Excerpt
Table of Contents
Acknowledgements
Abstract
List of Figures
List of Tables
Table of Abbreviations
1 Introduction
2 Theoretical Foundation and Hypothesis Development
2.1 Corporate Social Responsibility
2.2 The Role of ESG in Corporate Finance
2.2.1 ESG and Firm Performance
2.2.2 ESG and Cost of Capital
2.3 Credit Risk in the ESG - Cost of Debt Relation
3 Data and Methodology
3.1 Data for the ESG-CoD Relationship
3.1.1 Firm-Level Argument
3.1.2 Loan-Level Argument
3.2 Variable Construction for the ESG-CoD Relationship
3.2.1 Firm-Level Argument
3.2.2 Loan-Level Argument
3.3 Methodology for the ESG-CoD Relationship
3.3.1 Firm-Level Empirical Model
3.3.2 Loan-Level Empirical Model
3.4 Empirical Model for the Credit Risk Impact
4 Results
4.1 Analysis on the ESG-CoD Relationship
4.2 Analysis on the Credit Risk Channel
4.3 Robustness Checks
4.3.1 Alternative Model Specifications
4.3.2 Addressing Endogeneity Concerns
5 Conclusion
References
Appendix
Acknowledgements
With completing this master’s thesis, the unforgettable experience of studying at Rotterdam School of Management, Erasmus University (RSM) comes to an end - a challenging but tremendously rewarding experience at the same time. More than thankful, I look back on the nourishing academic environment I found at RSM that helped me sharpen my professional skillset and inspired me to become a force for positive change myself. Noteworthy, having the privilege of taking part in the 2020 Honours class broadened my horizon towards the real-world challenges I, as a finance graduate, can take on solving. Ultimately, the stimuli of thinking in an integrated fashion rather than within the narrow boundaries of traditional financial theory not only made me find my thesis topic in the dynamic area of sustainable finance, but also my career start in international project finance with one of the leading global financiers for renewable energy projects. More important even, I am grateful for all the inspiring people I met during my time at RSM. Most notably, I would like to thank Professor Mancy Luo and Professor Marno Verbeek for their insightful suggestions and helpful guidance for writing this thesis. Lastly, I cannot thank enough my friends, family, and especially my girlfriend Silvana for all their support, patience, and encouragement throughout my academic journey from Germany over China and the UK to the Netherlands and back. Thank you all so much!
Florian Porzel Rotterdam, June 2020
Abstract
This thesis examines the effect of Corporate Social Responsibility (CSR) expressed through Environmental, Social, and Governance (ESG) scores on firms’ cost of debt on two distinct layers with a particular interest on the economic mechanism through which sustainability performance unfolds. While I establish three distinct economic channels for the effect of corporate sustainability on the cost of debt capital, namely governance strengths, information asymmetry, and credit risk, I provide evidence that the latter is primarily responsible for lower debt premia to sustainable borrowers. First, on firm-level, I show that superior ESG performance can offset cost of debt by 0.45% for a one standard deviation improvement on ESG performance. Using propensity-matched loan facilities granted to European firms, I strengthen findings on firm-level by uncovering the expression of ESG performance in spread differences between conventional and sustainability-linked loan facilities. Leveraging the information efficiency of private debt markets, the additional loan-level analysis minimizes the likelihood of the alternative information asymmetry explanation and reveals that high ESG performing firms dominantly raise sustainability-linked debt. This translates into significant initial spread discounts of up to 69.42 basis points for ESG-linked loans as compared to regular loans. Lastly, the analyses expose an altered credit risk profile as the driving force behind the negative relation whereby ESG concerns are found to play a dominant role in the manifestation of credit risk and that lenders increasingly account for sustainability at low levels of traditional credit risk measures. These findings remain robust against alternative model specifications and endogeneity concerns, confirming the dominant influence of ESG performance beyond the effects of alternative economic mechanisms and that it is not a proxy for other firm attributes. Overall, obtained results provide academic evidence on the value of sustainability for reducing credit risk and hence for improving firm financing costs. As such, the results are in support of the theoretical foundations, primarily backing the risk-mitigation view. Moreover, the provided evidence holds practical implications for raising funds through sustainability-linked loan instruments and for including measures of CSR in the traditional assessments of credit risk.
Keywords: Corporate Finance, Sustainable Finance, Corporate Social Responsibility, ESG, Cost of Capital, Cost of Debt, Credit Risk, Sustainability-Linked Loans
List of Figures
Figure 1: Transition of Management Practices based on CSR
Figure 2: Conceptual Framework
List of Tables
Table 1: Descriptive Statistics
Table 2: Firm-Level Regression Results
Table 3: Loan-Level Regression Results
Table 4: Credit Risk Regression Results
Table 5: Marginal Effects Analysis
Table 6: Robustness Test - Alternative ESG Specification
Table 7: Robustness Test - Alternative Credit Risk Specification
Table 8: Robustness Test - Cross Sectional Variation
Table 9: Robustness Test - Endogeneity
Table of Abbreviations
Abbildung in dieser Leseprobe nicht enthalten
1 Introduction
In current times of global climate stress with environmental anomalies happening at a daily rate, business actions are understood to play the pivotal role in fighting the most pressing concern in contemporary human history - the transition towards a sustainable economic model. In fact, in order to reach the ambitious sustainability target set by the European Union to reach carbon neutrality by 2050, additional sustainable investments of EUR175 to EUR290 billion are in demand annually. As such, the scaling up of private sector investments entered as key element into the EU Sustainable Finance Action Plan launched by the European Commission in 2018. And although the first climate change bankruptcy with PG&E Corp. failing to meet potential liabilities of around USD30 billion resulting from wildfires already occurred, the need for sustainable adaptation of business practices is erroneously evaluated against hypothetical scenarios or a debatable model of the long-term effects of change (Mui, 2019).
