Exploring the insurance-like benefit of corporate social responsibility in the context of a systemic crisis

(1)

Exploring the Insurance-like Benefit of Corporate Social

Responsibility in the Context of a Systemic Crisis

Date and version: 01/07/2016, final version Name and student number: Nadine van Meel, 10263314

Supervisor: dr. P. Vishwanathan

(2)

2

STATEMENT OF ORIGINALITY

This document is written by Student Nadine van Meel who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

ABSTRACT

Literature on the relationship between corporate social performance (CSP) and corporate financial performance (CFP) in the context of a systemic crisis is scant and inconclusive. This empirical study investigates whether the insurance-like benefit of social responsibility persists in recessionary times. It is hypothesized that the negative consequences of an economic downturn – among which a loss of purchasing power is prominent – moderate the CSP-CFP relationship in such a way that the positive association that exists prior to the financial and economic crisis of 2007-2009 is reduced or entirely eliminated. Taking a longitudinal approach, the regression model is tested using a sample set of S&P500 companies in the period 2001-2013. Contrary to the hypotheses, the research shows a significantly positive relationship between CSP and CFP that occurs only under circumstances of systemic crisis for companies that perform well with respect to social responsibility. The findings suggest that an insurance-like benefit of CSR is only present in times of economic turmoil for high-performing companies. Furthermore, the national income growth rate (the measure used to capture macro-economic changes) does not seem to be of any moderating influence under systemic crisis conditions. Lastly, analysis is presented concerning how companies adjusted their investments in social responsibility during the 2007-2009 crisis.

(4)

4

1 INTRODUCTION

Albeit that social responsibility caught the attention of a wider public only in recent years, the relationship between corporate social performance (CSP) and corporate financial performance (CFP) has been under scrutiny of scholars since the late 1970’s. From then onwards, meaningful and influential studies (Barnett, 2007; Godfrey, Merrill & Hansen, 2009; King & Lenox, 2002; Orlitzky, Schmidt & Rynes, 2003) indicate a positive and significant association between the two constructs.

From the late 2000’s, corporate social responsibility (CSR) research concentrates on testing the causal mechanisms that underlie the CSP-CFP relationship. The meta-analysis of Aguinis and Glavas (2012) provides an extensive overview of all predictors and outcomes of CSR activities, and the mediators and moderators that affect the relationship between CSR and its outcomes that have been studied in the academic field – either conceptually or empirically. These factors are originally researched on a single level of analysis only and can be classified into factors operating on institutional, organizational and individual levels of analysis. The authors synthesized and integrated the body of CSR literature into a theoretical guiding framework with a multilevel perspective. Their ‘Multilevel and Multidisciplinary Model of CSR’ constitutes a valuable contribution to literature. However, a potential moderator that has been overlooked by CSR literature is the impact of circumstances that affect the entire economic system (Ducassy, 2013; Mio & Fasan, 2012; Schnietz & Epstein, 2005). To what extent the relationship between CSP and CFP is affected by severe economic market shocks remains largely unexplored. This is remarkable, given that negative market shocks can have far-reaching consequences: e.g. a downturn of economic activity, a loss of purchasing power of both businesses and consumers, and increased unemployment (Flammer & Ioannou, 2015; McKibbin & Stoeckel, 2010). In light of the recent economic and financial crisis, it is highly relevant for diverse purposes – theoretical as well as managerial – to

(6)

6 examine the potential moderating effect of tough economic times on the CSP-CFP relationship.

In this study, the change in economic returns resulting from social responsible activities during tough economic times will be examined. I argue that a systemic crisis moderates the relationship between CSP and CFP in such a way, that the positive association is offset – or at least weakened – by the negative effects of an economic downturn. Under normal economic circumstances, the beneficial consequences of CSR encompass among others creating business value, consumer loyalty, enhancing organizational identification and insuring against reputational risks (Aguinis & Glavas, 2012; Auger et al., 2003; Einwiller, et al., 2006; Margolis, Elfenbein & Walsh, 2007; Orlitzky & Benjamin, 2001; Peloza, 2006). An economic deterioration prevents the aforementioned benefits associated with CSR activities from occurring, resulting in overall non positive economic results from CSP. For the purpose of comparability, this relationship will be investigated by use of U.S. data only. Factors of influence on the CSP-CFP relationship other than economic circumstances should preferably be eliminated in order to draw meaningful conclusions from the research results. The use of U.S. data satisfies this ‘ceteris paribus’ requirement best, since the similarities between the States’ institutional environments minimize regulatory differences that could affect the CSP-CFP relationship. Therefore, the impact of the financial crisis is largely unambiguous in the U.S.

Contributions of this study to the field of CSR research are threefold. First, this research integrates CSR perspectives from an organizational and macro-economic level, thereby responding to the call to conduct CSR research based on a multilevel perspective (Aguinis & Glavas, 2012). More specifically, this study takes into account the macro-level effects of a systemic crisis, and incorporates a moderator for these effects into in an empirical investigation of the CSP-CFP relationship at the organizational level of analysis. In addition,

(7)

7 Aguinis and Glavas (2012) highlight that research methodologies tend to be cross-sectional. Since CSR activities usually do not show benefits immediately after implementation (Jo, Kim & Park, 2015; Rangan, Chase & Karim, 2015; Salama, 2005), but instead involve longer-term processes, there is a need for longitudinal research methodologies. Long-term studies are preferable over portfolio studies and event studies because of higher reliability (Ambec & Lanoie, 2008). This study takes a longitudinal approach to CSR research. Second, this study highlights the importance of the moderating role of economic circumstances on the relationship between companies’ social performance and their financial performance. When this moderator is pushed more to the forefront of the research area, it will hopefully catch scholarly attention in order to acquire more knowledge upon this topic. Finally, when managers in turn understand the effects of severe economic circumstances on the profitability of their CSR policy, they will be able to make more deliberate business decisions regarding their company’s socially responsible activities.

This research is empirical in nature. The sample set consists of 291 companies from the Standard & Poor’s 500 (S&P500) index in the time range 2001-2013. Using financial and non-financial company data, and net national income per capita growth as a macro-economic variable, the average effect of the financial crisis on the CSP-CFP relationship will be estimated by linear regression technique. Here, a distinction is made between companies that perform well on CSR and companies that perform low relative to the sample set.

The paper is structured as follows: paragraph 2 will review the existing literature on CSR research per se and in the context of the financial crisis. Here the relevant theories are outlined and the hypotheses are formulated. Then paragraph 3 raises the matter of an adequate research design for testing the hypotheses. Also, the characteristics of the data and selection criteria for investigation will be described into more detail. Subsequently, paragraph 4

(8)

8 provides the empirical results of the research, followed by an analysis of the results in paragraph 5. Lastly, the conclusion presents an overview of the main findings of this study and its contributions to theory and practice.

