Navigation – Plan du site

AccueilNuméros25-1Corporate Social Responsibility a...

Corporate Social Responsibility and Firm Default Risk Mitigation: The Moderating Role of the Legal Context

Responsabilité sociale des entreprises et réduction du risque de défaillance : l’effet modérateur du système juridique
Mohamad Hassan Shahrour, Isabelle Girerd-Potin et Ollivier Taramasco

Résumés

Cet article explore dans différents contextes institutionnels l’impact de la Responsabilité Sociale des Entreprises (RSE) sur le risque de défaillance. En utilisant les notations de RSE et de crédit de 1153 entreprises dans 45 pays, nous constatons que la RSE réduit le risque de défaillance et que l’effet est plus fort dans les pays de civil law, en particulier de droit allemand et scandinave, que dans les pays de common law. La composante environnementale est celle qui a l’impact le plus fort. Les résultats, utiles aux entreprises pour leur survie, sont robustes à l’usage de différentes mesures et classifications.

Haut de page

Notes de l’auteur

Acknowledgement: We are grateful to the Vigeo Eiris (a part of Moody’s ESG Solutions Group) social rating agency for their generosity in providing the ratings they produce. We would like to thank the French National Research Agency in the framework of the Investissements d’Avenir program (ANR-15-IDEX-02) for their support.

Texte intégral

1. Introduction

1Since CSR became an object of academic study, a fundamental question has arisen regarding the impact of CSR on a firm’s performance and risk. Can a firm improve its performance and reduce its risk by engaging in CSR? Despite the huge amount of research regarding the impact of CSR on firms’ financial performance, neither a unified definition of CSR (Campbell, 2006; Martínez et al., 2016) nor certainty about its impact (Margolis et al., 2009; Orlitzky et al., 2003; Wallis and Klein, 2015) have been reached. If the literature does not provide a simple answer, it may just be that there is none. Researchers have gradually been led to explore the factors that may influence the relationship between the socially responsible (SR) behaviour of firms and their financial outcomes. Assessing the impact of CSR on performance and risk requires specifying the context within which the firm operates and the targeted stakeholders.

2The existing literature is not solely in favour of CSR. As being socially responsible represents an additional cost for firms, it may not find favour with small corporations with limited resources nor benefit shareholders’ profit maximisation interests. According to related arguments based on shareholder theory, if the firm policy is devoted exclusively to the interests of its shareholders, with socially responsible (SR) commitment considered only as a cost, the firm would never pursue a SR policy. In contrast, stakeholder theory suggests that being socially responsible provides an advantage for firms and enhances society’s welfare. As a result, engaging in CSR is a firm’s choice to find a balance between the additional costs of an SR policy and the risk reduction it generates.

3In recent decades, CSR has received increasing attention in both academic literature and the business world. Recent research has examined whether CSR can be seen as a risk mitigation tool. Within that field, some studies focus on linking CSR to default risk, which is an important forward-looking indicator of a firm’s financial health (Badayi et al., 2020; Boubaker et al., 2020; Goss, 2009; Jiraporn et al., 2014; Rizwan et al., 2017; Sun and Cui, 2014). Although the number of studies is rather small, they provide significant results regarding CSR impact on reducing default risk. Those studies are mainly carried out on US samples. Is that impact specific to the United States? To answer that question, we examine whether CSR influences default risk level by considering an international sample.

  • 1 For an extensive literature review on the latter, please refer to Margolis et al. (2009).
  • 2 By definition, stakeholders are any group or individual who can affect or is affected by the achie (...)

4Even if that relationship also holds in other countries, that does not mean that it will be of the same magnitude. For instance, “CSR activities can be attributed to variation across nation-level institutions” (Ioannou and Serafeim, 2012, p. 835). Scholars state that the institutional framework of countries influences firms’ social behaviour (Ioannou and Serafeim, 2012; Walker et al., 2019) and moderates the CSR–firm value relationship (El Ghoul et al., 2016).1 On a normative management level, firms have to form their sustainable vision by integrating the organisational culture (Baumgartner, 2014). In that vein, the aforementioned causal relationship in, say, the French context can differ from that in the US context. That highlights the importance of considering the institutional differences in cross-country samples. Few studies have assessed the moderating role of the institutional context in the relationship between CSR and firms’ financial outcomes. On a related theme, Yan et al. (2019) show that the degree of prevalence of financial logic in a country has an impact on the creation of SR funds. That effect is itself moderated by other institutional forces, namely labour unions, churches and the representation of green parties in government. Institutions can be seen as the norms and rules that guide interactions within societies (Scott, 1995). Aguilera and Jackson (2003) state that researchers contrast two dichotomous models of Anglo-American and continental European corporate governance through applying various labels. For instance, the Anglo-American model is also labelled as outsider, common law, liberal law, shareholder-centred or market-oriented, while the continental European model is labelled as insider, civil law or stakeholder-centred, among others. Williams and Aguilera (2008) state that the importance of CSR issues in business operations is influenced by a country’s business culture (in particular, whether a country is more shareholder- or stakeholder-oriented). As CSR can be seen as a response to stakeholder2 demands (Jackson and Apostolakou, 2010), the impact of CSR on default risk is likely to vary between shareholder-oriented countries and stakeholder-oriented countries. In shareholder-oriented countries, individuals are expected to place less value on SR behaviour, and more attention is paid to shareholders. Therefore, firms’ engagement in CSR activities could be a waste of resources if it does not create sufficient positive feedback from employees, customers, suppliers and others. As a result, we expect to observe little or no relationship between CSR and default risk in shareholder-oriented countries. In contrast, in stakeholder-oriented countries, stakeholder demands are valued. The benefits for firms of being committed to CSR include greater customer loyalty and enhanced employee commitment and loyalty. As such, CSR behaviour positively influences the firm. Hence, we expect to observe a negative relationship between CSR and default risk in those countries.

5It should be noted that the classification of countries according to whether they are shareholder- or stakeholder-oriented is a delicate issue. Regarding the institutional environment defined by Scott (1995), i.e. it is composed of three main elements: regulative, normative and cultural-cognitive, we focus mainly on the regulative component because the legal environment within which firms operate is a key determinant in firm policies. A legal system has implications far beyond visible contracts in that it reflects society as a whole and has a strong influence on the prevailing institutional logics (Kock and Min, 2016). It is associated with individual behaviours, desires and beliefs. Regarding CSR, the legal system – as a mirror of society – has an impact on the effects of a firm’s SR policy on performance and risk. Therefore, we consider the legal origin to distinguish between shareholder- and stakeholder-oriented countries, and we use it as a moderator influencing the causal effect of CSR on default risk. A moderator variable should have an effect on both dependent and independent variables. The legal origin, which we explain hereinafter, satisfies that condition.

  • 3 According to Metzler (2001, p. 321), “establishing and maintaining organisational legitimacy is at (...)

6According to the traditional view in law and finance, firms tend to comply with the applicable regulations to avoid lawsuits. However, obeying the law is not just about being a “good corporate citizen” (Colbert et al., 2003); it is above all a means of achieving legitimacy (Windolph et al., 2014).3 The importance of the legal environment is recognised, as it shapes firms’ policies and priorities (e.g. shareholder wealth vs stakeholder demands). In other words, a country’s legal origin may shape “various country-level institutions and the firm-level contracting environment” (Liang and Renneboog, 2017, p. 855). That differentiation could be notably recognised by commercial laws in terms of the protection and importance given to shareholders/stakeholders. La Porta et al. (1998) distinguish between the two main legal origins that shape commercial regulations: common law (consisting of English law countries) and civil law (consisting of French, German and Scandinavian law countries). In the context of CSR, the legal environment will shape the contracts between shareholders and other stakeholders in different ways (e.g. through its effect on corporate governance). Common law countries are more “shareholder-oriented” and favour shareholder protection, whereas civil law countries are more “stakeholder-oriented” and favour stakeholder protection. Given the difficulties in classifying countries into shareholder-oriented and stakeholder-oriented, we deepen the analysis by classifying more precisely the civil law countries with German, French and Scandinavian law and by relating the results to the Hall and Soskice (2001) classification of the varieties of capitalism.

7Our intuition in this study is that the CSR–firm default risk relationship depends on how individuals and national legislations view and protect stakeholders. With that in mind, we examine the impact of CSR on firm default risk and consider the legal origin as a moderator of that relationship. The overall results suggest that CSR is negatively linked to default risk (i.e. higher CSR ratings result in higher credit ratings and higher Altman Z scores) and that the impact of CSR on default risk is stronger in civil law countries (especially in German and Scandinavian law countries) than in common law countries. We also examine the relationship between the sub-ratings of CSR and credit ratings. The results reveal that the environment component has the highest impact on default risk. We provide several robustness tests to ensure the validity of our results.

8Our findings help in understanding what determines firms’ SR behaviour and what shapes the CSR–firm default risk relationship. This paper contributes to the literature by broadening the scope of the research by using an international sample while offering a complementary view by addressing the moderating role of legal origins. We shed light on the results with additional analysis based on the literature on varieties of capitalism. This research has several potential implications for risk managers, portfolio managers, corporations and investors, as discussed further in the paper.