Still, market constraints and the prevalence of short-termism hinder tangible action-taking in the near-term. More alarming however, since weak economic incentives and uncertain relations between sustainable performance and business value persist, governmental and business efforts in the transition towards a new economic model remain misaligned (Schoenmaker & Schramade, 2019a). By establishing this long-missing practical incentive for connecting sustainable performance and firm value, sustainable lending has become a recent phenomenon of interest in the corporate world - a phenomenon implying that corporate social responsibility has in fact the potential to offer tangible economic value by improving creditworthiness, decreasing the risk of default, and thus reducing borrowing costs for sustainable firms. Hence, the apparent contradiction between hesitant corporate action-taking and lenders’ willingness to economically incentivize sustainable performance gives rise to the following research question:
R: What is the impact of corporate ESG Performance on Cost of Debt through Credit Risk?
Motivation and Academic Background
In academic literature, the link between corporate social responsibility (CSR) and firm performance has sparked interest for decades resulting in an empirically well-backed consensus view. To illustrate, by performing a review analysis of more than 2,200 unique studies, Friede, Busch, and Bassen (2015) discovered that c. 90% of studies find at least a nonnegative relation between firms’ achievements on environmental, social and governance aspects (ESG) and their performance. Although frequently used indistinguishably in research contexts, going forward CSR is viewed as the conceptual narrative whereas ESG as measurable indicators of its facets.
In addition to the favourable aspects of CSR on firm performance, researchers largely observed a negative relationship between superior CSR and cost of capital (see e.g. Sharfman & Fernando, 2008; El Ghoul, Guedhami, Kwok, & Mishra, 2011; Albuquerque, Koskinen, & Zhang, 2018) under the premise that CSR would affect systematic risk and thus drive down cost of equity. Since CSR efforts and their resulting benefits can vary greatly among different industries and individual firms (see Khan, Serafeim, & Yoon, 2015), sustainable performance was also found to negatively relate to idiosyncratic risk (Mishra & Modi, 2013) suggesting the same effect of CSR on cost of debt as on cost of equity. Surprisingly however, among the significantly fewer studies engaging with the impact of CSR on cost of debt, several articles suggesting counterintuitive results are found (see Sharfman & Fernando, 2008; Goss & Roberts, 2011 or Magnanelli & Izzo, 2017). Given the non-conclusive findings for the CSR- cost of debt relationship, the question how CSR is impacting borrowing costs persists.
Moreover, to which extent credit risk plays a role in this relationship remains poorly understood (see Jiraporn, Jiraporn, Boeprasert, & Chang, 2014; Liu & Ge, 2015). If elevated credit risk is indeed the driving force behind the expected higher cost of debt capital for poor CSR, sustainability performance should affect the probability of default as the main determinant of cost of debt (see Merton, 1974). Thus, I expect that the performance along relevant sustainability criteria interacts with the probability of default proxied for by firms’ long-term credit ratings. The assumed interplay between CSR and credit risk would then expectedly manifest in significantly lowered financing costs overall, i.e. on all contractual liabilities.
With growing public and academic awareness for CSR since the financial crisis, additional financial innovations have emerged that enable businesses to capture tangible benefits for integrating sustainability issues into financing decisions. Given the importance for European firms to employ bank loans as funding source (IMF, 2019), Europe has pioneered the structuring of sustainable debt with the arrangement of the first sustainability-linked loan facility by ING Bank N.V. to Koninklijke Philips N.V. in 2017. While this implies an incentive for all firms to improve sustainable performance during loan life, especially CSR frontrunners, i.e. already well-performing firms, are expected to dominantly raise sustainability-linked debt. Hence, if sustainable efforts do in fact negatively relate to the probability of default on debt, the decreased risk to lenders should equally manifest in significant differences in inaugural spreads on conventional (as investigated by Goss & Roberts, 2011) and sustainability-linked bilateral debt. Thus, in conclusion, the prevailing uncertainty about the CSR-cost of debt relation and the lack of understanding of its underlying economic mechanisms lead towards the research question as stated in the introduction and the testable hypotheses discussed in the following section.
The methodology used to test aforementioned hypotheses builds on the work of Liu and Ge (2015), who discovered a negative relation between CSR performance and bond yield spreads via improved credit ratings. First, testing whether the sample in this study shows an impact of ESG performance on cost of debt, I perform an empirical analysis on two distinct layers to provide both firm- and loan-level evidence. While other studies primarily use CSR data from KLD STATS, this study employs ESG scores from Thomson Reuters DATASTREAM/ASSET4. More uniquely, this thesis covers the post-financial-crisis period from 2013 to 2020 and focusses on European firms. The entirety of ESG-rated firms consecutively rated over 3 years within this period and incorporated in Europe thus forms the data set consisting of 1,052 unique firms. On this basis, the firm-level empirical analysis is carried out by estimating a simple panel regression of cost of debt on ESG Combined scores and accounting-related control variables revealing a weakly significantly negative effect of ESG performance on cost of debt by 0.45% for a one standard deviation increase in firms’ ESG Combined scores. Strengthening the argumentation on firm-level and mitigating the likelihood of the alternative information asymmetry explanation, the expression of ESG performance in loan spread differences between regular loans and sustainability-linked facilities is tested. For this purpose, data on loans granted to the 1,052 unique firms is obtained, after which conventional and sustainability-linked facilities are matched in their characteristics. Backing the firm-level results, the loan-level analysis indicates that high ESG performing firms dominantly raise sustainability-linked debt manifesting in a statistically significant initial spread discount of up to 69.42 basis points on average for ESG-linked versus regular loans.
