Moderating role of risk management effectiveness on corporate social responsibility- corporate performance relationship

Abstract This study is designed to provide empirical evidences on the conditional influence corporate social responsibility (CSR) on corporate accounting and market performance. This study considers risk management (RM) as a conditional or moderating variable. Accounting performance was proxied by return on assets (ROA), earnings per share (EPS), and net profit margin (NPM), while market performance was proxied by Tobin-Q and stock price. Furthermore, those proxies form corporate performance variable based on factor analysis. CSR measured using the Global Reporting Initiative (GRI G4) Index. Meanwhile, RM was measured using total risk management proxy. By applying purposive sampling method, 253 non-financial companies were selected as a sample of this study. The Structural Equation Modeling (SEM) with the WarpPLS approach was used for data analysis. This study found that the corporate social responsibility has a positive effect on corporate performance, and the relationship between corporate social responsibility and corporate performance will be stronger if corporate risk management is applied optimally. The findings of this study imply that the good corporate performance will be achieved since a company discloses more CSR information and runs effective risk management as well.


Introduction
Corporate activities are expected not only to focus on profit but also to be sensitive to environmental sustainability (planet) and community welfare (people) or known as the triple bottom line concept (Friedman, 1982) which was later popularized by Elkington (1997). Corporate social responsibility (CSR) includes three main dimensions, namely profit, community empowerment (people), and preservation of nature/earth (planet). Simionescu and Gherghina (2014) argued that the concept of CSR has developed in the last two decades, both in terms of literature and practice. From the literature perspective, CSR is divided into two areas: development studies (definition, theory, standards, and frameworks), and studies of the relationship between CSR and corporate financial performance and market performance. Although the results of research on the relationship between CSR and company performance have not shown a consistent relationship. However, subsequent literature developments have an impact on the practical side, where companies have begun to realize the importance of implementing CSR which is understood to improve company performance so that many companies have started to implement CSR guidelines into their corporate business strategies (Porter & Kramer, 2002).
The motivation for CSR disclosure is also realized by company management that stakeholders are starting to need information related to the company's CSR activities so that CSR disclosure becomes the company's good image and reputation in the eyes of its stakeholders (Boilar, 2013). This has been proven by the results of empirical research, among others, Brinkman (2003) concluded that corporate commitment to CSR will improve company performance. The research results of Simionescu and Gherghina (2014) show that CSR has a positive effect on market performance (EPS), as well as the results of research by Godfrey et al. (2009) who found that CSR disclosure can protect the company and will improve company performance. The same thing was concluded by Nekhili et al. (2017) stated that CSR disclosure has a positive effect on firm value (Tobin's Q). However, in terms of corporate performance measurements, most previous research using both fundamental performance and market performance individually. This study contributed in developing corporate performance measurement by combining some proxies of accounting and market performance based on factor loading of each proxy.
On the other hand, realizing the importance of CSR, the government makes regulations. In Indonesia Limited Liability Company, Law No. 40 of 2007 has stipulated that companies conducting business activities related to or impacting natural resources must implement and disclose social and environmental responsibilities. This law is the basis for implementing CSR reporting in Indonesia, even though the reporting is still voluntary. Until now, many companies in Indonesia have started to disclose their social and environmental activities, using various media, such as annual reports, sustainability reports, their respective websites, online media, and advertisements (Perks et al., 2013).
Furthermore, according to Simionescu and Gherghina (2014) that when a company implements CSR, the company's risk will be more controlled, so that the company can avoid losses and will gain a competitive advantage. In this case, CSR creates risk management benefits for the company (Kim et al., 2020), so that investment in CSR is not only intended to minimize corporate risk (Orlitzky & Benjamin, 2001;Sun & Cui, 2014), but also opportunities to improve performance. This research aims to prove that there is an increase in company performance after implementing risk management, such as research by Gordon et al. (2009), Hoyt and Lybenberg (2011), Florio and Florio and Leoni (2016, Saiful et al. (2019), and Silva et al. (2018) who show that the implementation of risk management can improve company performance. Conversely, several studies show that the implementation of risk management has no effect on company performance (Husaini Pirzada et al., 2020;Pagach & Warr, 2010;Wirawan et al., 2020).
Since the results of previous studies about the relationship between CSR and company performance are inconclusive, it indicated this relationship should be conditional. Waheed et al. (2021) explore institutional investors as conditional variables in the relationship between CSR and corporate market performance. They found that the influence of CSR on company performance is getting stronger for the companies with the proportion of shares owned by institutional investors is relatively large. However, some study found the potential of risk management as the conditional variable for CSR-company performance relationship. This is reasonable since risk management has been explored by some previous study for the case of corporate governance-firm performance relationship. It is also because of a substitute or complement point of view for CSR disclosures and company performance relationship. Therefore, the application of risk management is expected to strengthen the relationship between CSR disclosure and company performance. So, testing the interaction of risk management implementation and CSR disclosure is the main novelty of this study. This study also contributes in terms of risk management and company performance measurement point of view. Some researchers measured risk management using: (1) enterprise risk management (ERM) disclosure (Florio & Leoni, 2016;Gordon et al., 2009;Husaini Pirzada et al., 2020;Silva et al., 2018;Wirawan et al., 2020) and (2) the presence of a risk management committee (Hoyt & Lybenberg, 2011;Pagach & Warr, 2010). This study measure risk management using total risk management developed by Andersen (2008) with additional relevance and validity appropriate test. The study also using combination of both financial performance and market performance for company performance measurement This article is presented as follows. In the second section, a literature review and hypothesis development are presented. The third part describes the research methodology in the form of population, sample and data source, variable, measurement, and model, as well as the results of testing the validity and reliability of constructs. The fourth part presents the research results and finally, the fifth part is the conclusion.