2 THEORY AND HYPOTHESES

2.1 The CSP-CFP relationship, stakeholder theory and reputation

It cannot go unnoticed that CSR has become common practice for companies nowadays. Many companies have separately owned charitable foundations, and reducing the economic footprint is a mainstream business activity. But what exactly is CSR? Several definitions of CSR are proposed both in the scholarly literature and by institutions. McWilliams and Siegel (2001, p. 117) define CSR as “actions that appear to further some social good, beyond the interests of the firm and that which is required by law.” The World Bank (2016, January 23) lists different terms that refer to the practice of CSR (e.g. sustainable development, triple-bottom line and corporate ethics), and captures its notion by stating that these terms are bound together by “the expectation that corporates (private and public enterprises alike) behave ethically vis-à-vis a broad group of stakeholders – workers and their families, communities and the wider society.” The European Commission (2001) offers the following definition: “CSR is a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis”; Vogel (2006) refers to this by using “beyond compliance”. The similarity of the above definitions of CSR is found in that they put emphasis on the voluntary and discretionary involvement in the solution of a variety of social issues (Banerjee, 2008; Brammer & Pavelin, 2006). The term corporate social (or environmental) performance (CSP) is frequently used to refer to the observable and measurable output of CSR (Kitzmueller & Shimshack, 2012).

(9)

9 Having established that CSR means going beyond obeying legal or regulatory requirements, this raises the question why companies engage in CSR activities. What is in it for companies themselves? Over the past decades, researchers have investigated the relationship between CSP and CFP. With some exceptions, the relationship has been found to be significantly positive. Inconclusive results with regard to this relationship principally concern companies operating in the service sector, and small enterprises or new ventures (Garay & Font, 2012; Lee, 2008; Wang & Bansal, 2012). Yet meta-analyses in the field of CSR research find an overall positive effect for the CSP-CFP relationship (Margolis, Elfenbein & Walsh, 2007; Orlitzky, Schmidt & Rynes, 2003). Stakeholder theory helps explaining the positive association. Founder of stakeholder theory is R. Edward Freeman (1984), who came up with the notion that companies ought to create value for all stakeholders. His broad definition of stakeholders includes all groups and individuals who can affect, or are affected by, the strategic outcomes of a company. The theory is built on the assumption of an efficient market. In this way, the company can be conceived as being at the heart of a structure of connected stakeholders who exchange all kinds of resources (e.g. products, knowledge and money) (Harrison, Bosse & Phillips, 2010). According to stakeholder theory, welfare-optimization is achieved when the needs of the most important stakeholders in this nexus are met. Rather than focusing on the requests and desires of one or merely a few stakeholders, this perspective encourages companies to operate in the interests of many diverse stakeholders, since these companies may enjoy higher levels of financial performance (Harrison, Bosse & Phillips, 2010).

Stakeholder theory can well be applied to CSR research; CSR can be envisioned as a form of managing for stakeholders. A study conducted by Godfrey, Merrill and Hansen (2009) indicates that companies that carry out institutional CSR activities enjoy ‘insurance-like’ beneﬁts. Put differently, because stakeholders attach positive attributions to those

(10)

10 activities, goodwill is created, which in turn mitigates possible negative judgments and sanctions towards the firm. Similarly, Orlitzky, Schmidt and Rynes (2003) adopt the stakeholder perspective to explain the CSP-CFP relationship. As well as a positive correlation of CSP with CFP across studies and industries, they demonstrate that a company’s reputation mediates this relationship. Research (Aguinis & Glavas, 2012; Orlitzky, Schmidt & Rynes, 2003) sums up several external benefits of CSP from the reputation-effects perspective; communication about the company’s CSP contributes to image-building with customers, investors, bankers and suppliers (Fombrun & Shanley, 1990); companies may attract better employees (Turban & Greening, 1997) or increase existing employees’ loyalty and work ethic, which will likely increase financial performance (Waddock & Graves, 1997); high reputational ratings may signal trustworthiness to bankers and investors and hence facilitate accessing finance (Spicer, 1978).

2.2 The insurance-like benefit of CSP

In their review of CSR literature based on 588 journal articles and 102 books and book chapters, Aguinis and Glavas (2012) report reduced risk as an outcome for companies undertaking CSR activities. The phenomenon of reduced risk as a result of CSR is known as the insurance-like benefit or reservoir of goodwill. Godfrey, Merrill and Hansen (2009) investigated the CSP-CFP relationship in the context of negative company publicity and hypothesized that moral capital is responsible for the emergence of such an insurance-like property of CSP. In particular, CSR activities signal that a company takes an ‘other-regarding’ stance rather than a ‘self-regarding’ disposition towards its network of stakeholders. As these stakeholders notice that the company engages in activities they consider desirable from a social or moral perspective, the company builds favourable attributions or moral capital, which in turn protects against or mitigates negative publicity. This perspective is consistent

(11)

11 with literature on corporate associations and consumer-company identification by respectively Ellen, Webb and Mohr (2006) and Einwiller et al. (2006). Empirical research provides evidence for the insurance-like benefit of CSP; i.e. socially responsible activities yield an insurance mechanism that preserves the company’s financial performance when the company suffers a negative event (Bansal & Clelland, 2004; Godfrey, Merrill & Hansen, 2009; Peloza, 2006). Another study on companies engaging in controversial business practices (e.g. tobacco, gambling and alcohol) even indicates that support for the risk-reduction hypothesis through CSR activities is more significant for these controversial industry companies than for non-controversial industry companies (Jo & Na, 2012).

Risk reduction as an outcome of CSR activities fits well with a relatively recent stream of theoretical literature that incorporates crisis management into the stakeholder perspective. Alpaslan, Green and Mitroff (2009) contribute to this stream of literature by presenting a stakeholder theory of crisis management. They argue that in the context of a crisis, managing for stakeholders will result in more successful outcomes to the company than would be the case under managing in accordance with shareholder theory. These outcomes may arise in the form of early detection of warning signals or fast recovery. The authors propose three arguments for their rationale. First, managing for stakeholders implies that managers will have ties to a more diverse group of stakeholders than merely shareholders, which enables them to recognize a crisis from different perspectives and allows for a more realistic notion of the company itself and its environment. Second, managers may draw upon the resources and information available of these stakeholders. Third, managers may benefit from noticing early warnings of different stakeholders and prepare for their consequences.

(12)

12 2.3 Systemic crisis situations

Scholars of the aforementioned studies on the insurance-like benefit of CSR and Alpaslan, Green and Mitroff (2009) as well, establish their arguments and test their hypotheses in the setting of negative, company-specific events, i.e. organizational crises. Pearson and Clair (1998) propose the following definition: “an organizational crisis is a low-probability, high-impact event that threatens the viability of the organization and is characterized by ambiguity of cause, effect and means of resolution, as well as by a belief that decisions must be made swiftly” (p. 61). This company-specific type of crisis is rather different than a systemic (i.e. global) crisis which affects the entire economy; an organizational crisis caused by the company itself is likely to have a negative impact on a company’s reputation, whereas a global economic deterioration will not necessarily depreciate reputation.