9The paper proceeds as follows: Section 2 lays out the theoretical background and hypotheses development. Section 3 describes the data and the empirical methodology. Section 4 reports the empirical results and robustness tests. Section 5 concludes the paper.

2. Theoretical Background and Hypotheses Development

2.1. Theoretical framework

10Over recent decades, various theories have been proposed to explain what constitutes CSR and why its impact on firms varies between countries. Stakeholder theory and institutional theory can provide a prologue to this paper, particularly in terms of explaining how regulations within countries − more precisely, the legal origin and stakeholders’ legal protection − can determine firm and individual behaviour and the resulting impact on the relationship between SR commitment and default probability.

  • 4 For further explanations, please refer to Henisz et al. (2014).

11Stakeholder theory – in its instrumental view – suggests a positive relationship between CSR and a firm’s financial performance (Freeman, 1984; Orlitzky et al., 2003). According to that theory, firms are concerned about all groups or individuals that are affected by their activities. The reasons behind that can be either normative (e.g. for ethical reasons) or instrumental (e.g. for financial reasons).4 By relying on that theory, scholars argue that CSR contributes to building a good reputation for the firm (Benlemlih and Girerd-Potin, 2017; Stellner et al., 2015), increases customer loyalty (Arouri and Pijourlet, 2017; Lichtenstein et al., 2004; Luo and Bhattacharya, 2006) and influences financial performance (Brooks and Oikonomou, 2018; Flammer, 2015). On the other hand, the good stakeholder relationship established through CSR means that those firms are considered desirable partners and therefore attract high-quality employees. According to Edmans (2011, 2012), that could be positively correlated with shareholder returns. Moreover, CSR can result in better financing opportunities by decreasing agency costs and information asymmetry (Cheng et al., 2014; El Ghoul et al., 2016). Such results are in line with the literature asserting that firms engage in profit-maximising CSR (McWilliams and Siegel, 2001). In that vein, CSR is seen as value-creating (based on the promise that CSR is a differentiation strategy) and, as argued by Albuquerque et al. (2019), a way of reducing systematic risk.

12Institutional theory, on the other hand, states that institutional contexts influence firm behaviour (Kim et al., 2013; Morgan and Kristensen, 2006). Scholars refer to those contexts as the institutional environment, where firm practices are a function of rules determined by that environment. The existing theory has generated valuable insights into the process of understanding what influences the organisational adoption of homogeneous practices (Yin, 2017). Since this paper focuses on legal origins, we use the regulative component of the institutional environment to understand variations in the impact of CSR on firm default risk among firms operating in countries with differing legal origins.

2.2. Hypotheses on the link between CSR and firm default risk based on the theory and the empirical literature

13Orlitzky and Benjamin (2001) find that firms enjoying higher social performance have lower financial risk. Boutin-Dufresne and Savaria (2004) show that a negative relationship exists between CSR and firm-specific risk, where the residual volatility obtained for responsible portfolios is lower than that of irresponsible portfolios. Mishra and Modi (2013) find that CSR significantly reduces idiosyncratic risk and that financial leverage has a moderating effect on that relationship. The impact of CSR on risk reduction is lower for firms with a high level of financial leverage. Hsu and Chen (2015) highlight an inverse relationship between CSR performance and credit risk such that engaging in CSR activities reduces firms’ financial risks. Jo and Na (2012) find that CSR commitment is inversely related to a firm’s total risk. Benlemlih and Girerd-Potin (2017) show that CSR reduces idiosyncratic and systematic risks only in stakeholder-oriented countries, not in shareholder-oriented countries.

14In terms of default risk, recent studies by Boubaker et al. (2020) and Badayi et al. (2020) examine the impact of CSR on the probability of default. The former study considers a sample of 1,201 US firms over the period 1991–2012, with the MSCI ESG ratings used as a CSR measure and the Altman (1968) model as a default risk measure. The latter study considers 496 firms distributed over 17 developing countries over the period 2010–2017, with Thomson Reuters Datastream ESG ratings used as a CSR measure and, similar to the former study, the Altman (1968) Z-score as a default risk measure. Both studies confirm the risk mitigation effect arising from the CSR commitment.

  • 5 Primary stakeholders-related.
  • 6 Secondary stakeholders-related.

15Rizwan et al. (2017) examine the CSR–firm default risk relationship using data from 1,119 non-financial US firms during the period 2000–2012. Those authors distinguish between technical CSR (TCSR)5 and institutional CSR (ICSR).6 Using Merton’s (1974) model as a default risk measure provides results indicating that a high CSR index score can result in a lower probability of default. TCSR appears to have a significant negative relationship with default risk, whereas ICSR has an insignificant one.

16Based on a sample of 829 observations for 303 US firms, Sun and Cui (2014) examine the impact of CSR on default risk and how it can help in risk reduction. Those authors examine the impact of CSR on default risk while considering the moderating effects of environmental dynamism/complexity and firm capability. The credit ratings provided by Standard and Poor’s are used in that study as a measure of firm default risk level. The results show that CSR can help reduce default risk and suggest that CSR generates confidence and welfare for debt holders. Moreover, the impact of CSR on default risk is stronger in less stable environments, thus it acts as insurance-like protection.

17According to the literature, financial distress may be associated with social factors (e.g. social pressures and management quality), and firms that enjoy high levels of CSR face a lower probability of default. For instance, given that CSR is related to management quality (Goss, 2009) and that management inadequacies are related to default risk (Altman and Hotchkiss, 2006), one would expect that firms with better management through CSR would experience less risk of financial distress. Goss (2009) examines the relationship between default risk and CSR by considering a sample of 1,295 firms and 93,062 observations covering the period 1991–2003. For that purpose, that author considers, along with other models (e.g. discrete hazard model), the Merton (1974) model as a default risk measure. CSR is measured using the KLD database. The results reveal that CSR can be considered as a significant determinant of distress and that firms that enjoy high levels of CSR face lower probabilities of default.

18Jiraporn et al. (2014) state that the level of a firm’s commitment to CSR influences its credit rating: “an increase in CSR by one standard deviation improves the firm’s credit rating by as much as 4.5%” (p. 505). Firms with higher credit ratings have better access to financing opportunities (Cheng et al., 2014), thus enabling them to meet their debt obligations and reduce default risk.

19The above developments ensure that CSR maintains stable relations with the financial community, reduces information asymmetry, builds customer loyalty, ensures stable cash flows, etc. Given that, we argue that CSR plays an important role in decreasing the level of firm default. Furthermore, international awareness of CSR is accompanied by a questioning of which of its components most influence default risk. The 2018 final report of the High-Level Expert Group on Sustainable Finance points out that “the stock of natural capital is currently at risk, as it is deteriorating beyond its rate of renewal” (p. 88). That highlights that environmental challenges go beyond climate change to the extent of potentially calling into question a firm’s reputation and wider issues such as human rights and health security. The environmental management literature indicates the importance of the environment in improving firms’ competitive positions and in defining CSR policy. The Porter hypothesis – formulated in 1995 – states that a firm can improve its long-term commercial competitiveness by following strict environmental regulations and strategies.

  • 7 They refer to it as the “green swan” risk, which 0063ould be behind the next systemic financial cri (...)

20Chava (2014) shows that investors expect higher returns on stocks excluded by environmental screens. That study further suggests that environmentally sensitive lending and socially responsible investing can impact the cost of debt and the cost of equity of firms with environmental concerns. In line with that, Flammer (2013) argues that environmental CSR is a resource with decreasing marginal returns and insurance-like features. Cai et al. (2016) find that firms with a higher commitment to the environmental component can benefit from lower risk. That negative relationship holds even after controlling for various firm characteristics. Klassen and McLaughlin (1996) suggest that an improvement in corporate financial performance is a result of environmental performance. Other scholars find that the environment appears to be a key stakeholder concern in firms’ SR efforts (Kassinis and Vafeas, 2006; Welford et al., 2008). Amor-Esteban et al. (2018) state that a firm’s survival depends to a large extent on stakeholder relations and some studies link environmental records to stakeholder relations. Wang and Berens (2015) indicate that firms that have a strong environmental record could initiate and maintain better trust among broad groups of stakeholders. Gregory-Smith et al. (2017) find that consumers are willing to pay more for environmentally-friendly products. Sharfman and Fernando (2008) show that firms could benefit from a lower cost of equity by improving environmental risk management. Some recent working papers published by the Bank for International Settlements recommend considering environmental risks among the variables to be used when modelling systemic risk.7 Therefore, based on the environment’s importance for firms survival, we state that, of the various components, the environment component has the highest impact on the CSR–firm default risk relationship. The developed hypotheses are as follows:

H1: A negative relationship exists between corporate social responsibility and the default risk level of firms.

H2: The impact of the environmental component of CSR on firm default risk is higher than that of the other components.