Agency theory, stakeholder theory and signaling theory
This research explored three theories related to the cause and effect of CSR disclosure practices, total risk management, and company performance. The first theory related to the problem of this research is agency theory, which in this theory explains that the agency relationship between agents (management) and principals (shareholders) creates a gap or is often referred to as information asymmetry. This is where the role of disclosing information such as CSR plays a role in minimizing this gap, although it must be paid by an increase in agency costs to provide this information (Jensen & Dan Meckling, 1976), as well as the implementation of risk management also aims to reduce agency costs with the aim of monitoring agent behavior and at the same time creating benefits on increasing firm value (Schröck, 2002).
The second theory of this research is the stakeholder theory. In this theory, Freeman and Reed (1983) suggest that stakeholders, either as a group or individually, can influence the company in achieving its goals. Therefore, company management is expected to be able to carry out important activities related to stakeholders and report on these activities, one of which is the company's CSR activities, so that the company's existence does not conflict with the interests of the stakeholders. Thus, the company will find it easier to achieve its goals by improving company performance.
Finally, the signaling theory that assumes all information released by the company is a signal in the form of good or bad news. Therefore, information on CSR disclosure and risk management is an information prospect for investors and other stakeholders that have an impact on performance.

CSR and corporate performance
The relationship between CSR and corporate financial performance has long been debated in the various scientific literature. However, the argument that motivates companies to invest in CSR programs is based on stakeholder theory (Argandoña, 1998;Freeman, 1984). Stakeholder theory underlying the management policy to invest in CSR program in order to reduce transaction costs and to improve the company's future performance. Moreover, CSR disclosure will reduce information asymmetry between managers and stakeholders (Kordsachia, 2021). Marti et al. (2015) found that companies with better implementing CSR strategies report higher financial performance compare to companies with implementing traditional management strategies. Meanwhile, Godfrey et al. (2009) stated that the implications of CSR disclosure can protect companies and at the same time improve company performance. In contrast, Hashim et al. (2019) examined 17 telecommunication companies in ASEAN countries that did not find a significant relationship between CSR and company performance. Elouidani and Zoubir (2015) documented a negative and significant effect between CSR and corporate financial performance.
In the context of market performance, Hannah et al. (2020) and Simionescu and Gherghina (2014) documented that the implementation of CSR has a positive effect on market performance for EPS proxy. Moreover, Nekhili et al. (2017) stated that CSR disclosure has a positive effect on market performance for Tobin's Q proxy. Meanwhile, Odeh et al. (2020) suggests that the implications of increasing information and understanding of corporate stock prices can be explained by spending on CSR. Awaysheh et al. (2020) also found that best-in-class companies receive higher relative market ratings than their industry counterparts due to the company's attention to CSR. Ender and Brinckmann (2019) also conclude that news related to CSR has a significant influence on the shareholder value of a company represented by its share price. In addition, the company CSR activities positively and significantly impact on stock returns (Adina & Oanea, 2017).
Furthermore, Ferrero and Frías-aceituno (2015) investigated the relationship between CSR and corporate financial performance, using 1960 listed multinational non-financial companies.
The results showed evidence of a positive relationship between CSR and company financial performance. Tse Hou (2018) also examines the relationship between CSR and corporate financial performance in the electronic industry in Taiwan, by using CSR awards as an indicator of social responsibility, the findings showed that companies which have better social responsibility achieve superior financial performance than companies that do not have CSR initiatives.
Akben-Selcuk (2019), using a sample of non-financial public companies listed in the Borsa Istanbul (BIST) −100 index during the 2014-2018 period, shows that corporate social responsibility has a positive relationship with corporate financial performance. Likewise, the results of Famiyeh's (2017) study in Ghana show that companies tend to enjoy overall performance improvements such as profitability, sales growth, better return on investment, and increase market share when they invest in CSR activities. Wang et al. (2016) with a meta-analysis approach, based on research results from 42 empirical studies on the relationship between CSR and financial performance, conclude the general argument that CSR improves corporate financial performance, in which financial performance is positively related to previous social responsibility. Likewise, the research results of Galdeano et al. (2019) on the banking industry in Bahrain, that structural equation modeling concluded that CSR has a significant impact on financial performance. Therefore, the first hypothesis of this study is as follows.