In contrast to the high number of studies testing the CSP-CFP relationship in the context of ‘business as usual’ or an organizational crisis, how CSP may impact CFP in systemic crisis situations has been investigated only very scarcely in the empirical literature. This omission is remarkable, since literature does describe how extensive and widespread the consequences of the global economic and financial crisis of 2007-2009 are; businesses were destroyed, GDP contracted, stakeholder relationships were unsettled and the unemployment rate raised significantly (Flammer & Ioannou, 2015; Keeley & Love, 2010; McKibbin & Stoeckel, 2010). Furthermore, literature documents that systemic crises often are accompanied by increased uncertainty and risk of company failure (Bhattacharjee et al., 2009; Bloom, 2014); radical changes in regulations and policy (Baker, Bloom & Davis, 2013; Pastor & Veronesi, 2012); higher cost of capital and restricted access to capital sources (Chodorow-Reich, 2014); a decline in consumer demand (Mian, Rao & Sufi, 2013), and disruptions in supply chains (Cattaneo, Gereffi & Staritz, 2010). In short, an economic downturn generates dramatic consequences for individuals, societies, companies and governmental institutions

(13)

13 globally. Keeley and Love (2010) claim that although the most destructive phase of the recent economic and financial crisis has passed, it will take several years for economic growth and unemployment rates to recover from this market shock. If a financial crisis can have such a devastating impact on the economy as a whole, one might expect it to affect the CSP-CFP relationship as well. Does the insurance-like benefit of CSP persist under systemic crisis conditions rather than under organizational crisis conditions?

Only a few studies focus on the potential impact of CSP or the construct of reputation for CSR on financial performance in systemic crisis situations. In their conceptual paper, Selvi, Wagner and Türel (2010) argue that CSR serves as a strategic management tool that helps to deal with systemic crisis situations, since it provides a means for for-profit companies to engage with policymakers and society at large to contribute to the common good. They even go as far as to claim that the recent economic and financial crisis has partially been caused by a lack of CSR.

The event study of Jones, Jones and Little (2000) tests for the buffer effect of reputation for CSR on the stock market. The results are mixed; for one event the insurance-like property seems to exist for companies with a reputation for CSR (i.e., these companies are affected significantly less severely by the market shock as compared to companies without a reputation for CSR), whereas for the other event of investigation – a more severe and unexpected downturn – there was no support for CSR reputation as a reservoir of goodwill. It is assumed that for the latter event rational decision making for investment was precluded by investor panic. Also Schnietz and Epstein (2005) conduct empirical research on the insurance-like benefit of social responsibility. They find support for the financial value of companies’ reputation for CSR by examining the investor reaction to the 1999 Seattle World Trade Organization (WTO) failure. Although this event did not cause a truly systemic (i.e. global) crisis, its effects on the stock prices of U.S. companies were significant. Relying on

(14)

14 transaction cost theory (Coase, 1960), the main explanation proposed for the results of their study is the social responsibility effect: “[f]irms that focus not only on explicit contractual claims (the easy-to-specify-contractual claims of investors or bondholders on firms), but also on implicit claims on non-investor stakeholders (such as the promise of continuing service or product quality to customers, or job security to employees) will realize higher market valuations than firms which ignore or discount the implicit claims of non-investor stakeholders on future revenues” (p. 331).

In a similar manner, Mio and Fasan (2012) empirically investigate the specific role of social performance in the context of Lehman Brothers’ bankruptcy. Mainly based on literature on stakeholder relations and the stakeholder theory of crisis management (Alpaslan, Green & Mitroff, 2009), the scholars propose three mechanisms that may have linked social and financial performance. The first mechanism – Implicit Claim Management and Regulatory Costs – refers to the company’s ability to manage implicit claims and to avoid imposition of regulatory costs due to the enforcement of laws from the government. The recent economic and financial crisis increased demand for transparency and ethic behaviors in business, since lack thereof has been pointed out among the main causes of the crisis. Mio and Fasan (2012) argue that Lehman Brothers’ bankruptcy may have led stakeholders to exercise implicit claims on companies, thereby causing financial distress, and that governments might be more inclined to impose new laws and regulations. They hypothesize that companies with better stakeholder relations are able to manage implicit claims more efficiently and that companies with higher levels of CSP bear less risk of increased regulatory cost (Baron and Diermeier, 2007; Blacconiere & Patten, 1994; Maxwell, Lyon & Hackett, 2000). Therefore, socially responsible companies are expected to achieve higher CFP. Second, the rationale for the Resource Availability and Withholding mechanism holds that companies managing in accordance with the stakeholder model rather than the shareholder model may enjoy more

(15)

15 favorable crisis management outcomes because of the availability of resources and information from stakeholders. Socially responsible companies – i.e. having better stakeholder relations (measured through CSP) – are likely to experience less resource withholding as compared to non-socially responsible companies, and are therefore expected to achieve higher CFP. Third, the mechanism of Corporate Reputation for CSP moderates the relationship between the other two mechanisms and expected financial performance. More specifically, a reputation for CSP may signal to investors the quality of stakeholder relations, which constitute the cornerstone for both the Implicit Claim Management and Regulatory Costs mechanism, and the Resource Availability and Withholding mechanism. The findings of the event study show that CSP generates a buffer effect in the context of a systemic crisis.

Finally, Ducassy (2013) examines whether CSP of a selection of French companies acts like an insurance-like protection in times of uncertainty, and takes for this purpose a more longitudinal approach. The results of the study demonstrate that CSP is positively associated with CFP for the first six months of economic turmoil (i.e. second-half of 2007), but there is no evidence for a significant connection between the two constructs in the period before the initiation of the financial crisis, nor in the period after these first six months. This implies that investors would only attribute importance to CSR-levels in times of economic uncertainty; the buffer role of CSR might be of temporal nature.

2.4 The CSP-CFP relationship and macro-economic effects in systemic crisis situations As the studies described above illustrate, research on the relationship between social and financial performance – particularly under systemic crisis conditions – is inconclusive. Also, the small-scale or short-term oriented research methodologies of the studies may preclude generalization of the results. Literature reviews in the CSP-CFP field document a lack of measurement uniformity across studies (Callan & Thomas, 2009; Margolis & Walsh, 2003).

(16)

16 Therefore, this study aims to test the CSP-CFP relationship in a systemic crisis situation in a more generalizable manner.