Second, in order to identify the extent to which credit risk explains the effect of CSR performance on cost of debt, a non-linear multiple panel regression of cost of debt on ESG Combined and Controversies scores that introduces the interaction effect between long-term credit ratings and ESG performance is estimated. The methodology thus contrasts the approach taken by Goss and Roberts (2011) or Liu and Ge (2015), who merely controlled for credit risk by incorporating Altman’s Z-score or bond ratings when regressing cost of debt proxies on raw KLD scores. Ultimately, the results on the credit risk channel expose an altered credit risk profile to act as the driving force in the observed negative relation by partly capturing the impact of sustainability on cost of debt and vice versa. Yet, in line with prior research, the findings confirm the statistical significance of ESG concerns for altering the corporate credit risk profile whereas net ESG performance is found to lack evidential power. Ultimately, by performing various robustness tests, the possibility of an alternative channel other than credit risk driving the ESG-cost of debt relation can be rejected with sufficient confidence.
Research Contribution Given the research findings obtained, this thesis contributes towards enhancing the nonconclusive understanding of the ESG-cost of debt relation and extents previous academic work to the post-financial-crisis period. By focussing on Europe as geographical area of little academic attention but high practical relevance, this study also aims to fill the gap in research on European bilateral debt. Particularly, the work at hand expands current research on a not-yet studied phenomenon in corporate lending, namely sustainability-linked loans, and thus proposes a meaningful contribution to sustainable finance literature. Addressing the inconsistent view on the ESG-cost of debt link and its potential channels, this study, however, is not only relevant for scholars. Also, it is important in a practical setting given its implications for including measures of CSR into traditional credit risk assessments by rating agencies, for raising capital through sustainability-linked loans by corporations as well as for the pricing and underlying credit risk assessment of such facilities by financiers. Not least, this study establishes a compelling case for sustainability-linked loans in Europe and provides academic support on the tangible value benefits achievable in facilitating the transition towards a more sustainable and resilient economic model.
Concluding, the remainder of this thesis is structured as follows. The second chapter starts with providing an extensive overview of the relevant academic literature with increased focus on the role of CSR in corporate finance and its relation to cost of capital. This section is further concerned with the formulation of testable hypotheses to provide an answer to the research question raised. In the third chapter, the process of sampling the data and constructing the associated variables as well as the methodology used for testing these hypotheses are described. The fourth chapter highlights the main results of this thesis and presents robustness checks for alternative model specifications and for addressing endogeneity concerns. Lastly, the fifth chapter concludes with a critical assessment of limitations and suggestions for future research.
2 Theoretical Foundation and Hypothesis Development
The following chapter reviews the existing academic literature with respect to the role of CSR in financial theory and thus illustrates the research context of this study. Ultimately, testable hypotheses are formulated based on the identified gaps in the present state of literature.
2.1 Corporate Social Responsibility
Absent of a generally accepted definition of the term corporate social responsibility, the European Commission (2011) views CSR as “the responsibility of enterprises for their impacts on society”. The Commission clarifies that, in order to fully adhere to their CSR, firms are required to “integrate social, environmental, ethical, human rights and consumer concerns into their business operations and core strategy in close collaboration with their stakeholders [...]”. As such, the understanding put forward by the commission contrasts the traditional shareholdercentric perspective for which businesses’ only social responsibility lies in increasing its profits and thus maximizing shareholder value (Friedman, 1962; 1970).
Rather, this understanding emphasizes a broadened attention towards a larger group of related parties, i.e. the firms’ stakeholders. The concept of CSR is therefore, on the one hand, rooted in the stakeholder approach suggested by Freeman (1984) arguing that the corporate objective is interconnected with the individual interests of distinct internal and external groups. In fact, CSR is understood to go beyond this instrumental perspective by emphasizing the normative and managerial dimension of the stakeholder theory (Donaldson & Preston, 1995).
On the other hand, CSR simultaneously builds on the idea of the triple bottom line, arguing that performance should be assessed against the three dimensions social justice, economic prosperity and environmental quality. Therefore, the measurement approach suggested by Elkington (1997; via Elkington, 1998) goes beyond the traditional view of taking financial performance as sole indicator but simultaneously focusses on “people, planet and profit.” In this vein, CSR can be viewed as a “concerted effort to incorporate economic, environmental and social considerations into a company’s evaluation and decision-making processes” (Wang & Lin, 2007). Similarly, Schoenmaker and Schramade (2019b) define a dynamic sustainable finance framework evolving from shareholder value focus towards creation of a common good or system value as a last stage (cf. Tirole & Rendall, 2017). Hence, the concept of CSR resembles the transition from a shareholder centred perspective over an integrated shared value and towards a system value approach (see Figure 1).