H1
: CSR disclosure has a positive effect on corporate performance

CSR, risk management, and corporate performance
The risk management perspective views CSR as a tool to protect reputation value, in which companies begin to develop technology to measure and control potential environmental and social problems. CSR policies and activities are more focused on the activities of companies with the highest potential risk (Castelló & Lozano, 2009). Husted (2005) explains that the real options theory specifically shows that CSR is negatively related to business risk, which is the opposite view of corporate risk management (Devie et al., 2019). Lu et al. (2020) found that companies with better CSR performance tend to adopt integrated risk management practices. Moreover, the effective risk management will be ensuring the company's resilience to risk and followed by better organizational performance (Khan et al., 2020). Tran et al. (2019) found that CSR is the optimal action to minimize risks and improve financial performance of Vietnamese textile companies. Albuquerque et al. (2018) provided empirical evidence on the relationship between CSR reduces systematic risk and increases firm value. Andersen (2008) states that the idea of risk management is the ability to respond to market factors outside management's control to stabilize company earnings. This in turn will lead to increased trust by investors and stakeholders which results in improved performance. Several studies have proven that risk management can improve company performance (Florio & Leoni, 2016;Gordon et al., 2009;Hoyt & Lybenberg, 2011;Kommunuri et al., 2015;Silva et al., 2018). Silva et al. (2018), Husaini Pirzada et al. (2020, Saiful et al. (2019) also found that the implementation of RM can improve company performance for Indonesian public listed companies. However, Pagach and Warr (2010), Wirawan et al. (2020), Husaini Pirzada et al. (2020 found that RM has no effect on company performance. This inconsistency of previous studies findings indicate that the relationship between CSR and risk management is a substitution relationship so that it can strengthen the relationship between CSR disclosure and company performance. Bassen et al. (2006) stated that the implementation of CSR makes risk management more controlled and the company can reduce losses to gain a competitive advantage so that it has the opportunity to create value for the company.
Furthermore, Sun and Cui (2014) found that CSR has a strong influence in reducing the risk of default, this relationship is stronger for companies in highly dynamic environments than in static environments. Lee et al. (2015) and Devie et al. (2019) found that CSR has an effect on reducing corporate risk. In this case, CSR functions as a control mechanism to reduce optimal risk-taking deviations and helps reduce excessive risk avoidance (Harjoto & Laksmana, 2016).
Based on the above arguments, the second hypothesis of this study is as follows.

H2:
The relationship between CSR and corporate performance will be stronger when the implementation of corporate risk management is high.

Population, sample, and data source
The population of this study is non-financial companies listed on the Indonesia Stock Exchange (IDX). We applied purposive sampling technique based some criteria (see Table 1). Finally, 253 companies were selected as a sample of this study with 1265 observations for the years 2014-2018.
The sample of this study includes eight non-financial industries with the number of companies and the percentage of samples for each industry presented in Table 2. Based on Table 2, we conclude that the sample of this study refresent companies in eight industries proportionally dependence on the total numbers companies in each non-financial industries. Companies that use currencies other than Rupiah (USD) Companies that do not have complete data The number of companies that meet the criteria for the study sample 253