Whereas a loss of purchasing power for both businesses and consumers has been highlighted as a major consequence of a systemic crisis, academic empirical literature on the relationship between CSP and CFP does not seem to take this macro-economic construct into account. Examining the relationship in the context of a systemic crisis requires aggregation of the macro and organizational levels of analysis. I hypothesize that a decrease in purchasing power weakens or entirely eliminates the positive relationship between social and financial performance. More specifically, I argue that due to the loss of purchasing power as a consequence of a systemic crisis, companies cannot sustain a competitive advantage and/or protect their financial performance simply by sustaining their CSR expenditures. Although it is suggested that a company’s socially responsible reputation tends to decrease price sensitivity of consumers (Chernev & Blair, 2015), Eckhardt, Belk and Devinney (2010) find that economic rationality prevails among consumers; price is the single most important aspect in the consumer decision-making process. Assuming ‘business as usual’ conditions, the average price premium consumers are willing to pay for socially conscious products and services is 10 percent (Cotte & Trudel, 2010; De Pelsmacker, Driesen & Rayp, 2005). According to a study of Deloitte (2008) however, consumers will only buy the socially responsible products and services if quality, performance, and price are equal compared to regular products and services. Research thus clearly shows that human nature has strong preferences for cheapness. A loss of purchasing power is an additional stimulus to buy cheap and can likewise refrain consumers from taking into account a company’s CSR activities. Hence, considering both general consumer preferences and a loss of purchasing power, I assume that in the context of a systemic crisis consumers are not willing and/or able to reward companies for their CSR activities with price premiums or increased purchases; i.e. CSP

(17)

17 doesn’t provide an insurance-like benefit for financial performance anymore. CSP comes with a cost to companies and is in conflict with the company’s primary goal of profit maximization. Hence, under severe economic circumstances being socially or environmentally responsible does not yield any financial profits anymore, but instead has an effect of reducing financial performance. Consequently, I propose

Hypothesis 1a: Under normal economic circumstances, high-performing companies with respect to CSR activities enjoy a positive relationship between CSP and CFP, whereas low-performing companies do not.

Hypothesis 1b: In the context of a systemic crisis, the positive CSP-CFP relationship for high-performing companies with respect to CSR activities is significantly weaker than under normal economic circumstances. The difference between high- and low-performing companies in terms of strength of their CSP-CFP relationship is eliminated, implying that the insurance-like benefit of CSP for high-performing companies disappears.

Please note that besides an increase in revenues or insurance-like protection of financial performance, particular CSR activities might also provide a company with cost efficiencies. For instance, waste prevention is a form of CSP that is shown to be beneficial to companies in terms of financial profits (Hart & Dowell, 2010; King & Lenox, 2002; Marcus & Fremeth, 2009; Porter & Van der Linde, 1995; Quelch & Jocz, 2009), irrespective of economic circumstances and associated losses or increases in purchasing power.

2.5 Expenditures on CSR and systemic crisis situations

Although the relationship between CSP and CFP is expected to be weaker in the context of a systemic crisis, and despite the cost-cutting pressures companies experience in these circumstances, I hypothesize that companies do not adjust downwards their spending on CSR. More specifically, I argue that companies do not decrease their expenditures on CSR activities

(18)

18 to not risk losing their carefully built reputation – especially not in recessionary times. This is contrary to the belief of Selvi, Wagner and Türel (2010) that in times of economic turmoil, CSR cost cuts are inescapable. They argue that social responsibility is not a core business of the company and maintaining social responsibility programs therefore constitutes a threat to the company’s survival.

Literature provides evidence that CSP is linked closely to reputation, and reputation in turn is an important vehicle for financial profits (Hillenbrand & Money, 2007; Mio & Fasan, 2012; de Quevedo-Puente, de la Fuente-Sabaté & Delgrado-García, 2007). It might very well be the case that reputation deteriorates at a faster rate as a consequence of companies adjusting downwards their expenditures on CSR activities, than that it would increase as a result of companies undertaking additional CSR activities. A company that by reason of decreasing its CSR activities has fallen out of favour, may experience more than commensurate negative effects on its reputation. Supportive for this logic is the research finding that consumers show higher sensitivity to negative CSR information than to positive CSR information (Sen & Bhattacharya, 2001). Einwiller et al. (2006) confirm that negative publicity is likely to have a strongly negative impact on consumers’ formation of corporate reputation since negative information is given greater weight in the consumers’ mind than equally extreme positive information (Herr, Kardes & Kim, 1991; Klein & Dawar, 2004). Likewise, a negative reputation is likely to negatively influence the company’s financial performance and stock price. Moreover, the attention-getting nature of negative information induces negative word-of-mouth, which in turn hurts the company’s reputation even further. In addition to this, Fernández-Feijóo Souto (2009) argues that customer sensitivity with respect to negative information is much more likely to break out in crisis periods. Based on this logic I assert

(19)

19

Hypothesis 2: In the context of a systemic crisis, companies do not decrease their expenditures on CSR activities.

This section presented the theoretical argumentation and the formulated hypotheses on the expected CSP-CFP relationship and about the expected expenditures on CSR in the context of a systemic crisis. The next paragraph describes the empirical testing of the hypotheses and the data used for this purpose.

3 DATA AND METHODOLOGY

3.1 Sample construction

The data for this study come from several sources. The sample of companies qualifying for investigation has been drawn from the S&P500 stock market index. The index consists of relatively large, publicly traded companies in the U.S. Therefore, making use of this index ensures the sample’s representativeness of a larger population. The companies in the sample set are matched with the MSCI ESG KLD STATS DATA SET for social ratings; S&P500 firms that are not rated on social performance by this research and analytics institute are excluded from investigation. The period of investigation concerns the years 2001 up to and including 2013. This data range has been chosen because KLD social ratings are available only up till the year 2014. The period 2001-2013 includes the global economic and financial crisis of 2007-2009.

Financial data on the companies in the sample set complement the data on social responsibility. These data are obtained by use of Datastream. This database contains financial company-level information, company share prices and economic data on a macro-level of analysis. Unfortunately, data on purchasing power in itself are not available, but national income per capita rather accurately reflects changes in purchasing power. In particular, a

(20)

20 decrease in net income implies a proportionate decrease in purchasing power, provided that there is no deflation. However, since occurrence of deflation is rather exceptional, one can assume that adjusted national income growth is an appropriate measure to capture changes in purchasing power. The database of the World DataBank provides adjusted net national income per capita, expressed as annual percentage growth. Furthermore, to account for macro-level effects of systemic crises on the CSP-CFP relationship, data on U.S. unemployment rates during the period of interest are collected from the database of the Organization for Economic Co-operation and Development (OECD).

3.2 Dependent variable

This study tests for the potentially moderating effect of a systemic crisis on the relationship between CSP and CFP. The meta-analysis of Orlitzky, Schmidt and Rynes (2003) finds that CSP is more highly correlated with accounting-based measures of CFP than with market-based indicators. Following studies of other scholars (e.g. Aguinis & Glavas, 2012; Barnett & Salomon, 2012; Waddock & Graves, 1997) and considering that for long-term studies by use of regression analysis accounting-based performance measures are most appropriate (Ambec & Lanoie, 2008), the dependent variable in this study is the return on assets (RoA) for a given firm. RoA is defined as net income divided by the company’s total assets.