2.3. The moderating role of the legal context in the CSR–firm default risk relationship

21How firms interact with societies and how they respond to the legal requirements is a wide area of research. For example, firms in different legal regimes respond differently to shocks (e.g. recessions). The relationship between firms and society could be acrimonious, but regulations often reshape that relationship to allow corporations to achieve their primary objective (i.e. profit maximisation) while preventing adverse effects on society. Generally speaking, a country’s institutional framework shapes firm and individual behaviour (Benlemlih and Girerd-Potin, 2017; El Ghoul et al., 2011; Liang and Renneboog, 2017; Matten and Moon, 2008), so their behaviour varies between countries. Amaeshi et al. (2016) indicate that “CSR requires certain conditions and institutional arrangements to function, i.e. strong government, market and civil society” (p. 135). Walker et al. (2019) add stakeholder pressures as a condition. In other words, a variation in the institutional framework will influence the magnitude of the causal effect of firms’ behaviours on firms’ outcomes. In that vein, the impact of CSR on default risk will be influenced by the legal origins where firms operate. To address the legal origin aspect, we rely on La Porta et al. (1998). Those authors argue that common law countries have stronger legal protection for investors than do civil law countries. That must then translate into shareholder privilege differences, the understanding of which paves the way to discerning variations in firms’ actions.

22The classification proposed by La Porta et al. (1998) makes a distinction between “shareholder-oriented countries” and “stakeholder-oriented countries”. Shareholder theory considers that the only social responsibility for a firm is profit maximisation, with taking stakeholder interests into account regarded as a waste of resources that reduces shareholder wealth (Friedman, 1962). In a shareholder-oriented country, that belief will be shared by the wider society, not just by shareholders and managers. In such a context, a firm that adopts a strong SR commitment would not be well perceived by shareholders or other stakeholders. In this respect, the legal system is not only about the organisation itself – otherwise it would be a driving force – but also about the way in which different stakeholders assess the SR behaviour of firms.

23Legal jurisdictions and national legislations have shaped multiple financial and social outcomes. Legal boundaries are at the heart of corporate finance and drive financial decisions, while institutional frameworks change the performance impact of firms’ social commitment. It is necessary to understand how the institutional and legal environment affects first the average level of CSR commitment and, second, the impact of that commitment on default risk.

24To address the first issue, Liang and Renneboog (2017) empirically examine the impact of legal origins on the extent of CSR commitment. Those authors distinguish between several legal origins, namely: English, French, German, Scandinavian and Socialist law origins. Firms operating under a civil law origin (i.e. French, German, Scandinavian and Socialist law origins) are more likely to behave in a socially responsible way (more precisely, firms operating in countries with a Scandinavian civil law origin have the highest CSR ratings). In contrast, firms operating under a common law origin (i.e. English law) do not consider social responsibility to be a priority.

25The impact of the institutional environment on the strength of the relationship between CSR and default risk is yet to be understood. Such understanding would amount to explaining why the legal system plays the role of moderator. To that end, it is necessary to examine the influence of legal origin on the relationship between CSR and default risk, all other things being equal. For the sake of illustration, we can consider the situation of two firms identical in all respects except for the legal system in their home countries. We expect that the impact of CSR on default risk would not be the same for the two firms. According to institutional theory, the institutional environment would influence the behaviours and beliefs of firms’ various stakeholders (e.g. customers, employees and suppliers) with respect to CSR. For instance, in civil law countries, customers tend to be more patient with SR firms and the employees are more loyal. The response and the impact in common law countries would not be of the same magnitude. It is expected that the SR commitment of a firm would have little or no positive impact on employee productivity, the cost of financing by stockholders, bondholders and banks, or on sales in relation to the firm’s social reputation. That then leads to the following hypothesis: the legal origin will affect the CSR–firm default risk relationship. From an econometrical perspective, the legal origin can be seen as an interaction term in the CSR–firm default risk relationship which influences the magnitude of the causal effect of CSR on default risk. However, the civil law classification is a generalisation of the French law origin, the German law origin and the Scandinavian law origin. In line with Liang and Renneboog (2017), we disentangle that classification because our interest is in investigating whether all the sub-classifications of civil law origin yield a higher impact of CSR on default risk than for firms operating under an English law origin. The hypotheses are as follows:

H3: The negative impact of CSR on firms’ default risk level is stronger in civil law countries than in common law countries.

H4: The negative link between CSR and default risk is lower in English law countries than in French, German and Scandinavian law countries.

3. Data and Empirical Strategy

3.1. Data

26To investigate the moderating role of the legal environment in the CSR–firm default risk relationship, this study considers the credit ratings reported by Bloomberg from the three major rating agencies, namely: Moody’s Analytics, Standard and Poor’s and Fitch Ratings. We consider credit ratings as a proxy for the dependent variable default risk. The CSR commitment, which is the explanatory variable of interest, is measured by Vigeo Eiris ratings. The control variables are extracted from Datastream and Compustat databases in US dollars. Our initial sample was subject to the availability of data in Vigeo Eiris. Following that, we extracted the credit ratings from Bloomberg. After merging the two databases with our control variables and removing firms with incomplete data, the final sample consisted of 1,153 firms for a total number of 3,946 observations covering the period 2004–2017. Table 1 provides a list of the countries under study in terms of their legal origin, as classified in the literature, together with the relative distribution of observations and the average CSR score for each legal origin.

Table 1 : Sample breakdown by country and CSR average by legal origin

27This table presents the sample distribution and the CSR average across 45 countries, for a total number of 3,946 observations covering 1153 firms. The F, G and S distinction made for civil law countries stands for French, German and Scandinavian law origin, respectively.

3.2. Measuring firm default risk

28Rating agencies aim to evaluate the creditworthiness of firms. Those agencies specialise in ranking firms according to their assets and liabilities, financial history and the extent to which they can meet their debt obligations. The major rating agencies include Moody’s, Standard and Poor’s and Fitch Ratings. For instance, the primary business of Moody’s is to provide credit ratings for the financial obligations of publicly traded firms, with the ratings ranging from Aaa to C. Lower ratings suggest higher default risks. Standard and Poor’s take credit rating to denote a firm’s ability to meet debt holders’ expectations. The credit rating can therefore be identified as a good measure of default risk. The Standard and Poor’s and Fitch Ratings categories range from AAA to D, with AAA being the lowest default risk and D the highest.

29In general, firms with high credit ratings are more likely to have access to financing opportunities, thus the likelihood is reduced that such a firm would not be able to meet its debts and obligations and, consequently, default. Firms with low ratings face high interest rates to compensate for the possible losses in the event of default. Various studies have considered credit ratings as a measure of firm default risk (Attig et al., 2013; Moussu and Ohana, 2016; Sun and Cui, 2014). Moreover, credit ratings can explain a large portion of credit default swap spread variability (Ötker-Robe and Podpiera, 2010). In that vein, we consider the credit ratings provided by Moody’s Analytics, Standard and Poor’s and Fitch Ratings as a proxy for default risk. Bloomberg publishes those ratings in the form of alphabet letters. We transform those ratings into numerical values ranging from 19 to 1, from the highest to lowest rating, respectively. To use the ratings as a default risk proxy, we calculate an average of the credit ratings provided by the rating agencies. To calculate the average, we follow the previous literature by taking into consideration the magnitude of each rating system so that all the classifications have the same magnitude (Gex et al., 2018). By implementing that strategy, we expect to have more accurate results and reduce the number of observations lost due to lack of information. Table 2 presents the method implemented to transform the ratings into numerical values while considering a unified magnitude for all the agencies.

Table 2 : Credit ratings and their respective numerical values

30This table presents the credit ratings provided by Moody’s Analytics, Standard & Poor’s and Fitch Rating. The table shows the transformation from letter ratings into numerical ratings while taking into consideration the same magnitude of all the ratings provided.

3.3. Measuring CSR

  • 8 The methodology used for constructing the global rating is that employed by Girerd-Potin et al. (20 (...)

31The social ratings come from Vigeo Eiris (a part of Moody’s ESG Solutions Group). They are widely used as a CSR proxy in the literature (Benlemlih and Girerd-Potin, 2017; Gex et al., 2018; Girerd-Potin et al., 2014; Liang and Renneboog, 2017; Quéré et al., 2018). The CSR score we calculate in this study consists of six sub-ratings concerning the firm, namely: human resources; environment; business behaviour; corporate governance; community involvement; and human rights. We follow Girerd-Potin et al. (2014) by creating a CSR global rating scale8 – hereafter, CSR score – that transforms the six sub-ratings provided by Vigeo Eiris into numerical values ranging from 0 to +4, where 0 represents the least socially responsible firms and +4 equates to the most responsible. If firms obtain the maximum score of +4, that does not mean that they have perfect social performance, just that they have the best level of behaviour among the rated firms in the same industry. Ceteris paribus, a CSR score larger than +2 implies that the firm’s behaviour is better than the average in its industry. In our regression models, we consider the calculated CSR score and the sub-ratings.

3.4. Country-level institutions

32We refer to La Porta et al. (1998), Spamann (2010) and Liang and Renneboog (2017) to classify countries according to their legal origin (i.e. civil versus common law countries) and to break that classification down into English, French, German and Scandinavian law origin. To differentiate between common law and civil law countries, we create two dummy variables. The first variable takes the value of 1 if the country has a civil law origin, and 0 otherwise. The second variable takes the value of 1 if the country has a common law origin, and 0 otherwise. To test that moderating effect, we create two interaction variables between the calculated CSR score and the dummy variables. The first interaction term refers to common law countries. The same methodology is applied when classifying countries as liberal and coordinated market economies (see Section 4.3). We also examine our hypotheses under two different approaches. In the first approach, we classify the socialist countries as German or French civil law (according to their pre-socialist law origin), as indicated in Spamann (2010) and Liang and Renneboog (2017). In the second approach, we follow Liang and Renneboog (2017) based on Aghion et al. (2010) to exclude firms operating in socialist countries, since they are in a transition state (regarding their economies) and not at equilibrium.