Variable, measurement, and model
The dependent variable in this study is corporate performance (CP) that measured by four accounting (financial) performance indicators (ROA, ROE, NPM, and EPS) and 2 market performance indicators (Tobin's Q and stock price). The independent variables of this study consist of CSR Disclosure and Total Risk Management. CSR disclosure is measured using the Global Reporting Initiative (GRI G4) Index, which amounts to 91 disclosure items. The CSR disclosure index is calculated by comparing the amount disclosed with the amount that should be disclosed. Disclosure is stated in a dummy, namely given a value of 1 if disclosed and 0 if not disclosed. The Total Risk Management (TMR) variable is measured by the standard deviation of sales compared to the standard deviation of return on assets (ROA) and standardized with industry risk. Furthermore, some of the control variables in this study are company size (SZ) and Leverage (LV), and company age (AGE).
The conceptual model of this research is in the following equation.
Data analyze using structural equation modeling (SEM) with partial least squares (PLS) approach (WarpPLS) or known as SEM-PLS because of two reasons. Firstly, SEM-PLS can analyze measurement models of reflective, formative, and latent variables with one indicator without causing identification problems (Solihin & Ratmono, 2013). Secondly, SEM-PLS also does not require the assumption of a normal distribution (Sholihin et al., 2011). Meanwhile, in order to determine moderating of TRM, we follow some data analyzing step that proposed by Baron and Kenny (1986) and Hair et al. (2011).

Validity and reliability
We run factor analysis in order to determine validity and reliability of variable measurement of this study. The results of the validity and reliability of the constructs of this study tests are presented in Tables 3 and 4. Table 3 shows that the loading of the CP variable indicator fulfills the convergent validity even though there is loading below 0.70, but the p-value is significant (<0.05). According to Ho and Taylor (2013) loading between 0.40 and 0.70 can be maintained if it has an impact on increasing the Average variance extended (AVE). Table 3 shows the results of the reliability and collinearity testing, that the composite reliability and Cronbach's alpha value for the CP variable is 0.644, and the AVE value is 0.518. These results meet the requirements for composite reliability and Cronbach's alpha>0.60 or have an AVE value above 0.50 (Fornell and Lacker, 1981). Furthermore, the value of Full collinearity VIFs for all variables  shows a value below 3.3 so that it can be stated that the model is free from collinearity problems.   Next, the CSR variable has an average of 0.173 with a standard deviation of 0.110. These results indicate that the average CSR disclosure of the sample companies is only 17.3% (still low), however, there are companies whose CSR disclosures reach 61.5%. The average risk of the company is 0.59 with a standard deviation of 1.26. This shows that the average application of corporate risk management in Indonesia is still relatively in the moderate category.

Descriptive statistics
1,927 with a standard deviation of 5,506, in this case the stock price of the sample companies also has quite a large variance. The Tobin Q variable shows an average of 1.627 with a standard deviation above the average of 2.256; however, these variable data does not have a large variance.

Hypothesis testing
The results of testing this research model indicate that both the first and second models meet the requirements of the goodness-of-fit model as follows. Based on the test results on both models, it can be stated that the R-Squared (R 2 ) value indicates the percentage of endogenous construct variants and can be explained by the predictor variables in the model, which means that the endogenous and exogenous variables have a direct or indirect causal relationship. Next, the ARS value shows the accuracy of the path model's ability to describe the influence between one independent variable with the predictive value (dependent variable).
This study did not occur multicollinearity because the VIF value which should have been ≤ 3,3 had been fulfilled, so that the inner model could be accepted. Likewise, the Average Full collinearity VIF is less than≤3.3, indicating that the model is free from problems of vertical, lateral collinearity and common method bias. The output shows that Q 2 (Table 4) is greater than zero, which is 0.239, so it can be  concluded that the research model shows good predictive validity and deserves to be continued. Furthermore, the results of hypothesis testing are presented in Table 6 and Figure 1 below.
The results of testing Hypothesis 1 show that CSR disclosure has a positive effect on company performance (β1 = 0.167, with p-value<0.001), these results indicate that the higher CSR disclosure, the company's performance will increase. The results of this test support hypothesis 1. The results of testing for hypothesis 2 indicate that there is a negative interaction between total risk management and corporate social responsibility (β1 = −0.101, with p-value<0.10). These results indicate that the implementation of risk management moderates the relationship between CSR and firm performance (pure moderation), or in other words that the relationship between CSR and company performance will be stronger when the risk is low or the company's total risk management is applied optimally. Therefore, CSR disclosure and risk management implementation are at least a complementary solution.
The results of the control variable test show that company size has a positive impact on CSR disclosure, this result is in line with (Dissanayak et al., 2019;Orazalin & Mahmood, 2019;Silva et al., 2018) that there is a positive relationship between company size and CSR. Large companies will tend to have more resources to achieve legitimacy, to demonstrate that their operating methods are compatible with social values (Guthrie & Parker, 1989;Ho and Taylor, 2007). The next control variable, namely leverage, has a negative relationship with CSR disclosure. This result is in line with the research of Orazalin and Mahmood (2019), in which companies that have high debt have a tendency to disclose lower CSR compared to companies that have low debt. The last control variable in this study shows that company age has a positive effect on CSR disclosure. These results are in line with research by Masum (2020) that company age has a significant positive effect on CSR. These results indicate that companies that have long been listed on the stock exchange tend to disclose more CSR than companies that have just listed on the stock exchange.