3.3 Independent variables

In accordance with common research practice in academic articles on social responsibility, KLD social ratings are used to measure the independent variable CSP (Barnett & Salomon, 2012; Callan & Thomas, 2009; Margolis & Walsh, 2003; Mio & Fasan, 2012). The independent MSCI ESG KLD STATS DATA SET offers a compilation of data on various dimensions of CSP. Information on each company is gathered by in-depth research and

(21)

21 analysis of CSR-related business practices. The institute subsequently assigns a score (a value of 1, zero or -1; respectively representing an area of strength, a neutral score and an area of weakness) to the various dimensions of CSP. Companies are rated on 13 dimensions of social responsibility, seven of which are key stakeholder attributes and the remaining six concern participation in controversial business activities. The variable KLDscore for a given company per year is found by aggregating the company’s scores on the seven dimensions concerning key stakeholder attributes.

Subsequently, following some other scholars (Ducassy, 2013; Jones, Jones & Little, 2000; Schnietz & Epstein, 2005), companies are classified as either ‘high’ or ‘low’ performers on CSR. Out of all companies contained in the sample set, the worst performing quartile with respect to average KLD score over the entire period of investigation (2001-2013) is classified as low-performing company, whereas the best performing quartile is classified as high-performing company. Hence, the companies that show moderate performance on CSR (half of the companies in the data set) are excluded from the regression analysis, since the hypothesized effects probably merely occur for companies that perform either very low or very high rather than moderate. A second distinction is made between the period prior to the recent economic and financial crisis, and the period that captures the crisis and its aftermath. Since multiple events have happened that could mark the dawn of the crisis, the exact start date is hard to identify. However, a significant contraction of the national income growth rate and a significant increase of the unemployment rate is visible from January 2007. Therefore, January 2007 marks the start of the crisis for the purpose of this research.

In order to capture the macro-level effect of a systemic crisis on the companies’ financial performance, the growth rate of national income per capita (labelled: Income) is included as an independent variable. The World DataBank provides this measure as a yearly percentage growth rate of net national income per capita compared to the previous period, and

(22)

22 seasonally adjusted. Additionally, since I assume that the consequences of a systemic crisis – a loss of purchasing power is among the most prominent ones – moderate the relationship between CSP and CFP, the interaction variable 𝐾𝐿𝐷𝑠𝑐𝑜𝑟𝑒 × 𝐼𝑛𝑐𝑜𝑚𝑒 has been constructed. The coefficient of this variable indicates to what extent macro-economic circumstances moderate the effect of CSP on a company’s financial performance.

3.4 Control variables

A company’s financial performance is not affected merely by its social performance and purchasing power of consumers and businesses. Leaving out factors that have been identified to be (potentially) of influence on CFP would introduce bias into the coefficients of the aforementioned independent variables. In order to prevent from omitted variable bias, this study includes the variables specified below to control for factors that could systematically affect financial performance.

Aguinis and Glavas (2012) ascertain that many scholars found company Size to be positively related to a company’s financial performance. The number of employees is often used as a measure of company size (Barnett & Salomon, 2012; Martin, Swaminathan & Mitchell, 1998; Nachum & Zaheer, 2005). Also, company Leverage is computed as the ratio of the company’s long term debt divided by total assets. Following prior research in this academic field (Aguinis & Glavas, 2012; Barnett & Salomon, 2012; Callan & Thomas, 2009; King & Lenox, 2002; McWilliams & Siegel, 2001), measures of R&D intangibles and capital investments are included. R&D intensity is defined by dividing expenditures on research and development by total sales, and Capital intensity is expressed as the ratio of capital expenditures to total sales (both variables expressed as a percentage). Lastly, the

Unemployment rate (expressed as an annual growth percentage) as provided by the

(23)

23 consequence from a systemic crisis, and job loss implies a decrease of national income per capita.

3.5 Data collection

Data on the variables mentioned above has been collected for S&P500 companies in the period 2001-2013. For 183 out of 504 companies, the MSCI ESG KLD STATS DATA SET provides only incomplete data sets of KLD social ratings for the period of investigation, or no data at all. These companies are excluded from the dataset. Subsequently, in order to compute each company’s net KLD score, the ratings for strengths and concerns of the seven dimensions of CSP (environment, community, employee relations, diversity, product, corporate governance and human rights) have been aggregated.

Datastream also reports some missing values for the variables requested. Archival research on annual reports of the respective companies has been conducted to complement the data for as many companies as possible. Nevertheless, 30 more companies are excluded from investigation due to missing values for one or multiple variables (whether dependent, independent and/or control variables). This process of data cleaning yields a final dataset containing 291 companies from the S&P500 (Appendix Table 1). After classification of these companies as either high-, low- or moderate performer with respect to their CSR-levels, regression analysis is performed for 75 high-performing companies and 74 low-performing companies. Hypothesis 2 however, is tested for the entire sample set (thus including the companies performing moderately on CSR).

3.6 Statistical method

The statistical analysis aims to determine the effects of the variables KLDscore and Income on financial performance, and the effect of the interaction variable containing these two

(24)

24 components. CFP as represented by RoA is specified as a linear function of the independent variables, the interaction variable and control variables for company 𝑖 at time 𝑡 of the sample set, with 𝜀𝑖𝑡 being the residual term:

𝑅𝑜𝐴_𝑖𝑡 = 𝛽₀+ 𝛽₁𝐾𝐿𝐷𝑠𝑐𝑜𝑟𝑒_𝑖𝑡+ 𝛽₂𝐼𝑛𝑐𝑜𝑚𝑒_𝑡+ 𝛽₃𝐾𝐿𝐷𝑠𝑐𝑜𝑟𝑒_𝑖𝑡× 𝐼𝑛𝑐𝑜𝑚𝑒 + 𝛽₄𝑆𝑖𝑧𝑒_𝑖𝑡 + 𝛽₅𝐿𝑒𝑣𝑒𝑟𝑎𝑔𝑒_𝑖𝑡+ 𝛽₆𝑅&𝐷 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦_𝑖𝑡+ 𝛽₇𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝑖𝑛𝑡𝑒𝑛𝑠𝑖𝑡𝑦_𝑖𝑡 + 𝛽8𝑈𝑛𝑒𝑚𝑝𝑙𝑜𝑦𝑚𝑒𝑛𝑡𝑡+ 𝜀𝑖𝑡

An Ordinary Least Squares (OLS) regression – appropriate for estimating the unknown parameters in a linear regression model – is performed to determine the significance of the variables included in the model in the two periods and according to the classification of a company as a high- or a low-performer with respect to CSR. Therefore, four regression analyses are performed. The software package STATA is the statistical tool used for this purpose.