Figure 1: Transition of Management Practices based on CSR
Abbildung in dieser Leseprobe nicht enthalten
Source: Own illustration based on International Integrated Reporting Council [IIRC], 2019, p. 11
Behind this theoretical appeal for the shared value idea, the concept of CSR may be understood as closely related to the corporate strategy and as such a potential influence on the competitive standing of the individual firm. However, companies’ individual CSR approaches are still rather fragmented and entangled from their core business which leads to misconceptions about the value of CSR for creating new business opportunities beyond being a mere zero-sum game (Porter & Kramer, 2006). Rangan, Chase, and Karim (2012) thus similarly argue that the implementation of a company-specific CSR strategy requires a generally accepted framework of how to reconcile various CSR initiatives in a structured fashion. In other words, the transition from the shareholder perspective viewing CSR as a redundant cost item towards acknowledging the opportunities from integrated value thinking remains rather slow as there “[...] are different catalysts for the various CSR programs that a company might undertake, and depending on the motivations, the expected outcomes of the programs are also quite varied” (Rangan et al., 2012).
Beyond this issue, there is still little clarity about how companies can assess their progress on sustainability issues in a structured and consolidated manner even though it is possible to measure CSR performance against indicators that relate to the goals of the triple bottom line (Elkington, 1998) - a problem amplified by capital markets increasingly demanding transparency over firms’ integrated performance (Chauhan & Kumar, 2018). As such, for the individual firm to possibly capture competitive advantages from CSR by adapting towards true stakeholder orientation requires regular assessments and disclosures of those non-financial indicators in a coherent way (see Embong, Mohd-Saleh, & Sabri Hassan, 2012).
Environmental, Social and Governance Criteria
Following this need for quantifying CSR efforts, the use of measurable environmental, social and governance criteria (ESG) has not only become an established practice among researches (see e.g. Sharfman & Fernando, 2008; Albuquerque et al., 2018). Also, employing ESG criteria for operationalizing CSR in financial analysis and decision-making is increasingly adopted by market participants (Bassen & Senkl, 2011). However, a large number of ESG measures available with a lack of standardization and regulation across providers hinder wide-scaled ESG adoption among practitioners such as credit rating agencies (see Schoenmaker & Schramade, 2019b). Given these differences in methodologies, Semenova and Hassel (2014) find that environmental scores assigned by major providers on a sample of 466 US firms from 2003 to 2011 do not converge on average. Similarly, Chatterji, Durand, Levine, and Touboul (2016) document little correlations (at most 0.53 on average) among the scores of six leading providers on an international firm sample between 2002 and 2010. As such, these findings imply that market participants, rather than relying blindly on aggregate ESG performance, should pay increased attention to single indicators that matter most to the operations of the individual firm.
Behind this background, Eccles, Krzus, Rogers, and Serafeim (2012) thus demand the application of sector-specific reporting standards for ensuring public disclosure of material nonfinancial information. In this context, the concept of materiality for ESG factors builds on the definition for financial information relating to the extent of their substantial impact on decisionmaking and value creation (see e.g. the International Integrated Reporting Council [IIRC], 2019). Although the adoption of its guidelines remains voluntarily to date, the Sustainability Accounting Standards Board (SASB) presents the most prominent approach of establishing sector-specific standards for ESG reporting by indicating which criteria have a material impact on companies within a certain industry. Applying material ESG criteria as deemed by SASB, Khan et al. (2015) provide empirical evidence on the importance of ESG materiality. Particularly, by examining 2,307 unique US firms, the authors find that firms performing better on SASB-material ESG issues exhibit superior profitability measures. Similarly, Grewal, Serafeim, and Yoon (2016) studied 2,665 shareholder proposals between 1999 and 2013 and demonstrate that filings on SASB-material ESG issues are associated with appreciation in firm value whereas proposals on immaterial issues are associated with a decline. These findings thus stress the idea that the disclosure and effective management of industry critical ESG issues only can yield a competitive edge to the respective company. Further, the evidence implies that cross-sector comparisons of ESG performance should account for the material ESG issues of the respective industries observed.
2.2 The Role of ESG in Corporate Finance
This materiality of CSR within the boundaries of the firm was long neglected under the neoclassical finance paradigm. Most prominently, Friedman (1970) formulated the idea that the sole responsibility of the firm was to increase its profits, whereas social and environmental issues were strictly of governmental concern. Essentially, the neoclassical view regards CSR as a redundant cost item depleting the resource base of the firm for which it would incur a competitive disadvantage (see e.g. Aupperle, Carroll, & Hatfield, 1985; Jensen, 2002). A similar argument labelled as overinvestment hypothesis (see Jo & Harjoto, 2012) states that firms overinvest into CSR for the reputational benefit of managers and at the expense of both shareholders and creditors ultimately evoking agency conflicts (Barnea & Rubin, 2010).
With the emergence of the alternative perspective to align business actions with the interests of a broader group of related parties, i.e. stakeholders (see Freeman, 1984), CSR performance is increasingly viewed as a potentially value-relevant opportunity for the firm. Under the assumption that CSR would be essential for obtaining particular resources, improving the firm’s resource base, and securing stakeholder support (see Jones, 1995), the multi-stakeholder view gained increased acceptance among scholars and corporations alike. Behind this background, one line of reasoning states that strong ESG performance yields firm-level reputational benefits which manifest in stronger external relationships to customers, suppliers, financiers etc. Most important however, a positive firm image potentially improves the quality of firms’ human capital by attracting better-skilled, highly motivated employees (see for example Russo & Fouts, 1997; Waddock & Graves, 1997; Orlitzky, Schmidt, & Rynes, 2003).