3.5. Control variables

33In line with the prior literature on default risk (Bharath and Shumway, 2008; Goss and Roberts, 2011; Hillegeist et al., 2004; Jackson and Rathert, 2015; Merton, 1974; Molina, 2005; Sun and Cui, 2014; Vassalou and Xing, 2004), we set various control variables in order to eliminate several factors that might affect the default risk level. The literature points to the importance of size in examining default risk, especially as large firms tend to be more diversified and face a lower probability of default. Therefore, we control for firm size. (SIZE) using the logarithm of total assets, and we expect to see a positive relationship with credit ratings (i.e. a negative relationship with default risk). Our second control variable is profitability (PROF),since profitable firms will yield a higher credit rating. We consider the return on assets as a measure of a firm’s profitability, and we expect to observe a positive sign. The third control variable is the book-to-market ratio (BTM), which is measured as the ratio of the book value of assets to the market value of assets. We use that measure to control for a firm’s growth, and we expect to find a negative relationship with credit ratings. The fourth control variable is leverage (LEV) Previous studies state that leverage has a great impact on default risk and can be considered as a major determinant. We control for leverage using the ratio of long-term debt to total assets. We expect to see a negative relationship with credit ratings.

3.6. Methodology

34By considering credit ratings as a proxy for default risk and as our dependent variable, we run various regressions on the interaction variables with our calculated CSR score while considering different control variables. Our initial models are as follows:

35Where (CR)it is the proxy of firm default risk measured by credit ratings; α represents the intercept; (CSR)it is the calculated score using the ratings provided by Vigeo Eiris; (CV*CSR)it is the interaction term between civil law and the calculated CSR score,while (CM*CSR)it is the interaction term related to common law; (SIZE)it is the logarithm of total assets; (PROF)it is the return on assets; (BTM)it is the book-to-market ratio; (LEV)it is the ratio of long-term debt to total assets; εit is the disturbance term. All the variables refer to firm i at time t.

  • 9 That is, for models that do not differentiate between countries in terms of legal origins or variet (...)

36To examine the various hypotheses, we develop several models based on our initial models. Model 1 is presented in Equation (1). Model 2 is similar to Model 1, but we replacewith the sub-ratings provided by Vigeo Eiris. Model 3 is presented in Equation (2). Model 4 examines the four different legal origins, namely: English, French, German and Scandinavian law, instead of the civil/common law classification. We follow the same methodology used for civil and common law classification to calculate the interaction terms related to those four legal origins. For models that consider the sub-ratings as a measure of CSR, we run a regression for each sub-rating separately to avoid multicollinearity. Our multivariate approach is based on annual data, and all the regressions are run with a time fixed effect, in addition to a country fixed effect where indicated.9 Moreover, for models that differentiate between legal origins, we examine those models with and without socialist countries.

4. Descriptive Statistics and Empirical Results

4.1. Descriptive statistics

37Table 3 provides the descriptive statistics for our default risk proxy, the CSR score, the sub-ratings used in calculating the score and the control variables used in the regression analysis. Credit ratings have an average value of 12.01 and a median of 12. Regarding our main explanatory variable, the average of our sample is 2.05, with a median of 2.09. The data ranges from 0 to 4. Table 3 also presents the descriptive statistics for the sub-ratings. The descriptive statistics of the control variables indicate that our sample is diversified regarding size, profitability and leverage.

Table 3: Descriptive statistics for the regression variables

38This table reports the descriptive statistics for the variables used in the regression for 1,153 firms. Panels A, B and C present the mean, median, standard deviation, minimum, maximum, first quartile, third quartile and the total number of observations for the dependent variable credit ratings, the independent variable CSR score with the six sub-ratings used in calculating this score, and the control variables, respectively. The Appendix provides definitions for all the variables.

39Table 4 presents the Pearson correlation coefficients for the regression variables. The explanatory variable (CSR) is positively related to credit ratings, with the sub-ratings of CSR also positively correlated. The relationship there between the environmental component of CSR and credit ratings is the highest among the sub-ratings. The control variables also meet our expectations regarding their relationship with the credit ratings. We employ the variance inflation factor test (VIF) for our initial model in Equation (2) to ensure that there is no correlation between the independent variables. Table 5 reports the results for the VIF test. The mean VIF is 1.73, which indicates that there is almost no correlation between the independent variables.

Table 4: The Pearson pairwise correlation coefficients between the variables

40This table reports the Pearson pairwise correlation coefficients for the variables used in the main empirical model presented in Equation (1). Coefficients, which are highlighted in bold, are significant at least at the 5% level. The dependent variable is the credit rating. The independent variable is CSR, and the control variables are: size (SIZE), profitability (PROF), book-to-market (BTM) and leverage (LEV). The table also reports the correlation of the sub-ratings used in calculating the dependent variable (CSR),namely: human resources (HR); environment (ENV); business behaviour (BB); corporate governance (CG); community involvement (CI); and human rights (HRT). The sub-ratings are used later in our regression analysis. The Appendix provides definitions for all the variables.

Table 5 Variance inflation factor results

41This table reports the variance inflation factor (VIF) test for the regression coefficients in our initial model (Equation 2), which assesses how much the variance of the regression coefficients increase if the predictor variables are correlated. The Appendix provides definitions for all the variables.

42Figure 1 represents the box­ plot concerning the median of CSR ratings in common and civil law countries. The median in civil law countries is higher than that in common law countries. Nevertheless, and to confirm whether firms’ social behaviour is more pronounced in civil law countries than in common law countries, we employ the non-parametric Wilcoxon rank-sum (Mann–Whitney) test to compare the median of the CSR ratings across different legal origins. We first test the normality of CSR ratings using the Shapiro–Wilk (1965) test of normality to reveal that the CSR ratings are not normally distributed. We then employ the non-parametric Wilcoxon rank-sum (Mann–Whitney) test in our univariate analysis. The preliminary results reveal that civil law countries consistently score higher than common law countries. We provide more in-depth analyses in the following section.

Figure 1: The median of CSR ratings between civil and common law countries.

43This figure represents the box plot of the CSR ratings’ median in civil and common law countries.

4.2. Results of the four hypothesis tests

  • 10 Since we consider the credit ratings as a default risk proxy, we expect to obtain a higher credit r (...)

44We consider Model 1 to examine the relationship between CSR and firm default risk, as alluded to in our first hypothesis (H1). To validate H1, we expect a positive relationship between our main explanatory variable (i.e. credit ratings). Table 6 reports the results of Model 1. The results reveal that CSR is positively and significantly related to credit ratings (at the 1% level), thus validating hypothesis H1 that CSR is negatively related to default risk.10 The control variables are significant and consistent with our expectations regarding their relationship with credit ratings. For instance,) and firms with high leverage will face a higher probability of default. That is in line with our expectations and the existing literature.

Table 6: The relationship between CSR and credit ratings

45This table reports the results from regressing the dependent variable CR over the main explanatory variable CSR, and the control variables, namely: SIZE, PROF, BTM, LEV, with year and country fixed effects. The table reports the coefficients of the explanatory variables, where the t-value is reported in parentheses under its corresponding variable. The R2, adjusted R2, and the total number of observations are also reported. The Appendix provides definitions for all the variables.

46Table 7 presents the results of Model 2, which examines the impact of each of the CSR sub-ratings on credit ratings. To validate H2, we expect the coefficient of the environmental component of CSR to be higher (and more significant) than that of the other components (i.e. human rights, business behaviour, corporate governance, community involvement and human resources). The coefficient of the environment component of CSR in Model 2 is the highest of all the components in their relationship with CR. The coefficient ofis significant at the 1% level and is positively related to credit ratings, and the adjusted R2 for this model is the highest (0.485), thus indicating that the environment can explain the default risk better than any other component. Therefore, H2 has been validated. All our control variables are significant at the 1% level and are in line with our expectations.

Table 7: Examining the Vigeo Eiris sub-ratings impact on credit ratings

47This table reports the results from regressing Model 2. Each component of CSR (e.g. HR, ENV…) is run in a separate regression from the other components, in order to avoid multicollinearity. The main explanatory variable is the CSR component. The table reports the Adjusted R2 and the coefficient of the main explanatory variable, where the t-value is reported between the parentheses under the corresponding variable. The total number of observations is 3946. The Appendix provides definitions for all the variables.

48Table 8 reports the results of Models 3, in which we classify countries into civil/common law countries. We argue that CSR would have a higher impact on default risk in civil law countries than in common law countries. The results indicate that the coefficient of CSR in civil law countries is higher than in common law countries, no matter whether our sample includes or excludes socialist countries. The relationship is positive between the interaction variables and the credit ratings, so a higher social score will lead to higher credit ratings. The 90% confidence intervals (CI) for CV*CSR and CM*CSR intersect when we include socialist countries. However, CI(CV*CSR) is superior to CI(CM*CSR) and the intervals do not intersect after excluding socialist countries from the sample. The overall results indicate that the impact of CSR on default risk is stronger in civil law countries than in common law countries. Therefore, H3 is validated.