In an auxiliary analysis, a paired t-test is performed on the companies’ average KLD scores in two different periods (2001-2006 and 2007-2013) to determine whether companies sustained or adjusted their level of social performance during the economic and financial crisis. The test statistic is calculated as follows:

t = XD − μ0 SD / √n

In the above equation, X_D and S_D represent the mean and standard deviation of the differences between each company’s average KLD scores in the two periods. μ₀ is a constant that takes a non-zero value when testing whether the average of the difference is significantly different from μ0, otherwise it will be zero. The number of observations is given by 𝑛, and (𝑛 − 1) represents the degrees of freedom. Since each company is used as its own control –

(25)

25 thereby eliminating random between-company variation – it is assumed that the variances of the pairs are equal.

3 RESULTS

Tables 2 and 3 below present respectively descriptive statistics for all variables used in this study and the correlations between these variables.

First, four OLS regressions have been performed in order to investigate whether the formulated hypotheses are true. Results of the analyses are reported in Table 4 below. A first key observation in reviewing the regression analysis is that findings are rather unexpected. Despite the fact that the adjusted R2’s are relatively low, conclusions can still be drawn about how changes in the independent variables are associated with changes in the dependent

(26)

26 variable (CFP as represented by RoA) since the majority of the predictors is statistically significant.

The regression coefficient of the variable KLDscore for low-performing companies in the period before the crisis is negative and significant (𝛽 = −0.6957, 𝑝 < 0.001), implying that these companies obtain negative financial returns from CSR activities. The coefficient of the interaction variable Kld_income is significantly positive (𝛽 = 0.5037, 𝑝 < 0.001). This implies that a decrease in net national income per capita generates an even lower financial return from CSR, whereas a significant increase will yield profits resulting from CSR activities. Surprisingly, the coefficient of KLDscore for high-performing companies in this period (2001-2006) is slightly negative (𝛽 = −0.0229, 𝑝 = 0.767), but not significant. In contrast, the coefficient of the interaction variable is positive and significant (𝛽 = 0.2247, 𝑝 < 0.001). The results indicate that financial returns from CSR decrease (increase) for decreasing (increasing) net income per capita.

In the period that captures the crisis and its aftermath (2007-2013), the KLDscore coefficient for low-performing companies is negative and insignificant (𝛽 = −0.0269, 𝑝 = 0.209), and also the coefficient of the interaction variable is insignificant (𝛽 = 0.0014, 𝑝 = 0.918); in times of economic turmoil, CSR activities are not associated with CFP for low-performing companies. Lastly, for high-low-performing companies under systemic crisis conditions, the relationship between CSP and CFP is significantly positive (𝛽 = 0.2895, 𝑝 < 0.001), but the coefficient of the interaction between KLDscore and Income is insignificant (𝛽 = −0.0026, 𝑝 = 0.812). Hence, the benefits associated with high CSR performance outweigh the additional costs incurred for doing so.

Except for Size, and in the regression for high-performing companies in 2007-2013 also Leverage, all control variables included are statistically significant in all four regression analyses.

(27)

27 Subsequently, the results of the paired t-test in Table 5a show that the overall average KLD score in the period 2007-2013 is significantly higher than in the period 2001-2006 (𝑡(290) = 16.83, 𝑝 < .001 – both for a two-sided test and a one-sided positive test). However, this effect might be due to increased public pressure on companies to engage in CSR and to publish social responsibility reports starting from the early 2000’s (Callan & Thomas, 2009; Eccles, Ioannou & Serafeim, 2014; Schnietz & Epstein, 2005). Therefore, a second paired t-test (Table 5b) has been performed to compare the average KLD score in the period 2005-2006 (a subset of the original time frame before the start of the crisis) with the average of the period 2007-2013. This yields the same result as the previous test; on average, companies significantly increased their level of social performance (𝑡(290) = 17.29, 𝑝 < .001 – again, both for a two-sided test and a one-sided positive test).

(28)

28

4 ANALYSIS

4.1 Interpreting the regression results

The empirical study generates results that are not consistent with the theoretical argumentation for conducting this research. In brief, engaging in socially responsible activities seems to provide insurance-like benefits only to high performing companies in systemic crisis situations.

The negative or neutral association between CSP and CFP for companies that perform poorly on CSR activities is not striking. It indicates that simply engaging in some socially responsible activities is not enough in order to generate financial returns from CSR. Researchers find that congruence of the company with the cause is important, just as the level of commitment to the cause (Cornwell, Weeks & Roy, 2005; Ellen, Webb & Mohr, 2006; Webb & Mohr, 1998). Hence, for companies that perform relatively low on CSR, the financial benefits generated by their CSR practices do not offset the costs incurred for these practices, or at most amount to the costs incurred.

A noteworthy observation however, is the order in which the negative and neutral association between CSP and CFP for low-performing companies occur. The coefficient of

(29)

29 period 2007-2013 it is not significantly different from zero; i.e., the CSP-CFP relationship is negative prior to the crisis and non-existent during the crisis and its aftermath. This effect might be due to the increased pressure of stakeholders on companies to engage in socially responsible behaviour. Specifically, companies that engaged in CSR to a lesser extent in the period prior to the crisis are likely to have felt this pressure more severely than moderately- or high-performing companies. Possibly, these companies increased their socially responsible investments to such a large extent that they outperform the initial moderately- or high-performing companies. However, it has not been investigated whether companies that are classified as low-performers in the period prior to the crisis are also performing poorly during the crisis and its aftermath or whether they have improved their performance rather significantly and need to be reclassified as moderate- or possibly even high-performers in the period 2007-2013. Instead, the classification remained unaltered throughout the entire period of investigation (2001-2013).

Even more surprising is the finding that the coefficient of KLDscore for high-performing companies under systemic crisis conditions is significantly positive. This may be explained by the logic that CSR activities not only could generate profit, but they could also reduce costs (Ambec & Lanoie, 2008; Hart & Dowell, 2010; Marcus & Fremeth, 2009; Porter & Van der Linde, 1995). High-performing companies with respect to CSR may be better able to strategically align their social responsibility programs with core business activities in order to simultaneously enjoy cost reductions and higher profits. For the fact that systemic crisis conditions significantly increase the pressure to reduce costs, companies may be incentivized to emphasize both aspects even more, and this yields the greatest returns for companies performing well on CSR.