Discussion
The first finding of this study shows that CSR disclosure positively influence to company performance. This finding indicates that the higher availability of companies social responsibility activities will be followed by firm better performance. This finding in line with agency theory which state that corporate information disclosure including CSR plays an important role in order to minimize the information asymmetry between agents (management) and principals (shareholder). Moreover, information asymmetry reducing will of course be followed by company performance increasing. This result also supports the stakeholder theory which state the company's CSR activities disclosure is in line with the interests of stakeholders, thereby increasing company performance. This finding also indicate companies a positive signal or good news through CSR information disclosure, then investor will response to those signal that showed by company performance increasing. This result in line with Godfrey et al. (2009) In the context of risk management role, this study found that total risk management implementation strengthens the relationship between CSR and company performance. This finding is in line with agency theory which mentioned that the effective risk management can reduce agency costs in order to monitor agent behavior and at the same time create benefits for improving company performance. In this case, CSR is seen as a tool to protect reputation value, where CSR policies and activities are more focused on company activities with the highest risk potential or negatively related to business risk, which is an inverse view of corporate risk management (Devie et al., 2019). This finding also is in line with Lu et al. (2020) who found that companies with better CSR performance tend to adopt risk management practices, where effective risk management capabilities will ensure the company's resilience to risk, which will ultimately be achieved better organizational performance (Khan et al., 2020). The results of this study are also in line with Tran et al. (2019) who stated that CSR is the optimal action to minimize risk and improve financial performance. The same results were also reported by Albuquerque et al. (2018) that CSR reduces systematic risk and increases firm value. Moreover, the application of risk management can improve company performance (Florio & Leoni, 2016;Silva et al., 2018;Kommunuri et al., 2015;Hoyt & Lybenberg, 2011;Gordon et al., 2009;Husaini & Saiful, 2017;Husaini et al., 2020a;Husaini et al., 2020b). The results of this study are also in line with Bassen et al. (2006), Lee et al. (2015), Devie et al. (2019) and Sun and Cui (2014) stated that the implementation of CSR makes risk management more controlled and companies can reduce losses to gain a competitive advantage, so that they have the opportunity to create value for the company. CSR serves as a control mechanism to reduce optimal risk-taking deviations, and helps reduce excessive risk avoidance (Harjoto & Laksmana, 2016). The results of the control variable test show that company size has a positive impact on CSR disclosure, this result is in line with (Albuquerque et al., 2018;Dissanayake et al., 2019;Orazalin & Mahmood, 2019) that there is a positive relationship between company size and CSR. Large companies will tend to have more resources to achieve legitimacy, to demonstrate that their operating methods are compatible with social values (Guthrie & Parker, 1989;Ho and Taylor, 2007).
The findings of this study ensure that the positive influence of CSR on company performance is very conditional. One of the conditions that strengthen such influence is the effective implementation of risk management. Therefore, it is not enough for companies to disclose information on their social activities when hoping to achieve good performance, but they must be able to implement effective risk management at the same time.

Conclusion
This study aims to examine the effect of CSR disclosure on the performance of public companies in Indonesia and to examine the application of total risk management in explaining the relationship between CSR disclosure and performance of public companies in Indonesia. The results showed that CSR disclosure has a positive effect on company performance. This result indicates that the higher the company's CSR disclosure, the company's future performance will increase. Next, total risk management moderates the relationship between CSR and company performance. These results indicate that the company's performance in the future will increase when the company can optimize the increase in total risk management. These results support the agency theory that disclosure of information such as CSR plays a role in minimizing information asymmetry between agents and principals. The results of this study also support the stakeholder theory that disclosure of the company's CSR activities is in line with the interests of stakeholders, thereby increasing company performance. The results also support the signaling theory that CSR disclosure is good news information that has an impact on company performance. The results of this study are an indication for companies in disclosing CSR and implementing total risk management, because the implementation of total risk management is mutually substituted with CSR, this is a positive signal for investors in investing.
The results of the study contributed to providing a strong reason for the government to require companies to carry out social activities and discourage these activities. Based on the results of this study, regulators can evaluate the application of Limited Liability Company Law No. 40 of 2007, especially related to CSR activities and risk management models that have been implemented by Indonesian public companies. The results of this study also contribute to investors and regulators in making decisions and improving policies that have an impact on strengthening the Indonesian capital market, which ultimately has an impact on economic growth. Saiful Saiful 2