A second argument targets the operative dimension under which ESG performance is associated with improved internal processes, increased efficiency, and more resilient information systems (see Orlitzky et al., 2003; Eccles, Ioannou, & Serafeim, 2014). On this basis, Russo and Fouts (1997) argue that redesigning production, service, or inter-company processes by adopting environmental policies would improve the quality and availability of competition-critical resources. In fact, with the transition from shareholder theory to resource-based view (c.f. Barney, 1991), investments in CSR are now widely accepted as prerequisite for securing critical resources, maintaining competitive standing, and increasing preparedness for external change. Although those distinct benefits of strong ESG performance might be difficult to isolate and potentially depend on markets’ perception of CSR efforts (c.f. Lo, 2005; 2017), their combined effect is expected to result in increased firm value via improving value drivers such as firm financial performance or cost of capital (see Schramade, 2016).
2.2.1 ESG and Firm Performance
Undoubtedly, investments in ESG criteria come at certain costs which are to be measured against subsequent financial benefits. As opposed to examining stock price performance as market-based proxy of firm performance, Orlitzky et al. (2003) suggest that it is especially accounting-based measures such as a firm’s return on assets, return on equity, or earnings per share that tend to be strongly influenced by responsible business practices. This idea is consistent with the findings of Friede et al. (2015), who, by performing a review analysis of more than 2,200 unique studies, find only a weak relation between ESG and investor performance but discovered that c. 90% of studies find at least a nonnegative link between ESG and firm performance. Hence, sought-after tangible benefits for the individual firm are expected to predominantly manifest in the direct link between ESG and accounting-metrics rather than the indirect effects on stock or investor performance (cf. Schoenmaker & Schramade, 2019b).
Providing first evidence, Russo and Fouts (1997) regressed return-on-asset ratios (RoA) of 477 firms between 1991 and 1992 on their respective environmental rating to find a positive relationship between environmental and firm performance. Similarly, Waddock and Graves (1997) find a statistically significant positive impact of social performance on RoA as well as return-on-equity ratios (RoE) and return-on-sales ratios (RoS). The authors link their finding to the “good management theory” stating that increased attention to social aspects such as employee relations causes the improvement of firm performance (see stakeholder view in 2.2) Conducting a meta-analysis of 52 studies with a total of 33,878 observations, Orlitzky et al. (2003) confirm the positive correlation between both environmental and social aspects and firm performance. More recent evidence was put forward by Guenster, Bauer, Derwall, and Koedijk (2011), who document a positive, though asymmetric relationship between environmental resource efficiency and both firm value and operating performance as indicated by return on assets. In addition, Cheng, Ioannou, and Serafeim (2014) examined a cross-industry sample from 49 countries and find that high CSR performing firms face fewer financial constraints. Interestingly, the authors identify increased CSR disclosures (see 2.1) and better stakeholder engagement (see 2.2) as channels through which CSR performance affects financial constraints. Similarly, Eccles et al. (2014) find that high-sustainability firms are more likely to establish formal stakeholder engagement, tend to act more long-term orientated, and exhibit a higher degree of CSR disclosure for which they enjoy better values for RoE and RoA. Khan et al. (2015) further find that firms with high performance on material ESG issues (see 2.1) enjoy relatively better future RoS and sales growth whereas performance on non-material aspects would yield no relative differences in performance metrics .
As such, the link between ESG and firm performance seems overall well-established. The lack of corporate action-taking on sustainable matters thus contrasts this empirically well-backed view on the positive influence of ESG on traditional value drivers. This apparent contradiction implies that firms still see a mismatch in the evaluation of the future long-term benefits of sustainability against their short-term financial goals (see Schoenmaker & Schramade, 2019a). Therefore, the sought-after economic benefits incentivizing near term action-taking might be realized in another altered value driver, namely firms’ cost of capital.
2.2.2 ESG and Cost of Capital
Claiming that investments in ESG related issues would have the potential to alter a firm’s corporate risk profile, scholars assumed to find this manifested in the corporation’s cost of capital (CoC). While the bulk of research efforts is empirical in nature, Heinkel, Kraus, and Zechner (2001) propose an equilibrium framework for explaining how investments in ESG would impact CoC. The authors draw a simple risk-averse setting involving only investors fully ignoring ethical considerations and green investors who rule out investments in unethical firms. With the resulting lack of risk-sharing among green investors and thus in the equilibrium state, stock prices for polluting firms are expected to decrease and their CoC to increase.
First empirical evidence on the relationship between CSR and CoC has been provided by Sharfman and Fernando (2008). In a study of 267 public US firms, the authors apply the CAPM model to find a significantly negative relationship between a firm’s environmental risk performance and cost of equity (CoE) acting as the driving force for the lowered CoC. Other evidence relying on an asset pricing model to explain the ESG-CoE relationship is provided by Gregory, Tharyan, and Whittaker (2014), who find a 0.19% difference in CoE between “green” and “toxic” firm portfolios by using the Fama-French (1993) 3-factor model. In a later study, Gregory, Whittaker, and Yan (2016) employ the Carhart (1997) 4-factor model to reveal that the found inequality in CoE for high and low ESG firms can largely be attributed to industry affiliation. This finding is thus consistent with the idea that ESG materiality differs across industries (see 2.1). Regression-based evidence on the link between sustainable performance and CoE is found by El Ghoul et al. (2011), who estimate the ex-ante CoE implied in stock price data and analyst information by using various accounting-based valuation models. Based on a sample of 2,809 unique firm characteristics examined between 1992 and 2007, the authors find that companies scoring higher on corporate social responsibility matters enjoy statistically significant lower cost to their equity capital, partially due to lower perceived firm risk.