In sum, the research findings largely contradict the formulated hypotheses on the relationship between CSP and CFP. Contrary to hypothesized, the growth rate of net national

(30)

30 income per capita does not seem to have a moderating effect on the CSP-CFP relationship under systemic crisis conditions. Moreover, it has been hypothesized that the CSP-CFP relationship for high-performing companies with respect to CSR would be significantly positive prior to the recent economic and financial crisis, and that this relationship would be weaker during the crisis and its aftermath. However, this study indicates that there is no relationship between CSP and CFP for high-performing companies prior to the crisis, and that this relationship turns positive during the crisis and its aftermath. These results are in accordance with the findings of Ducassy (2013). Her study demonstrates a significant positive association between CSP and CFP for the period that captures the financial and economic crisis, but does not show a significant connection between the two variables prior to the crisis. This study provides additional support for the finding that the insurance-like benefit of CSR to high-performing companies is of temporal nature and that importance is attributed to CSR-levels in recessionary times only. Additionally, it has been hypothesized that low-performing companies would not enjoy financial profits from CSR under any circumstances – neither prior to the crisis, nor during the crisis and its aftermath. The findings of this study are supportive for this hypothesis and consistent with stakeholder theory. They validate empirical work that has been conducted on the CSP-CFP relationship with regard to low-performing companies.

4.2 Comparing CSR expenditures prior and during the crisis

In times of crisis, the need for companies to minimize their costs is straightforward. Selvi, Wagner and Türel (2010) argue that expenditures on activities that do not constitute the core business of the company are the first to be reduced. For that reason, commitment to social responsibility under severe and uncertain economic circumstances is a threat for the company’s survival; i.e. CSR costs cuts are inevitable. However, consistent with hypothesis 2,

(31)

31 research findings of this study indicate that companies increased their level of CSR activities during the financial and economic crisis and its aftermath, compared to the period prior to the crisis. This is in line with a study recently conducted by Flammer and Ioannou (2015). They test how companies adapt their investment strategies for key strategic resources in times of economic recession. The authors hypothesize that a focus on intangible resources enables companies to “i) become more efficient and innovative, ii) adapt more easily to shifting needs and demands of suppliers, consumers, and other stakeholders, and iii) enhance their organizational resilience” (p. 5). They find that “[companies] did not decrease their R&D and CSR investments, despite the cost-cutting pressures, heightened uncertainty, and other challenges inherent to periods of recession’ (p. 7), whereas employees are laid off and capital expenditures are reduced significantly (Berns et al., 2009; Delevingne, 2009; Jaruzelski & Dehoff, 2009; Scheck & Glader, 2009). Also Whiteman, Walker and Perego (2013) indicate that companies’ investments in CSR initiatives remain significant despite tough economic circumstances, “illustrating that sustainability is not simply an ‘add on’ when times are good” (p. 308). Moreover, Flammer and Ioannou (2015) document that companies that sustain their investments in R&D and CSR achieve higher performances once the economy recovers compared to companies that reduce investments in these intangible resources. The findings provide evidence that innovation capabilities and stakeholder relationships are instrumental in sustaining a competitive advantage during and beyond times of economic turmoil.

Although it may seem paradoxically, research findings of this study provide evidence that companies increase their expenditures on social responsibility under severe economic circumstances. There are three different rationales for this phenomenon. First, as hypothesized in paragraph 2.5, companies may refrain from decreasing their CSR expenditures for the sake of reputation issues. Reputation may deteriorate more than commensurate as a consequence of negative publicity, in turn affecting companies’ financial profits. Second, from this study, it

(32)

32 turns out that high-performing companies with respect to their CSR activities enjoy a positive relationship between CSP and CFP in recessionary times. This could be linked to the research finding that companies that sustain their CSR activities outperform companies that do not once the economy recovers (Flammer & Ioannou, 2015). Also, there is considerable evidence that consumers will demand an ‘unsustainability’ discount for regular products and services (Cotte & Trudel, 2010). This could have a stronger negative impact on financial profits than sustaining investments in social responsibility. Lastly, in contrast to the above two arguments for which financial performance is the main reason for sustaining CSR levels, maintaining investments in CSR activities may as well derive from personal beliefs. Several CEO’s and board members of leading companies declared that investing strategically in people and environment is more compelling than ever during economic downturns (Flammer & Ioannou, 2015). Selvi, Wagner and Türel (2010) share this vision and argue that CSR is a means to engage with policy makers and members of society in order to contribute to the common good, and that this means must be exploited in particular in recessionary times. The arguments to invest in stakeholder relations attest to an intrinsic motivation for sustaining or increasing CSR activities.

5 CONCLUSION

This study aims to provide more insight into the relationship between CSP and CFP in the context of a systemic crisis. More specifically, does CSP yield an insurance-like benefit so that financial profits are preserved in times of an economic downturn? I theoretically argue that the negative consequences of a systemic crisis – among which a loss of purchasing power which is reflected in the contraction of the growth rate of national income per capita – have a moderating effect that makes the CSP-CFP relationship weaker.

(33)

33 A linear regression technique is used to investigate this question empirically. Using the national income per capita growth rate to account for the macro-economic effects of the financial and economic crisis of 2007-2009, and distinguishing both between systemic crisis conditions and normal economic circumstances, and between high- and low-performing companies with respect to CSR activities, this study indicates under which conditions an insurance-like benefit operates. A noteworthy key finding of this study is that there is no positive association between CSP and CFP, except for high-performers on social responsibility during the crisis and its aftermath. This result suggests that only in times of systemic crisis importance is attributed to CSR activities and that the insurance-like benefit is of temporal nature. Second, the national income growth rate as a measure to reflect disruptive changes on a macro-economic level does not seem to be of any moderating influence under systemic crisis conditions. Furthermore, in an auxiliary analysis, it is demonstrated that on average companies increased their investments in social responsibility in the period 2007-2013 compared to the period prior to the crisis.

5.1 Contributions

This study makes at least three contributions to literature on the CSP-CFP relationship. First, in a response to a call for a multilevel theoretical approach to CSR (Aguinis & Glavas, 2012), this study brings together two currently disparate fields; theory of the relationship between CSP and CFP on an organizational level of analysis, and theory of systemic crises (macro-economic level). Other researchers as well acknowledge that it is astonishing how sparse literature is with regard to the impact of economic crises on decision-making on an organizational level and on competitive strategies taking into account the severity and frequency with which major market shocks occur (Agarwal et al., 2009; Flammer & Ioannou, 2015; Garcia-Sanchez, Mesquita & Vassolo, 2014). This study explores the influence of

(34)

34 macro-economic changes on the CSP-CFP relationship on an organizational level, and contributes to bridging the micro-macro divide in doing so.

Second, the finding that high-performing companies with respect to CSR do not enjoy financial profits from their socially responsible activities under normal economic circumstances, raises questions about the existence of a direct positive association between CSP and CFP. Future research is needed to gain more insights in this relationship and/or in the circumstances under which the relationship seems to exist.

Lastly, on a managerial level the findings of this study contribute to an enhanced understanding of the financial value of CSR as an intangible asset under systemic crisis conditions. Provided with this knowledge, managers can more deliberately adjust their investments in social responsibility in times of an economic downturn.

