & Staer, 2020; Xue et al., 2022), an anticorruption campaign is expected to increase CSR. In November 2012, China's central government initiated an anticorruption campaign, with the Central Commission for Discipline Inspection (CCDI) serving as the executive agency for the campaign. CCDI deployed inspection teams to target specific provinces in 2013 and 2014, aiming to strengthen anticorruption legislation and mitigate rent-seeking behaviors.

To account for the influence of the anticorruption campaign, we incorporate the variable Anticorruption in the baseline regression. Following the approach of Kong et al. (2021), we create a dummy variable called Anticorruption, which takes a value of one if the province where a firm is headquartered has been inspected by CCDI in a given year, and zero otherwise. The result presented in Column 4 of Table 5 indicates that the coefficient of Justice remains positive and statistically significant even after controlling for the anticorruption campaign. This finding suggests that our main result regarding the positive impact of judge independence on CSR holds true in the presence of the anticorruption campaign.

Dynamic Models

The parallel trend assumption is at the heart of the DID design. To test the parallel trend assumption and the dynamic effects of the judicial reform, we investigate the dynamic model as follows.

$$CSR_{{i,c,p,t}} ~ = ~\alpha + \beta _{1} Justice_{{c,p,t}}^{{ - 2}} + \beta _{2} Justice_{{c,p,t}}^{{ - 1}} + \beta _{3} Justice_{{c,p,t}}^{0} + \beta _{4} Justice_{{c,p,t}}^{{ + 1}} + \beta _{5} Justice_{{c,p,t}}^{{ + 2}} + \beta _{6} Justice_{{c,p,t}}^{{ \ge + 3}} + \lambda ^{\prime} X_{{i,c,p,t - 1}} + \nu _{i} ~ + \nu _{t} + \varepsilon _{{i,c,p,t}}$$

(2)

where i, c, p, and t index the firm, city, province, and year, respectively. ${X}_{\text{i,c,p,t-1}}$ is the same set of controls as that in Eq. (1). The six dummy variables Justice ⁻², Justice ⁻¹, Justice ⁰, Justice ⁺¹, Justice ⁺², and Justice ^≥+3 are equal to one if a firm is headquartered in a city that will adopt the judicial delocalization reform in the next two years, the next year, adopts the judicial reform this year, adopted the reform one year ago, two years ago, or three or more years ago, respectively.

The results of the analysis are presented in Table 6. Prior to the introduction of the judicial delocalization reform, we find no significant difference in the change of CSR between the treatment and control groups. The coefficients of Justice ⁻² and Justice ⁻¹ are both statistically nonsignificant, indicating that CSR trends in both the control and treatment groups followed parallel paths prior to the reform. This supports the parallel trend assumption, which is important for the validity of the difference-in-differences estimation.

Table 6 Dynamic models for judicial reform and CSR

In Column 3 of Table 6, the coefficient of Justice ⁰ is positive but statistically insignificant. This suggests that there is no immediate impact of the judicial delocalization reform on CSR in the year of its introduction. However, one year after the reform, we observe a significant and sharp increase in CSR for the treatment firms compared to the control firms (0.94, statistically significant at the 5% level). This finding aligns with previous literature (Lei & Li, 2022) and suggests that the effects of the judicial reform take some time to materialize.

Moreover, the coefficients of Justice⁺² and Justice ^≥+3 are statistically significant and positive (1.01 and 1.17, respectively). This indicates that the increase in CSR for the treated firms relative to the control firms persists in the second, third, and subsequent years following the introduction of the judicial reform. These findings align with the evidence on the persistence of judicial reform effects reported in Liu et al. (2022). Figure 2 further illustrates the estimated coefficients capturing the dynamic effects over time.