Table 8: The moderating effect of the legal origin on the CSRfirm default risk relationship

49This table reports the results of Model 3. The interaction terms for Model 3 are CV*CSR and CM*CSR. The control variables are: SIZE, PROF, BTM, and LEV, in addition to time fixed effect. The table reports the coefficients of the explanatory variables, where the t-values are in parentheses under the corresponding variable. The R2, adjusted R2, total number of observations and the 90% confidence intervals are also reported. The Appendix provides definitions for all the variables.

50Model 4 examines the CSR–firm default risk relationship by considering four different interaction variables. This model is used to testH4, in which we build a classification based on those of La Porta et al. (1998), Spamann (2010) and Liang and Renneboog (2017). Table 9 reports the results, which indicate that all our variables are significant, at least at the 5% level. Firms operating in a German (GR) law origin appear to have the highest effect in the CSR–firm default risk relationship, with a coefficient of 0.686, while firms operating in French (FR) law countries have the lowest (0.140). Scandinavian (SCA) law countries have a higher coefficient of CSR than English (ENG) law countries (0.679 > 0.372). Both confidence intervals of GR*CSR and SCA*CSR are superior to ENG*CSR and FR*CSR, with FR*CSR being inferior to the confidence intervals of our main explanatory variables. Therefore, H4 is partially validated. All the control variables are in line with our expectations. Whether the sample excludes or includes the socialist countries does not affect the significance level of the coefficients and the results interpretation remains the same. The differences in results between the civil law countries show that those countries are not homogeneous with respect to our object of study. In particular, it appears that the impact of CSR on default risk in French law countries is weaker than in common law countries. One may wonder whether there is not a more effective method than the legal origins approach for classifying countries as stakeholder- or shareholder-oriented. We propose such an alternative approach in the following section.

Table 9: The impact of CSR on credit ratings in the legal origin sub-classifications

51This table reports the results from regressing the dependent variable CR over the interaction terms ENG*CSR, FR*CSR, GR*CSR and SCA*CSR, where ENG, FR, GR and SCA refer to English, French, German and Scandinavian law origin, respectively. The control variables included in the regression are: SIZE, PROF, BTM, and LEV, in addition to time fixed effect. The table reports the coefficients of the explanatory variables, where the t-values are in parentheses under the corresponding variable. The R2, adjusted R2, total number of observations and the 90% confidence intervals are also reported. The Appendix provides definitions for all the variables.

4.3. Shareholder-oriented vs stakeholder-oriented classification

52The intuition behind our assumptions is that firms and individuals in stakeholder-oriented countries do not have the same behaviour and beliefs as those in shareholder-oriented countries. We have addressed that distinction between countries by referring to legal origin. Any classification is a simplification of reality, as countries are more complex than any categorisation can reflect. The previous results show contrasting results among civil law countries, with French law countries standing out. To account for those results, we turn to another classification, that of Hall and Soskice (2001), between liberal and coordinated market economies. It could be argued that relying on the varieties of capitalism framework could be more general than the legal origins, as the concepts of coordinated market economies (CMEs) and liberal market economies (LMEs) presented by Hall and Soskice, could represent broader institutional contexts in which shareholder and stakeholder contributions can be analysed. CMEs are characterised by strong inter-firms networks and alliances, collaboration, non-market relationships and coordination. In contrast, LMEs are characterised by management-driven, top-down hierarchies, and strong competitive markets (Walker et al., 2019; Witt and Jackson, 2016). Hall and Soskice classify six countries as LMEs and ten countries as CMEs. The six LMEs are all common law countries, and the ten CMEs are all civil law countries. What is interesting is that of the ten CMEs, only two are French law countries. Hall and Soskice consider six countries, all French law (France, Italy, Spain, Portugal, Greece and Turkey), as being more ambiguous and unclassifiable. We are therefore left with a classification that does not include all countries (only sixteen countries are classified, as opposed to forty-five previously). That amounts to being more demanding about whether a country belongs to the stakeholder-oriented or shareholder-oriented category by working on a sub-sample of countries.

53Using that framework, Walker et al. (2019) find that firms in CMEs tend to have a greater CSR commitment and a higher financial performance (measured by Tobin’s q) than firms in LMEs. Given the characteristics of CMEs and LMEs, we expect a stronger impact of CSR on default risk in CMEs than in LMEs. The reason for that is that CME countries are more likely to be concerned about firms’ CSR behaviour, with firms being under stronger pressure to exhibit CSR in those countries, thanks to the built-in community relations. Firms operating in CMEs enjoy stronger stakeholder relationships that those in LMEs (Witt and Jackson, 2016); the employee relationship is also crucial for firms’ survival. Bauer et al. (2009) examine the link between the employee relationship and credit risk to find that firms with good employee relations have a lower cost of debt financing, a lower firm-specific risk and a higher credit rating. More generally, in market economies characterised by consensus-based decisions and strong reputations, firms have greater incentives to engage in CSR and are more likely to be financially rewarded when doing so (Walker et al., 2019). In contrast, in market economies characterised by unilateral decision-making, firms have less incentive to engage in CSR and are less likely to be financially rewarded when doing so. Assuming that the degree of SR engagement and related rewards differs from one type of economy to another, we hypothesise that:

H5: The negative impact of corporate social responsibility on firms’ default risk level is stronger in CMEs than in LMEs.

54To assess the validity of Hypothesis 5 (H5), we follow the same methodology used in Model 2, except we change the interaction terms to , respectively. We refer to this model as Model 5. Note that the sample size has decreased because not all countries are distributed according to this classification. Table 10 reports the results, which validate H5, thus indicating that the impact of CSR on default risk is stronger in CMEs than in LMEs. The 90% confidence interval (CI) of CME*CSR is distinct and much higher than that of LME*CSR.

55The previous results regarding the legal origins are reinforced by using this more demanding classification, which does not allow all countries to be classified. The legal framework and the coordination mode at the heart of the LMEs vs CMEs classification go hand in hand. As shown by Casper in Hall and Soskice (2001), the legal system and the form of business coordination are co-determined. Casper chooses the examples of Germany and the United States, with the German institutions being associated with a high level of non-market coordination, whereas the US institutions are compatible with market forms of coordination. Our results confirm the relevance of the legal origin and the sub-classification of civil law countries to distinguish between shareholder-oriented and stakeholder-oriented countries.

Table 10: The impact of CSR on credit ratings in CMEs and LMEs

56This table reports the results of Model 5. The interaction terms of Model 5 are CME*CSR and LME*CSR. The control variables are: SIZE, PROF, BTM, and LEV, in addition to time fixed effect. The table reports the coefficients of the explanatory variables, where the t-values are in parentheses under the corresponding variable. The R2, adjusted R2, total number of observations and the 90% confidence intervals are also reported. The Appendix provides definitions for all the variables.

4.4. Robustness tests

  • 11 We considered only one lag for the control variables. Nevertheless, we allowed in a second approach (...)

57The examination of the relationship between the CSR and default risk of firms raises various concerns. One of those is a potential endogeneity influencing our results. To address endogeneity, we assess the CSR–firm default risk relationship ­– with and without a moderator – using the Arellano–Bover/Blundell–Bond endogeneity correction, which has been proposed previously in the literature to address this issue (Gex et al., 2018). It could be argued that the use of CSR scores and credit ratings could be misleading with short lags, as both are likely to change infrequently. An additional concern is whether any reverse causality exists between the dependent and independent variables. To tackle those issues, we allow up to five lags in our independent variables.11 Table 11 presents the endogeneity correction results. The overall results indicate a significant positive relationship between CSR and credit ratings (at the 5% level). In addition, CSR appears to have a significant impact on default risk in civil law countries (at least at the 5% level), whereas no significant impact is observed in common law countries. That is in line with our expectations. After correcting for endogeneity, the previous results are confirmed.

Table 11: Endogeneity correction results

58This table reports the results of Models 1 and 3 after correcting for endogeneity. The table reports the coefficient for the independent variables, where the t-values are in parentheses under the corresponding variable. The Appendix provides definitions for all the variables.

  • 12 Due to missing data in calculating the Z-score, the sample size is reduced.

59In a second robustness test, we re-examine Models 1, 3 and 6 by replacing the initial default risk measure (i.e. the credit ratings) with another proxy. The proxy considered is the Altman (1968) Z-score, which is an accounting-based measure that uses financial ratios to predict the risk of default.12 In that seminal paper, Altman classifies the ratios into five categories that account for liquidity, profitability, leverage, solvency and activity. Those ratios are included in the measure with the coefficients from Altman, with a higher Z-score being preferable as it indicates a lower likelihood of bankruptcy. Table 12 reports the results obtained and indicates that CSR is negatively related to firm default risk. That is in line with our previous findings. We include the confidence intervals so that we can compare those concerning the interaction terms. We notice that the confidence interval of CV*CSR is superior to that of CM*CSR. That substantiates our previous arguments that the impact of CSR on default risk is higher in stakeholder-oriented countries.