5.2 Limitations and suggestions for future research

Unfortunately, this study does not investigate the reasons for the remarkable finding that for both high- and low-performing companies, the CSP-CFP relationship has improved during the crisis and its aftermath compared to the period prior to the crisis. It would be highly interesting to find out the reasoning for why this effect occurs. However, this also implies that it should be questioned whether the positive relationship really exists. Empirically testing the relationship or conditions under which social and financial performance go together appears a difficult task since there is no consensus on research methodologies (Callan & Thomas, 2009; Margolis & Walsh, 2003)

Furthermore, the use of the net national income per capita growth rate for the purpose of this study could be contested. In particular, the income growth rate shows a contraction only for the period 2007-2009, whereas it is assumed that consequences of such a devastating

(35)

35 meltdown have a longer-term effect (Keeley & Love, 2010). The national income growth rate does not seem to capture this longer-term effect.

Although it has been suggested that there is great overlap between the concepts of corporate reputation and CSR (Hillenbrand & Money, 2007), it would be an insightful contribution to investigate the impact of CSP on CFP by use of consumer perceptions of CSR rather than through CSR scores retrieved from an objective database. Moreover, scholars point out that the U.S and Europe have very different institutional and regulatory environments (King & Lenox, 2002; Matten & Moon, 2008). For example, socially responsible practices are much more implicitly regulated in Europe compared to the U.S. This difference could have implications for the CSP-CFP relationship. Finally, this study calls for more future research in other systemic crisis contexts in order to achieve a generally accepted consensus on the role of social responsibility in recessionary times.

ACKNOWLEDGEMENTS

I am grateful to Pushpika Vishwanathan for her valuable and insightful feedback on drafts of this study.

(36)

36

REFERENCES

Agarwal, R., Barney, J. B., Foss, N. J., & Klein, P. G. (2009). Heterogeneous Resources and the Financial Crisis: Implications of Strategic Management Theory. Strategic

Organization, Forthcoming.

Aguinis, H., & Glavas, A. (2012). What We Know and Don’t Know About Corporate Social Responsibility: A Review and Research Agenda. Journal of management, 38(4), 932-968.

Alpaslan, C. M., Green, S. E., & Mitroff, I. I. (2009). Corporate Governance in the Context of Crises: Towards a Stakeholder Theory of Crisis Management. Journal of

Contingencies and Crisis Management, 17(1), 38-49.

Ambec, S., & Lanoie, P. (2008). Does It Pay to Be Green? A Systematic Overview. Academy

of Management Perspectives 22(4), 45-62.

Auger, P., Burke, P., Devinney, T. M., & Louviere, J. J. (2003). What Will Consumers Pay for Social Product Features?. Journal of Business Ethics, 42(3), 281-304.

Baker, S. R., Bloom, N., & Davis, S. J. (2015). Measuring Economic Policy Uncertainty (No. w21633). National Bureau of Economic Research.

Banerjee, S. B. (2008). Corporate Social Responsibility: The Good, the Bad and the Ugly.

Critical sociology, 34(1), 51-79.

Bansal, P., & Clelland, I. (2004). Talking trash: Legitimacy, impression management, and unsystematic risk in the context of the natural environment. Academy of Management

Journal, 47(1), 93-103.

Barnett, M. L. (2007). Stakeholder Influence Capacity and the Variability of Financial Returns to Corporate Social Responsibility. Academy of Management Review, 32(3), 794-816.

(37)

37 Barnett, M. L., & Salomon, R. M. (2012). Does it pay to be really good? Addressing the shape of the relationship between social and financial performance. Strategic Management

Journal, 33(11), 1304-1320.

Baron, D. P., & Diermeier, D. (2007). Strategic Activism and Nonmarket Strategy. Journal of

Economics & Management Strategy, 16(3), 599-634.

Berns, M., Townend, A., Khayat, Z., Balagopal, B., Reeves, M., Hopkins, M., & Kruschwitz, N. (2009). The Business of Sustainability: Imperatives, Advantages, and Actions.

Boston, MA: The Boston Consulting Group.

Bhattacharjee, A., Higson, C., Holly, S., & Kattuman, P. (2009). Macroeconomic Instability and Corporate Failure: The Role of the Legal System. Review of Law and Economics

5(1), 1-32.

Blacconiere, W. G., & Patten, D. M. (1994). Environmental disclosures, regulatory costs, and changes in firm value. Journal of Accounting and Economics, 18(3), 357-377.

Bloom, N. (2014). Fluctuations in Uncertainty. Journal of Economic Perspectives 28(2), 153-176.

Brammer, S. J., & Pavelin, S. (2006). Corporate Reputation and Social Performance: The Importance of Fit. Journal of Management Studies, 43(3), 435-455.

Callan, S. J., & Thomas, J. M. (2009). Corporate Financial Performance and Corporate Social Performance: An Update and Reinvestigation. Corporate Social Responsibility and

Environmental Management, 16(2), 61-78.

Cattaneo, O., Gereffi, G., & Staritz, C. (2010). Global Value Chains in a Postcrisis World: A

Development Perspective (The World Bank, Washington, DC).

Chernev, A., & Blair, S. (2015). Doing Well by Doing Good: The Benevolent Halo of Corporate Social Responsibility. Journal of Consumer Research, 41(6), 1412-1425.

(38)

38 Chodorow-Reich, G. (2014). The Employment Effects of Credit Market Disruptions:

Firm-level Evidence from the 2008–9 Financial Crisis. The Quarterly Journal of

Economics, 129(1), 1-59.

Coase, R. (1960). The Problem of Social Cost. Journal of Law and Economics, 3(4), 1-44. Cornwell, T. B., Weeks, C. S., & Roy, D. P. (2005). Sponsorship-Linked Marketing: Opening

the Black Box. Journal of advertising, 34(2), 21-42.

Cotte, J., & Trudel, R. (2010). Socially conscious consumerism: A Systematic Review of the Body of Knowledge. Network for Business Sustainability.

Delevingne, L. (2009). Surprising survivors: Corporate do-gooders. Fortune, January, 20. Deloitte Consulting LLP (2008). Finding the green in today's shoppers – Sustainability trends

and new shopper insights. The Association of Food, Beverage and Consumer Products

Categories.

De Pelsmacker, P., Driesen, L., & Rayp, G. (2005). Do Consumers Care about Ethics? Willingness to pay for Fair‐Trade Coffee. Journal of Consumer Affairs, 39(2), 363-385.

Ducassy, I. (2013). Does Corporate Social Responsibility Pay Off in Times of Crisis? An Alternate Perspective on the Relationship between Financial and Corporate Social Performance. Corporate Social Responsibility and Environmental Management, 20(3), 157-167.

Eccles, R. G., Ioannou, I., & Serafeim, G. (2014). The Impact of Corporate Sustainability on Organizational Processes and Performance. Management Science, 60(11), 2835-2857. Eckhardt, G. M., Belk, R., & Devinney, T. M. (2010). Why don't consumers consume

Exploring the insurance-like benefit of corporate social responsibility in the context of a systemic crisis