More recently, Albuquerque et al. (2018) further contribute to this line of reasoning by regressing beta as systematic risk measure on an overall responsibility score, which demonstrates a statistically and economically significantly negative relationship between corporate sustainability and CoE. The largely observed negative relationship between ESG and CoE is further backed by single-country evidence for example from the UK (see Salama, Anderson, & Toms, 2011), France (see Girerd-Potin, Jimenez-Garcès, & Louvet, 2014), China (see Li, Liu, Tang, & Xiong, 2017) and Japan (see Suto & Takehara, 2017).
While the relationship between ESG and CoE has been empirically well-established, significantly less academic attention has targeted the influence of CSR on cost of debt (CoD). Moreover, among the few studies engaging with the impact of sustainable performance on CoD several articles suggesting counterintuitive results exist. For instance, Sharfman and Fernando (2008) examined a positive relationship between environmental performance and cost of debt. Further, Menz (2010) expected that high performing CSR firms would face lower credit spreads on a sample of 498 European corporate bonds between 2007 and 2010 but the results do not confirm a statistically significant relationship. Interestingly, Menz attributed this result to the relatively higher importance of credit ratings than CSR ratings for bond investors, i.e. the public debt market. Contrary to these findings, subsequent research by Bauer and Hann (2010) documented the opposite direction in the CSR-CoD relationship. The authors investigated bond issues of 582 public US firms between 1995 and 2006 and find that bond investors demand higher issue spreads from corporations with environmental concerns. These contrary results point towards information deficiencies with respect to CSR strengths and concerns in public debt markets (see discussion in 2.3). Examining private debt markets, Chava (2014) reports significantly higher interest rates on loans to corporations dealing with environmental concerns based on a sample of 5,879 loan facilities to 1,341 US firms. Using a sample of 3,996 loans to US borrowers, Goss and Roberts (2011) too propose that banks charge on average 7 to 18 basis points higher loan spreads to low-quality CSR firms. Notably, the reverse effect, i.e. lower loan spreads to high-quality borrowers, was not observed. Analogous, Magnanelli and Izzo (2017) investigated the effect of CSR on cost of bank debt and found a statistically significant positive relationship on an international cross-industry sample of 332 firms between 2005 and 2009. An interesting study is conducted by Suto and Takehara (2017), who examined a positive relationship between CSR and the ex-ante cost of debt on a Japanese firm sample between 2008 and 2013. Since the influence of CSR was statistically significant only between 2008 and 2010 however, their results imply that in the years after the financial crisis market participants became more aware of non-financial disclosures in general and ESG performance especially.
Most similar to this thesis, the studies of La Rosa, Libertore, Mazzi, and Terzani (2017) observe a negative ESG-CoD relationship on an European firm sample between 2005 and 2012 but limit their attention to firms’ social performance. Given these counterintuitive findings for the CSR- CoD relation, the question how CSR is impacting borrowing costs persists to date. However, if superior ESG performance does indeed improve firm financial performance and the corporate systematic risk sensitivity, then the expectation arises that ESG performance should equally have an impact on the firm-level cost of debt. Hence, the following hypothesis is formed:
Abbildung in dieser Leseprobe nicht enthalten
The importance of understanding the effect of CSR on CoD becomes further evident with the emergence of sustainable lending concepts. Historically, lending institutions emerged under the premise to foster the allocation of capital as well as for facilitating the trading and management of risks (Levine, 2004). First, with this longstanding role understanding comes a crucial responsibility of lenders in the transition towards a more sustainable economy (see for example Scholtens, 2017). This responsibility of facilitating the economic transition is especially pronounced for European banks since European firms traditionally rely on bank debt as primary source of external financing as opposed to their US counterparts (IMF, 2019). Such imbalance implies a relatively higher exposure to corporate default risks from lending to European firms for which European debt markets are expected to be especially sensitive to the influencing factors on borrowers’ financial performance in general and non-financial information such as ESG performance (see 2.2.1) especially. Moreover, grounded on relationship lending practices, private debt markets are commonly regarded as more information efficient than public debt markets (see Diamond, 1984; Altman, Gande, & Saunders, 2010) for which private lenders such as banks are understood as particularly well-aware of the different layers of the corporate risk profile. This is due to the fact that private lenders are seen to have a competitive advantage in obtaining (non)-financial information ex-ante in the lending process for which asymmetric information between borrower and lender are minimized - a condition providing fertile ground for studying the effect of ESG performance on the cost of bilateral debt for European firms.
Secondly, the successful transition requires lenders to value sustainability by internalizing environmental and social externalities along the value chain. Yet, while several non-regulatory initiatives have been launched to facilitate the integration of sustainability into lending operations (e.g. the Equator Principles or the Global Alliance for Banking on Values), the accounting for sustainability in day-to-day corporate borrowing was until recently non-existent.
Behind this background and given the well-established positive relation between CSR and firm performance, it comes as no surprise that European banks have played the pioneering role in internalizing ESG with the structuring of the first sustainability-linked loan facility by ING Bank N.V. to a European corporate, namely Koninklijke Philips N.V., in early 2017.