Fig. 2

Dynamic Test of Judicial Delocalization Reform and CSR. This figure plots the 90% confidence interval of coefficient estimates in Column 3 of Table 6

Alternative Sample

To address concerns regarding the potential influence of unobserved shocks on the positive relationship between the judicial reform and CSR, we conduct a robustness test by constructing an alternative sample. Drawing from the approach of Li et al. (2017), we narrow down the sample period to one year before and one year after a shock, specifically the years 2013 and 2014. This selection aims to enhance the identification strategy employed in our difference-in-differences analysis. Additionally, we limit the control group to neighboring provinces of those provinces affected by the judicial delocalization reform in 2014. By focusing on these neighboring provinces with comparable economic characteristics, we aim to minimize potential confounding factors. The final sample used in this robustness test consists of 3,240 firm-year observations. The results, as presented in Column 5 of Table 5, indicate that the estimated coefficient of Justice remains significantly positive, consistent with our previous findings.

Control for Additional Fixed Effects

In Column 6 of Table 5, we introduce industry and province fixed effects to account for time-invariant omitted characteristics specific to industries or provinces. This helps control for potential biases that could arise from unobserved factors at the industry or province level. By including these fixed effects, we aim to isolate the impact of judge independence on CSR more accurately. Consistent with prior research on CSR (Chen et al., 2023; Freund et al., 2023), we incorporate industry × year fixed effects in Column 7 to address concerns that time-varying shocks at the industry level may simultaneously influence the implementation of the reform and CSR activities. Notably, in Column 7, we do not explicitly include industry and year fixed effects as these are absorbed when controlling for industry × year fixed effects. Importantly, the estimated coefficient of Justice remains positive and statistically significant in Columns 6 and 7 of Table 5, even after accounting for these additional fixed effects.

Heterogeneity Tests

In this section, we test the heterogeneity of the effect of judge independence on CSR from the following three perspectives: litigation risk, political connection, and local government intervention.

Litigation Risk

To measure corporate litigation risk, we employ two variables in our analysis. Firstly, following the approach of Wang and Li (2016), we categorize certain industries as high litigation risk industries, including computers and office equipment, drugs, retail trade, electronic and other electrical equipment and components (except for computers and equipment), and computer programming, data processing, and other computer-related services.¹¹ We create an indicator variable called Industry that takes a value of one if a firm operates in these high litigation risk industries, and zero otherwise. We incorporate both the Industry variable and the interaction term Justice × Industry into Eq. (1). The results, as reported in Column 1 of Table 7, reveal that the estimated coefficient of Justice × Industry is positive and statistically significant at the 5% level. This suggests that the positive relationship between judge independence and CSR becomes stronger among firms operating in high litigation risk industries.

Table 7 Heterogeneity tests for litigation risk

Secondly, considering that extremely opportunistic behavior exposes firms to greater litigation risk (Kim & Skinner, 2012), we follow the approach of Quan and Zhang (2021) to measure a firm's litigation risk by examining whether it has violated regulations in a given year. We define the variable Violation as an indicator that takes a value of one if a firm has violated a regulation, and zero otherwise. The findings reported in Column 2 of Table 7 reveal a statistically significant positive coefficient estimate for Justice × Violation. These results are consistent with our expectation based on Hypothesis 2a. Overall, the findings displayed in Table 7 demonstrate that the impact of judge independence on CSR is more pronounced among firms with higher litigation risk.

Political Connection

We employ two measures to capture political connections. Firstly, SOEs are often perceived as government entities established to maintain social stability. To capture this, we create a binary variable called SOE, which takes a value of one if a firm's controlling shareholder is the government or its agent, and zero otherwise. Secondly, politically connected non-SOEs also serve as government agencies to achieve state political objectives (Jiang & Kim, 2020). We obtain information on the background and work experience of CEOs and chairs from the China Listed Firm's Corporate Governance Research Database in CSMAR. Following the approach of Liu et al. (2016), we construct the variable Connection as a binary indicator that takes a value of one if the CEO or chair has a background or work experience in government or military service, or has served as a deputy of the Provincial/National People's Congress or the Chinese People's Political Consultative Conference. Otherwise, it takes a value of zero.