Table 12: Examining the CSR-firm default risk relationship using an the Altman measure

60This table reports the results of Models 1 and 3. The interaction terms of Model 3 are CV*CSR and CM*CSR. The control variables are: SIZE, PROF, BTM, and LEV, in addition to time and fixed effects. The table reports the coefficients of the explanatory variables, where the t-values are in parentheses under the corresponding variable. The R2, adjusted R2, total number of observations and the 90% confidence intervals (concerning Model 3) are also reported. The Appendix provides definitions for all the variables.

  • 13 Due to the decreased sample size, we decided to consider those variables/models in the robustness s (...)

61We present a third robustness test, which relates to the control variables used in the regressions. We wish to check whether the major findings are preserved, that CSR is negatively related to default risk and that the magnitude of this relationship varies across different legal origins. To that end, we re-examine the hypotheses using Models 1 and 3 with different and additional control variables. The unchanged control variables are : SIZE, BTM and LEV. Following Benlemlih and Girerd-Potin (2017), we control for profitability (N_PROF) using the ratio of earnings before interest and taxes to total capital. We include additional control variables: LIQ is the current ratio defined as the total current assets to total current liabilities; following Molina (2005),we control for the firm’s relative tangible assets position (RTA), which “serves to guarantee debt” (Molina, 2005, p. 1437). We control for tangibility using the ratio of net property plant and equipment to total assets. The choice of the control variables decreases the number of observations to 1,446.13 Table 13 presents the results, which reveal that changing the control variables in our model and adding variables provides no change from the previous findings. Regarding H1, CSR appears to have a significant effect on credit ratings. In accordance with H3, that effect is higher and more significant for firms operating in civil law countries. CSR appears to have no significant effect on default risk in common law countries when socialist countries are excluded.

Table 13: Examining the CSR−firm default risk relationship using different control variables

62This table reports the results of Models 1 and 3. The table reports the coefficients of the explanatory variables, where the t-values are in parentheses under the corresponding variable. The R2, adjusted R2, and the total number of observations are also reported. The Appendix provides definitions for all the variables.

  • 14 Available from the authors upon request.

63Additional robustness tests have been conducted.14 For instance, we regress our dependent variable first on the control variables, then we regress the obtained residuals on our main explanatory variables. The results are similar to our previous findings.

5. Conclusion

64This paper contributes to the debate concerning the corporate response to CSR and whether legal rules matter. We explore the impact of CSR on firm default risk across a large international sample offering varying institutional contexts. We first establish the existence of a negative link between CSR and a firm’s default risk level. Using credit ratings as a proxy for default risk, and the ratings provided by Vigeo Eiris as a measure of CSR, we find a positive link between credit ratings and CSR. We then investigate the role of the legal context by testing the moderating role of the legal origins on the CSR–firm default risk relationship. We first classify countries according to their civil law or common law origin. The results show that legal origins have a significant impact on firms’ social behaviours and on individuals’ social beliefs, thus generating a higher impact of CSR commitment on credit ratings in civil law countries than in common law countries. The difference in coefficients between the two categories of countries is even more pronounced if the socialist countries are excluded from the sample. Additional analyses concerning the CSR components reveal that the environment component has the highest impact on default risk. The robustness of the results is attested to first by taking endogeneity and reverse causality into account, then by using the Altman score as a measure of default risk instead of credit ratings and finally by modifying the control variables.

65Second, we break our classification down into English, French, German and Scandinavian law origin. We find that the impact of CSR on default risk is highest in countries with a German law origin. It is lowest for French law countries. That leads us to look at a more demanding classification, that of Hall and Sostkice (2001), which classifies a small number of countries as LMEs or CMEs. The LMEs are all common law countries and the CMEs are all civil law countries, with very few French law countries. The impact of CSR on default risk is much stronger in CMEs than in LMEs, with a larger gap than is presented when applying the civil law / common law distinction. Those results indicate the crucial role of the legal system under which a firm operates. They show that the legal origin creates distinctions between shareholder-oriented and stakeholder-oriented countries and substantiate the argument for the use of sub-origins within civil law countries.

66While empirical evidence is still scant regarding the impact of CSR on default risk, our findings contribute to a better understanding of the relationship. Our research has several potential implications for corporations and risk managers in terms of default risk and the enhancing of corporate financial performance. More precisely, it would be beneficial for firms operating in civil law countries to increase their concern for all stakeholders because SR commitment has a significant impact on reducing the risk of default. For investors and portfolio managers, our results reveal the need to consider the social ratings of firms, as well as the legal regimes under which they operate, when constructing their portfolios. Our findings also indicate that, on average, firms in common law countries – more generally, in shareholder-oriented countries – tend to invest less in CSR, probably because the benefit–cost ratio of SR commitment is lower than in civil law countries – more generally, in stakeholder-oriented countries. The legal context shapes the behaviours of all the stakeholders (managers, employees, customers, suppliers, etc.) more prone to favouring SR firms in civil law countries and thus helps those SR firms to survive.

67Building on the importance of the links between default risk, CSR and legal regimes for firms, investors and portfolio managers, this research raises some new questions. For instance, future research could study how government deregulation and legislative changes lead to an evolution in firms’ SR commitment and its impact. It would also be worth examining the case of multinational enterprises (MNEs) where the headquarters is in a country with a different legal origin than its subsidiaries. In particular, do subsidiaries adopt the social policy of the parent firm? Or do they adopt the social policy of the environment within which they operate? An additional concern that could be examined in further research is related to the rating agencies themselves. Financial analyses of firms now tend to integrate social and environmental criteria previously considered extra-financial. Those criteria are useful for assessing financial risks. The mergers and acquisitions occurring between credit rating agencies and social rating agencies (e.g. the acquisition of Vigeo Eiris by Moody’s in 2019) also demonstrate that convergence. The end of the independence of attribution of the two types of ratings generates questions about their future links. If the sole purpose of firms’ SR commitments is to improve financial risk and return, the value of the ESG rating as such is likely to diminish. On the other hand, if firms’ SR commitments are also a form of altruism desired by shareholders, the ESG ratings will retain their value. Future research on those aspects would be of great interest.

Haut de page

Bibliographie

Aghion, P., Algan, Y., Cahuc, P., & Shleifer, A. (2010). Regulation and distrust. Quarterly Journal of Economics, 125(3), 1015–1049.

Aguilera, R. V., & Jackson, G. (2003). The cross-national diversity of corporate governance: Dimensions and determinants. Academy of Management Review, 28(3), 447–465.

Albuquerque, R., Koskinen, Y., & Zhang, C. (2019). Corporate social responsibility and firm risk: Theory and empirical evidence. Management Science, 65(10), 4451–4469.

Altman, E. I. (1968). Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy. The Journal of Finance, 23(4), 589–609.

Altman, E. I., & Hotchkiss, E. (2006). Corporate Financial Distress and Banckruptcy (third edit). New York: Wiley Finance.

Amaeshi, K., Adegbite, E., & Rajwani, T. (2016). Corporate Social Responsibility in Challenging and Non-enabling Institutional Contexts: Do Institutional Voids matter? Journal of Business Ethics, 134(1), 135–153.

Amor-Esteban, V., García-Sánchez, I. M., & Galindo-Villardón, M. P. (2018). Analysing the Effect of Legal System on Corporate Social Responsibility (CSR) at the Country Level, from a Multivariate Perspective. Social Indicators Research, 140(1), 435–452.

Arouri, M., & Pijourlet, G. (2017). CSR Performance and the Value of Cash Holdings: International Evidence. Journal of Business Ethics, 140(2), 263–284.

Attig, N., El Ghoul, S., Guedhami, O., & Suh, J. (2013). Corporate Social Responsibility and Credit Ratings. Journal of Business Ethics, 117(4), 679–694.

Awazu, P. B.-M. D.-L., Da Silva, P., & Svartzman, F. S.-R. (2020). The green swan.

Badayi, S. A., Matemilola, B. T., Bany-Ariffin, A. N., & Wei Theng, L. (2020). Does corporate social responsibility influence firm probability of default ? International Journal of Finance & Economics, 1–19.

Bauer, R., Derwall, J., & Hann, D. (2009). Employee Relations and Credit Risk. European Centre for Corporate Engagement (ECCE).

Baumgartner, R. J. (2014). Managing corporate sustainability and CSR: A conceptual framework combining values, strategies and instruments contributing to sustainable development. Corporate Social Responsibility and Environmental Management, 21(5), 258–271.

Benlemlih, M., & Girerd-Potin, I. (2017). Corporate social responsibility and firm financial risk reduction: On the moderating role of the legal environment. Journal of Business Finance & Accounting, 1–30.

Bharath, S. T., & Shumway, T. (2008). Forecasting default with the Merton distance to default model. Review of Financial Studies, 21(3), 1339–1369.

Boubaker, S., Cellier, A., Manita, R., & Saeed, A. (2020). Does corporate social responsibility reduce financial distress risk ? Economic Modelling, 91, 835–851. Retrieved from https://doi.org/10.1016/j.econmod.2020.05.012

Boutin-Dufresne, F., & Savaria, P. (2004). Corporate Social Responsibility and Financial Risk. The Journal of Portfolio Management, 13(1), 57–66.