According to the Loan Market Association, Asia Pacific Loan Market Association, and Loan Syndications & Trading Association (2019) as the issuers of the so-called Sustainability Linked Loan Principles in March 2019, sustainability-linked loans generally involve a pricing adjustment during the loan life given the achievement of “[...] predetermined sustainability performance objectives [...] measured using sustainability performance targets [...]”. Important to note however, sustainability-linked loans, unlike green bonds, are not restricted in their use of proceeds and are in fact mainly granted for general corporate purposes. Furthermore, while the principles encourage the provision of information regarding the individual CSR strategy and targets, borrowers are not subject to increased reporting requirements to banks. Rather, the predetermined sustainability performance objectives are oftentimes expressed and measured using external ESG scores (see 2.1; LMA, 2019). In simple terms, while sustainability-linked loans are special in their pricing structure, they are comparable to conventional loans with respect to the use of proceeds, the regular reporting requirements, and hence the degree of information asymmetries in the lending relationship.
The emergence of sustainability-linked loans (or similarly referred to as ESG-linked loans) thus emphasizes lenders’ willingness to reward borrowers for improvements on predetermined ESG benchmarks during the life of the lending relationship. As such, sustainability-linked loans may serve as the potential vehicle through which banks can incorporate sustainability into lending and thus as the long-missing puzzle piece for providing the tangible economic incentives for corporates in the transition towards a sustainable economy. Given these potential benefits to borrowers, one would expect that especially ESG frontrunners have a strong incentive to raise sustainability-linked debt. Therefore, if there is in fact an observable significantly negative relationship between ESG and firm-level cost of debt, this should then equally manifest in reduced initial loan spreads on sustainability-linked loan facilities as compared to conventional debt contracts. Consequently, to leverage the information efficiency of private debt markets for creating a level comparison setting between sustainability-linked and regular debt, the first hypothesis will be supplemented by examining loan-level evidence. Ultimately, taking on a private debt market view supports the first hypothesis by investigating whether ESG efforts can pay off instantly, i.e. whether the expected ESG-CoD relationship manifests in initial loan spread differences between conventional and sustainability-linked loans.
2.3 Credit Risk in the ESG - Cost of Debt Relation
In light of the emergence of sustainability-linked loans as innovative lending concept it becomes increasingly visible that lenders are willing to financially reward borrowers upon improving ESG performance during loan life. As such, looking at the incentives for lenders to incorporate ESG information in addition to traditional financial indicators into lending practices is the crucial prerequisite for understanding the three distinct economic channels through which sustainability translates into reduced funding costs. According to Schoenmaker and Schramade (2019b), the integration of sustainability within debt markets lies on a spectrum from the full exclusion of credit to poor ESG firms to pure value-based lending supporting potentially meaningful borrowing causes. Commonly cited theories to explain creditors’ incentives and thus the ESG-CoD channels are however found in the midst of the spectrum, namely the conflict-resolution and risk-mitigation hypothesis.
The conflict-resolution view arises from the asymmetric pay-off of debtholders in relation to the residual claim of shareholders that is more symmetrical to the firm’s overall performance. This mismatch in the interests of debt and equity holders opens potential agency conflicts at the expense of lenders. Bhojraj and Sengupta (2003) thus relate lower CoD to especially good performance on the governance dimension of ESG, manifested either through higher institutional ownership or more outside board directors. More specific, Ashbaugh-Skaife, Collins, and LaFond (2006) argue that more anti-takeover mechanisms would relate to higher CoD. Consequently, governance strengths embedded in the conflict-resolution view may help to reduce potentially costly agency conflicts and thus act as one potential economic channel through which lenders are willing to charge lower interest rates to high ESG performing firms.
Besides, the risk-mitigation view is at the heart of traditional banking theory understanding credit risk as main determining factor for the pricing of debt (Merton, 1974). Hence, the assumed negative ESG-CoD link might be explained if ESG frontrunners are either perceived to be less risky or do in fact display less default risk rendering the justification for charging lower rates. Based on the former, Sengupta (1998) provides evidence that firms making timely and informative disclosures reduce information asymmetries and thus lenders’ perception of default risk for which they would charge lower risk premia. Further, Orens, Aerts, and Cormier (2010) find that disclosure of web-based non-financial information mitigates information asymmetries and yield lower CoD for large listed European firms. Due to the study’s limited focus and a missing distinction between positive and negative disclosure, this result may however not be generally applicable to the ESG-CoD debate (c.f. Semenova & Hassel, 2014).
Subsequent research on Chinese firms finds high CSR disclosure quality to be associated with lower cost of corporate bonds (see Gong, Xu, & Gong, 2018). Similarly, Fonseka, Rajapakse, and Richardson (2019) report a negative relationship between environmental disclosures and CoD for the Chinese market, which is comparable to the European debt market in terms of the relative importance of financial market segments (see Schoenmaker & Schramade, 2019b). On the contrary, Chauhan and Kumar (2018) examine the effect of ESG disclosures on the valuation of Indian firms but find no statistically significant relation between the level of ESG disclosures and CoD. Studying the channels through which such potential value effect unfolds based on a sample of public US companies from five polluting industries, Clarkson, Fang, Li, and Richardson (2013) find that the incremental value effect of voluntary environmental disclosures manifests in higher future profitability instead of reduced CoC. Overall, although academic evidence on the relationship between CSR disclosures and CoD is still limited and mixed, taken together, the results suggest that reduced information asymmetries might act as a potential second economic channel through which ESG unfolds in lower CoD.