In the baseline regression, we expand the model by introducing the political connection indicator and its interaction with Justice. The results, presented in Column 1 of Table 8, reveal that the coefficient of Justice × SOE is statistically significant and negative. Moreover, in Column 2 of Table 8, we limit the analytical sample to non-SOEs and include both the Connection variable and the interaction term Justice × Connection in Eq. (1). The estimated coefficient of Justice × Connection is also statistically significant and negative, in line with our expectation based on Hypothesis 2b. These findings indicate that the positive impact of judge independence on CSR is diminished for firms with political connections.

Table 8 Heterogeneity tests for political connection and local government intervention

Local Government Intervention

We utilize survey data from the China Justice Index 2014 to examine our prediction. The China Justice Index 2014 is based on direct survey data collected in 2014, which assesses the public's trust in local judicial justice in nine provinces.¹² To measure local government intervention, we create two variables. Firstly, Low Justice takes a value of one if the survey score for "Do you think the local judicial system will settle disputes justly?" in the province where a firm is headquartered is lower than the sample median, and zero otherwise. Secondly, High Intervention is set to one if the survey score for "How likely is it that judges in your district will be interfered with by the local government in the course of their work?" in the province where a firm is headquartered is higher than the sample median, and zero otherwise.

In our CSR regressions, we augment the model by including the proxy variables for local government intervention and their interactions with Justice. The results, presented in Columns 3 and 4 of Table 8, indicate that the estimated coefficient of Justice × Low Justice is positive and statistically significant at the 1% level. This suggests that the effect of judge independence on CSR is more pronounced in provinces with lower levels of judicial justice. These findings imply that a greater increase in judicial justice contributes to more improvement in CSR among firms. Furthermore, the estimated coefficient of Justice × High Intervention is positive and significant, indicating that the effect of judge independence on CSR becomes stronger in provinces with higher levels of local government intervention.¹³ Overall, the results demonstrate that the influence of judge independence on firms' CSR performance is more prominent in provinces characterized by higher levels of local government interference. This finding aligns with our expectation based on Hypothesis 2c.

Discussion and Conclusion

Our study examined the impact of judge independence on CSR in the context of China's judicial delocalization reform. Using a staggered DID test, we found a significant positive association between judge independence and CSR. The analysis also revealed a positive link between judge independence and corporate lawsuits, indicating that the reform reduced local government interference and improved the fairness of the legal system. Firms faced with increased litigation pressure following enhanced judge independence were more likely to engage in CSR activities as a means of protection. The effect of judge independence on CSR was stronger for firms with higher litigation risk but weaker for politically connected firms. Furthermore, the positive relationship between judge independence and CSR was more pronounced in provinces with heavy local government intervention or weak judicial justice prior to the reform.

This paper contributes to the existing literature in two significant ways. Firstly, our study enhances the literature on CSR by providing empirical evidence of the positive impact of judge independence (Lei & Li, 2022; Liu et al., 2022). The independence of judges plays a crucial role in shaping the institutional environment within which companies operate. Our research yields novel insights into the effects of judge independence on CSR, highlighting how increased independence, achieved by mitigating local government interference in the judicial process, leads to a higher level of CSR engagement.

Secondly, our research advances the understanding of CSR motivations. CSR serves as a risk management tool that offers protection against adverse events (Miras‐Rodríguez et al., 2015; Barnett et al., 2018; Xue et al., 2022). Our findings reveal that increased judge independence results in a higher number of corporate lawsuits, prompting firms to adopt more socially responsible practices as a means of mitigating potential legal risks. This study contributes to the literature by shedding light on the impact of judge independence on CSR, particularly within the context of risk management motives.

Our study is useful to policy makers because the finding provides evidence that the judicial delocalization reform has led to more socially responsible practices, which supports the implementation of this reform. The judicial reform guarantees that judges try cases according to the law rather than following the instructions of local governments, representing a remarkable shift to respect for the law in China (Lei & Li, 2022). More importantly, our new evidence suggests that great independence of judges has a positive impact on the willingness of firms to be socially responsible. These findings can inform policy makers to create an environment for judges to be able to hear cases independently under the law, with no interference from any source, to promote CSR.

Appendix

See Table 9.

Table 9 Variable definitions