Brooks, C., & Oikonomou, I. (2018). The effects of environmental, social and governance disclosures and performance on firm value: A review of the literature in accounting and finance. British Accounting Review, 50(1), 1–15.

Cai, L., Cui, J., & Jo, H. (2016). Corporate Environmental Responsibility and Firm Risk. Journal of Business Ethics, 139(3), 563–594.

Campbell, J. L. (2006). Institutional Analysis and the Paradox of Corporate Social Responsibility. American Behavioral Scientist, 49(7), 925–938.

Chava, S. (2014). Environmental externalities and cost of capital. Management Science, 60(9), 2223–2247.

Cheng, B., Ioannou, I., & Serafeim, G. (2014). Corporate Social Responsibility and Access to Finance. Strategic Management Journal, 35, 1–23.

Colbert, B., Wheeler, D., & Freeman, E. R. (2003). Focusing on Value: Reconciling Corporate Social Responsibility, Sustainability and a Stakeholder Approach in a Network World. Journal of General Management, 28(3), 1–28.

Edmans, A. (2011). Does the stock market fully value intangibles? Employee satisfaction and equity prices. Journal of Financial Economics, 101(3), 621–640.

Edmans, A. (2012). The link between job satisfaction and firm value, with implications for corporate social responsibility. Academy of Management Perspectives, 26(4), 1–19.

El Ghoul, S., Guedhami, O., & Kim, Y. (2016). Country-level institutions, firm value, and the role of corporate social responsibility initiatives. Journal of International Business Studies, 48(3), 1–26.

El Ghoul, S., Guedhami, O., Kwok, C. C. . Y., & Mishra, D. R. (2011). Does corporate social responsibility affect the cost of capital? Journal of Banking and Finance, 35(9), 2388–2406.

Flammer, C. (2013). Corporate Social Responsibility and Shareholder Value: The Environmental Consciousness of Investors. Academy of Management Journal, 56(3), 758–781.

Flammer, C. (2015). Does Corporate Social Responsibility Lead to Superior Financial Performance ? A Regression Discontinuity Approach. Management Science, 61(11), 2549–2568.

Freeman, E. R. (1984). Strategic management: A stakeholder approach. Boston: Pitman/Ballinger.

Gex, M., Girerd-Potin, I., Jimenez-Garcès, S., & Louvet, P. (2018). Does Sustainability Reduce Country Default Risk. Bankers, Markets & Investors, (150), 34–47.

Girerd-Potin, I., Jimenez-Garcès, S., & Louvet, P. (2014). Which Dimensions of Social Responsibility Concern Financial Investors? Journal of Business Ethics, 121(4), 559–576.

Goss, A. (2009). Corporate social responsibility and financial distress. ASAC, 30(29).

Goss, A., & Roberts, G. S. (2011). The impact of corporate social responsibility on the cost of bank loans. Journal of Banking and Finance, 35(7), 1794–1810.

Gregory-Smith, D., Manika, D., & Demirel, P. (2017). Green intentions under the blue flag: Exploring differences in EU consumers’ willingness to pay more for environmentally-friendly products. Business Ethics: A European Review, 26(3), 205–222.

Hall, P. A., & Soskice, D. (2001). An introduction to varieties of capitalism. Oxford University Press.

Henisz, W. J., Dorobantu, S., & Nartey, L. J. (2014). Spinning gold: The financial returns to stakeholder engagement. Strategic Management Journal, 35(12), 1727–1748.

Hillegeist, S. A., Keating, E. K., Cram, D. P., & Lundstedt, K. G. (2004). Assessing the probability of bankruptcy. Review of Accounting Studies, 9(1), 5–34.

Ioannou, I., & Serafeim, G. (2012). What drives corporate social performance the role of nation-level institutions. Journal of International Business Studies, 43(9), 834–864.

Jackson, G., & Apostolakou, A. (2010). Corporate Social Responsibility Mirror Europe : An Institutional Mirror or Substitute ? Journal of Business Ethics, 94(3), 371–394.

Jackson, G., & Rathert, N. (2015). February 2015, 1–26.

Jiraporn, P., Jiraporn, N., Boeprasert, A., & Chang, K. (2014). Does corporate social responsibility (CSR) improve credit ratings? Evidence from geographic identification. Financial Management, 43(3), 505–531.

Jo, H., & Na, H. (2012). Does CSR Reduce Firm Risk? Evidence from Controversial Industry Sectors. Journal of Business Ethics, 110(4), 441–456.

Kassinis, G., & Vafeas, N. (2006). Stakeholder Pressures and Environmental Performance. Academy of Management Journal, 49(1), 145–159.

Kim, C. H., Amaeshi, K., Harris, S., & Suh, C.-J. (2013). CSR and the national institutional context: The case of South Korea. Journal of Business Research, 66(12), 2581–2591.

Klassen, R. D., & McLaughlin, C. P. (1996). The Impact of Environmental Management on Firm Performance. Management Science, 42(8), 1199–1214.

Kock, C. J., & Min, B. S. (2016). Legal Origins, Corporate Governance, and Environmental Outcomes. Journal of Business Ethics, 138(3), 507–524.

La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1998). Law and Finance. Journal of Political Economy, 106(6), 11131–55.

Liang, H., & Renneboog, L. (2017). On the Foundations of Corporate Social Responsibility. The Journal of Finance, 72(2), 853–910.

Lichtenstein, D. R., Drumwright, M. E., & Braig, B. M. (2004). The Effect of Corporate Social Responsibility on Customer Donations to Corporate-Supported Nonprofits. Journal of Marketing, 68(4), 16–32. Retrieved from

Luo, X. L., & Bhattacharya, C. B. (2006). Corporate social responsibility, customer Satisfaction, and market value. Journal of Marketing, 70(4), 1–18.

Margolis, J. D., Hillary, E. A., & Walsh, J. R. (2009). Does it pay to be good? A meta-analysis and redirection of research on the relationship between corporate social and financial performance. MIT Sloan Management Review, 50, 61–68.

Martínez, J. B., Fernández, M. L., & Fernández, P. M. R. (2016). Corporate social responsibility: Evolution through institutional and stakeholder perspectives. European Journal of Management and Business Economics, 25(1), 8–14.

Matten, D., & Moon, J. (2008). “Implicit” and “Explicit” CSR: A Conceptual Framework for a Comparative Understanding of Corporate Social Responsibility. The Academy of Management Review, 33(2), 404–424.

McWilliams, A., & Siegel, D. (2001). Corporate social responsibility : a theory of the firm perspective. Academy of Mamagement Review, 26(1), 117–127.

Merton, R. (1974). On the Pricing of Corporate Debt: The Risk Structure of Interest Rates. The Journal of Finance, 29(2), 449–470.

Metzler, M. S. (2001). The certanity of organizational legitimicay to public relations practice. In Handbook of public relations (pp. 321–334).

Mishra, S., & Modi, S. B. (2013). Positive and Negative Corporate Social Responsibility, Financial Leverage, and Idiosyncratic Risk. Journal of Business Ethics, 117(2), 431–448.

Molina, C. A. (2005). Are Firms Underleveraged? An Examination of the Effect of Leverage on Default Probabilities. The Journal of Finance, 60(3), 1427–1459.

Morgan, G., & Kristensen, P. H. (2006). The contested space of multinationals: Varieties of institutionalism, varieties of capitalism. Human Relations, 59(11), 1467–1490.

Moussu, C., & Ohana, S. (2016). Do Leveraged Firms Underinvest in Corporate Social Responsibility? Evidence from Health and Safety Programs in U.S. Firms. Journal of Business Ethics, 135(4), 715–729.

Orlitzky, M., Schmidt, F. L., & Rynes, S. L. (2003). Corporate Social and Financial Performance: A Meta-analysis. Organization Studies, 24(3), 403–441.

Ötker-Robe, İ., & Podpiera, J. (2010). The Fundamental Determinants of Credit Default Risk for European Large Complex Financial Institutions. IMF Working Papers, 153, 1–31.

Quéré, B. P., Nouyrigat, G., & Baker, C. R. (2018). A bi-directional examination of the relationship between corporate social responsibility ratings and company financial performance in the European context. Journal of Business Ethics, 148(3), 527–544.

Rizwan, M. S., Obaid, A., & Ashraf, D. (2017). The Impact of Corporate Social Responsibility on Default Risk: Empirical evidence from US Firms. Business & Economic Review, 9(3), 36–70.

Scott, W. R. (1995). Institutions and Organizations. Sage: Thousand Oaks.

Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality. Biometrika, 52(3–4), 591–611.

Sharfman, M. P., & Fernando, C. S. (2008). Environmental Risk Management and the Cost of Capital. Strategic Management Journal, 29(6), 569–592.

Spamann, H. (2010). The “antidirector rights index” revisited. Review of Financial Studies, 23(2), 467–486.

Stellner, C., Klein, C., & Zwergel, B. (2015). Corporate social responsibility and Eurozone corporate bonds: The moderating role of country sustainability. Journal of Banking and Finance, 59, 538–549.

Sun, W., & Cui, K. (2014). Linking corporate social responsibility to firm default risk. European Management Journal, 32(2), 275–287.