Initial support for the third potential channel, i.e. that ESG performance in fact translates into less default risk is provided by Lee and Faff (2009), who demonstrate that socially responsible firms face lower idiosyncratic risk manifesting in reduced cash flow volatility. In this vein, Attig, El Ghoul, Guedhami, and Suh (2013) examined that rating agencies reward firms’ superior social performance with comparatively better credit ratings. Interestingly, factors relating to primary stakeholder management such as employee relations (see e.g. Bauer, Derwall, & Hann, 2009) are deemed particularly meaningful in assessing firms’ creditworthiness - an idea consistent with the stakeholder theory. Widely cited, Mishra and Modi (2013) find that overall ESG performance is equally associated with reduced idiosyncratic risk. However, the authors point out that the marginal effect of CSR on idiosyncratic risk becomes smaller with increasing levels of leverage. This conclusion renders the assumption that lenders might increasingly turn from non-financial indicators such as ESG performance towards traditional credit risk measures the closer firms get to entering financial distress. In this vein, Erragragui (2018) observes a panel of 214 US firms contributing that only environmental concerns would matter in assessing credit risk while governance concerns did not. This revealed “governance paradox” supports the risk-mitigation view in favour of the conflict-resolution theory. Consequently, given the mixed academic support on the role of governance strengths and information asymmetries in the ESG-CoD relationship, this thesis acknowledges the established alternative channels but focusses on the empirically well-backed credit risk channel under the risk-mitigation view as expected vehicle through which ESG performance unfolds.
Existing studies on the ESG-CoD effect have indeed acknowledged the importance of credit risk, mostly proxied for by employing firms’ credit ratings. Yet, only recently, very few scholars deliberately started considering the direct relationship between CSR and credit risk (instead of merely controlling for credit risk) as the potential channel through which the impact of corporate sustainability on CoD is manifested. The first study to investigate this channel was conducted by Jiraporn et al. (2014) who find that CSR strengths sourced from the KLD database are positively correlated with a firm’s credit rating. In fact, the authors show on a sample of 2,516 firm-year observations from 1995 to 2007 that an increase in CSR by one standard deviation could improve long-term credit ratings by up to 4.50%. While the authors subsequently do not examine the impact of CSR on CoD, they suggest that the negative relationship between CSR and credit ratings would ultimately result in lower funding costs.
Building on the work of Menz (2010), the most important research on the credit risk channel however is brought forward by Liu and Ge (2015). Based on a large sample of US firms during the period 1992-2009, the scholars use credit ratings as ex-ante CoD to first find that KLD strengths on environmental dimensions are associated with improved credit ratings. Second, after controlling for credit ratings in the regression of bond yield spreads on CSR performance, the results show significantly lower yield spreads for firms with pronounced CSR strengths whereby the credit risk channel partly captured the impact. Interestingly, the study reveals that while overall scores had an impact on CoD, for the dimensions “environment”, “community” and “governance”, only the strengths resulted in decreasing yield spreads in the sample. Similarly, Bae, Chang, and Yi (2018) find on a US loan sample over the period of 1991-2008 that ESG strengths are associated with lower firm-risk and thus reduced lending spreads while ESG concerns would be captured by firms’ credit ratings. As such, the beforementioned studies imply a differentiated effect between positive ESG performance and ESG concerns in their manifestation in an altered risk profile. Further indicated by those findings, if ESG performance significantly impacts credit risk then the effect on CoD should be partly driven by the interaction between the ESG score and credit risk. This implication builds on the risk-mitigation view under which lenders employ a risk-based rather than a value-based or other approach for integrating ESG ratings into debt pricing and that credit risk is indeed the main determinant of borrowing costs (see Merton, 1974). Ultimately, based on this theoretical foundation established, the following hypothesis is developed:
Abbildung in dieser Leseprobe nicht enthalten
3 Data and Methodology
The following chapter describes the data and methodology for examining the hypotheses developed. First, the process of data collection and the main characteristics of the samples are portrayed. Second, the construction of relevant variables and the empirical models for analysing the firm- and loan-level samples and the associated hypotheses are described. Within this conceptual framework developed (see Figure 2), comparisons are drawn between data and model characteristics in this thesis to those employed in related studies.
Figure 2: Conceptual Framework
Abbildung in dieser Leseprobe nicht enthalten
Source: Own Illustration; 1) In testing H2, ESG Performance relates to the aggregate of “net” ESG Performance (ESGC) and ESG Controversies
3.1 Data for the ESG-CoD Relationship
Contrary to other research, which samples ESG strengths and concerns from KLD STATS (see e.g. Jo & Na, 2012; Bouslah, Kryzanowski, & M'Zali, 2013; Oikonomou, Brooks, & Pavelin, 2014; or Liu & Ge, 2015), this study employs ESG Scores from Thomson Reuters ASSET4 due to their more objective nature and the availability of an overall score in contrast to KLD. In addition to an overall ESG score measuring firms’ “gross” ESG performance, Reuters further publishes an ESG Combined (ESGC) score available on all covered entities. The ESGC overlays the ESG score with ESG controversies to exhibit a more comprehensive scoring of firms’ “net” ESG performance. As such, Thomson Reuters was further utilized based on the suitability of the ESGC score for examining the specific research focus of this thesis.
[...]