Vassalou, M., & Xing, Y. (2004). Default Risk in Equity Returns. The Journal of Finance, 59(2004), 831–868.

Walker, K., Zhang, Z., & Ni, N. (Nina). (2019). The Mirror Effect: Corporate Social Responsibility, Corporate Social Irresponsibility and Firm Performance in Coordinated Market Economies and Liberal Market Economies. British Journal of Management, 30(1), 151–168.

Wallis, M. von, & Klein, C. (2015). Ethical requirement and financial interest: a literature review on socially responsible investing. Business Research, 8(1), 61–98.

Wang, Y., & Berens, G. (2015). The Impact of Four Types of Corporate Social Performance on Reputation and Financial Performance. Journal of Business Ethics, 131(2), 337–359.

Welford, R., Chan, C., & Man, M. (2008). Priorities for corporate social responsibility: A survey of businesses and their stakeholders. Corporate Social Responsibility and Environmental Management, 15(1), 52–62.

Williams, C. A., & Aguilera, R. V. (2008). Corporate Social Responsibility in a Comparative Perspective. Oxford Handbook of Corporate Social Responsibility, 452–472.

Windolph, S. E., Harms, D., & Schaltegger, S. (2014). Motivations for corporate sustainability management: Contrasting survey results and implementation. Corporate Social Responsibility and Environmental Management, 21(5), 272–285.

Witt, M. A., & Jackson, G. (2016). Varieties of Capitalism and institutional comparative advantage : A test and reinterpretation. Journal of International Business Studies, 47(7), 778–806.

Yan, S., Ferraro, F., & Almandoz, J. (John). (2019). The Rise of Socially Responsible Investment Funds: The Paradoxical Role of the Financial Logic. Administrative Science Quarterly, 64(2), 466–501.

Yin, J. (2017). Institutional Drivers for Corporate Social Responsibility in an Emerging Economy: A Mixed-Method Study of Chinese Business Executives. Business & Society, 56(5), 672–704.

Haut de page

Annexe

Appendix: Variables’ Definitions

Variables

Definition

Panel A. The dependent variable

CR

The default risk proxy measured by the average credit ratings provided by Moody’s Analytics, Standard & Poor’s, and Fitch Ratings.

Z-score

The default risk proxy calculated using the Altman (1968) model.

Panel B. Corporate social responsibility

CSR

The corporate social responsibility score based on the six rating dimensions provided by Vigeo Eiris. Following Girerd-Potin et al. (2014), we calculate the score in the following way:

CSR = (6√((1+nHR)(1+nENV )(1+nBB)(1+nCG )(1+nCI )(1+nHRT))-1),

Where nj for human resources, environment, business behaviour, corporate governance, community involvement and human rights.

HR

Human resources. Represents the constant improvement of industrial relations, career development, as well as the quality of working conditions.

ENV

Environment. Represents the protection, safeguard, prevention of attacks on the environment, implementation of an adequate managerial strategy, ecodesign, protection of biodiversity and reasonable control of environmental impacts on the overall life cycle of products and services.

BB

Represents the business behaviour while taking into account clients’ rights and interests, integration of social and environmental standards, both in the process of selection of suppliers and in the overall supplying chain, efficient prevention of corruption and respect of competition laws.

CG

Corporate governance. Represents the efficiency and integrity, insurance of both independence and effectiveness of the Board of Directors, effectiveness and efficiency of audit and control systems, and in particular inclusion of social responsibility risks, respect of shareholders’ rights and most of all of the minorities, transparency and moderation in executive remuneration.

CI

Community involvement. Represents the contribution to economic and social development of the territories of establishment and their human communities, concrete commitment in favour of the control of societal impact of products and services, transparency and participative contribution to causes of general interest.

HRT

Human rights. Measures the respect of trade unions’ freedom and promotion of collective negotiation, non-discrimination and promotion of equality, eradication of banned working practices, and prevention of inhumane or humiliating treatments.

Panel C. The control variables

SIZE

The logarithm of total assets, used to control for the firm size.

PROF

The return on assets, used to control for profitability.

BTM

The book-to-market ratio, used to measure the future growth of a firm.

LEV

Leverage; measured as the ratio between long-term debt and total assets.

N_PROF

Controls for profitability; measured by the ratio between earnings before interest and taxes (EBIT) and total capital.

RTA

The ratio of net property plant and equipment to total assets, used to control for assets’ tangibility.

LIQ

The current ratio defined as the ratio of total current assets to total current liabilities, used to control for liquidity.

Panel D. Legal origins

Civil law (CV)

A dummy variable that takes the value of 1 if the firm is operating in a civil law origin, and 0 otherwise. The classification is based on La Porta et al. (1998), Spamann (2010) and Liang and Renneboog (2017).

Common law (CM)

A dummy variable that takes the value of 1 if the firm is operating in a common law origin, and 0 otherwise. The classification is based on La Porta et al. (1998), Spamann (2010) and Liang and Renneboog (2017).

ENG, FR, GR, SCA

Four dummy variables that take the value of 1 if the firm is operating in the examined legal origin, and 0 otherwise. Where, ENG, FR, GR and SCA, stand for English, French, German and Scandinavian legal origin, respectively.

Panel E. Additional Classifications

CME

A dummy variable that takes the value of 1 if the firm is operating in a coordinated market economy, and 0 otherwise. The classification is based on Hall and Soskice (2001).

LME

A dummy variable that takes the value of 1 if the firm is operating in a liberal market economy, and 0 otherwise. The classification is based on Hall and Soskice (2001).

Haut de page

Notes

1 For an extensive literature review on the latter, please refer to Margolis et al. (2009).

2 By definition, stakeholders are any group or individual who can affect or is affected by the achievement of the firm’s objectives (Freeman, 1984, p. 25).

3 According to Metzler (2001, p. 321), “establishing and maintaining organisational legitimacy is at the core of most, if not all, public relation activities”.

4 For further explanations, please refer to Henisz et al. (2014).

5 Primary stakeholders-related.

6 Secondary stakeholders-related.

7 They refer to it as the “green swan” risk, which 0063ould be behind the next systemic financial crisis. For further information, see Awazu et al. (2020).

8 The methodology used for constructing the global rating is that employed by Girerd-Potin et al. (2014): , where represents the sub-rating concerning dimensions for human resources, environment, business behaviour, corporate governance, community involvement and human rights.

9 That is, for models that do not differentiate between countries in terms of legal origins or varieties of capitalism, we add a country fixed effect.

10 Since we consider the credit ratings as a default risk proxy, we expect to obtain a higher credit rating for firms that behave socially responsibly, thus we expect a positive relationship between credit ratings and CSR.

11 We considered only one lag for the control variables. Nevertheless, we allowed in a second approach (not reported) up to five lags for all the independent variables. The results obtained are in line with our previous findings.

12 Due to missing data in calculating the Z-score, the sample size is reduced.

13 Due to the decreased sample size, we decided to consider those variables/models in the robustness section.

14 Available from the authors upon request.

Haut de page

Table des illustrations

URL http://journals.openedition.org/fcs/docannexe/image/8784/img-1.png
Fichier image/png, 61k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-2.png
Fichier image/png, 34k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-3.png
Fichier image/png, 18k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-4.png
Fichier image/png, 18k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-5.png
Fichier image/png, 41k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-6.png
Fichier image/png, 22k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-7.png
Fichier image/png, 13k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-8.jpg
Fichier image/jpeg, 34k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-9.png
Fichier image/png, 27k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-10.png
Fichier image/png, 25k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-11.png
Fichier image/png, 52k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-12.png
Fichier image/png, 62k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-13.png
Fichier image/png, 38k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-14.png
Fichier image/png, 35k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-15.png
Fichier image/png, 44k
URL http://journals.openedition.org/fcs/docannexe/image/8784/img-16.png
Fichier image/png, 54k
Haut de page

Pour citer cet article

Référence électronique

Mohamad Hassan Shahrour, Isabelle Girerd-Potin et Ollivier Taramasco, « Corporate Social Responsibility and Firm Default Risk Mitigation: The Moderating Role of the Legal Context »Finance Contrôle Stratégie [En ligne], 25-1 | 2022, mis en ligne le 31 mars 2022, consulté le 20 mai 2022. URL : http://journals.openedition.org/fcs/8784 ; DOI : https://doi.org/10.4000/fcs.8784

Haut de page

Auteurs

Mohamad Hassan Shahrour

Université Côte d'Azur, IAE Nice, GRM, Nice, France
Corresponding Author
E-mail: Mohamad-Hassan.Shahrour@univ-cotedazur.fr
Telephone: +33 (04) 89 15 21 41

Articles du même auteur

Isabelle Girerd-Potin

Univ. Grenoble Alpes, Grenoble INP, CERAG, Grenoble, France
E-mail: Isabelle.Girerd-Potin@univ-grenoble-alpes.fr

Articles du même auteur

Ollivier Taramasco

Univ. Grenoble Alpes, Grenoble INP, CERAG, Grenoble, France
E-mail: Ollivier.Taramasco@univ-grenoble-alpes.fr

Articles du même auteur

Haut de page

Droits d’auteur

Tous droits réservés

Haut de page
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search