MSc Business Administration (Strategy track) Master Thesis
Study into the relationship between Downstream Sustainability, CSR considerations in relation to the IR framework and Corporate Financial Performance.
Name: Philipp Havel Student ID: 10826173
Supervisor: Sebastian Kortmann
Statement of originality
This document is written by Student Philipp Havel, 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.
Table of Contents:
Abstract 3
Introduction 4
Literature review 9
Hypothesis development 21
Data and Method 26
Discussion 37
Conclusion 43
Limitations 44
Abstract:
The aim of this study is to investigate the effect of several sustainability-related variables on Corporate Financial Performance (CFP). The core of the sustainability-related considerations is made up of several downstream sustainability variables (Marketing/ Contracting, Product Safety, Antitrust and Others), as well as a variable that relates to the extend to which CSR practices of the firm are globally integrated or locally responsive. The EBITDA margin represents the dependant variable, Corporate Financial Performance. The standard multiple regression analysis was chosen to analyse the relationship between the variables. The data that was used for this study was obtained for a sample of 124 companies over a time period of 5 years. The analysis of the panel data was further complemented by controlling for this time factor. Overall, the findings for the downstream sustainability variables and their relationship to CFP are mixed. On the other hand, the results do show that having globally integrated CSR policies is positively related to Corporate Financial Performance. This gives a potential indication for why firms would favour globally integrated over locally responsive CSR policies. The results obtained in this study are useful and important starting point for investigating CSR policies from a downstream as well as International Business point-of-view.
Introduction:
The topic for this thesis has been based on a research gap that was identified through extensively reading into the topic of the Integration - Responsiveness framework and linking this to the Downstream Sustainability Performance of MNEs’. The IR trade-off and sustainability choices will also be investigated in order to see if there are differences in this framework depending on the economic development level of the host country. In terms of the scope of the research, it will focus on company’s downstream value chain activities, such as Marketing strategies, Product Safety Concerns, CRM, different forms of customer collaboration etc. The defined sustainability performance that will be investigated is going to be specified according to the three-pillar model, which includes social, environmental and economic factors.
Corporate Social Responsibility involves managing a firm in such a way that it can be ‘economically profitable, law abiding, ethical and socially supportive’ (Carroll, 2010), something which is complicated when operating in a large number of different contexts with often diverging views of the role of business in society (Kolk & Tulder, 2010). Kolk & Tulder (2010) also argue that the impact of MNE’s in sustainable development is largely unclear and needs further investigation. Nevertheless, MNE’s CSR activities are seen as becoming increasingly strategic, in the sense that they affect the core business of the firm and its growth, profitability and survival (Kolk & Pinske, 2008), with CSR as a potential source of competitive advantage (Porter & Kramer, 2006).
The question is based on a research gap that is further specified in the literature review section of this document, namely being the lack of research into links between International Business (in this case the I-R framework) and downstream sustainability related theories. In today’s economic environment, where efficient global supply chains are often a deciding factor in a firm’s competitive nature, a wide range of stakeholders expect firm’s to
have globally reaching sustainability policies. It is only a matter of time until sustainability issues, both downstream and upstream, get incorporated into mainstream IB theory.
There are several reasons for why the Integration-Responsiveness framework and sustainability related issues are important areas for research that are worth combining into one stream of research. Performance in terms of sustainability related metrics, often called Corporate Sustainability Performance (CSP) in the literature, is becoming increasingly important. Especially internationally operating firms that are exposed to a wide set of stakeholders have to ensure that their performance in terms of CSP is within the expectations of all stakeholders. Managing the Global CSR policies of a firm is therefore more important than ever. Appropriately managing the firms CSR policies on a global scale is especially important because wrongdoings in one single location or market can have severe impacts on the perception of the organization on a global level. Appropriately managing the firms CSR policies is therefore of crucial importance. When looking specifically at the I-R spectrum of CSR policies, the question of whether a locally responsive, globally integrated, or a mix of both creates the greatest value for a firm, is therefore very interesting. Combining these two fields is therefore not only an important step in investigating the effectiveness of different kinds of CSR policies on an international scale, but can also potentially lead to interesting insight about the perceptions of firms when it comes to the effectiveness of global CSR policies.
Making the distinction between upstream and downstream sustainability is especially applicable, especially as there is already a wide range of research papers written on upstream sustainability issues. Downstream sustainability issues, especially in relation to IB topics like the I-R framework, have not been discussed much in recent literature. Part of the reason of the literature’s focus on upstream rather than downstream sustainability issues could well be that upstream issues, such as pollution and waste etc., have much more noticeable effects on
a firm’s performance. Furthermore, upstream issues also create much more pressing and dangerous situations, which is partially why stakeholders are much more likely to be actively concerned about these issues. The neglect of downstream issues becomes very obvious when one is searching current literature, which was one of my motivations for writing this thesis. Most importantly though, downstream CSR issues have a much more direct impact on the consumer than upstream issues. For example, Marketing/ Contracting and Product Safety issues are likely to affect the consumer much more directly than upstream concerns such as high pollution levels at a factory. It could be argued that downstream sustainability issues are therefore more concerned with the social and economic aspects of the triple bottom line, while upstream issues are related to all three parts. Even though downstream sustainability considerations affect a much smaller group of stakeholders than upstream considerations, researching into these issues and their effect on company performance can provide valuable insights nevertheless.
The two issues that can arguably affect the consumer the most and also lead to the gravest ramifications for a company are marketing/ contracting concerns and product safety concerns. These two areas should therefore be of great importance for companies that are trying to establish themselves as sustainable. From a research point of view, being able to determine the motivating factors for firms adopting sustainable downstream practises is very important. The biggest motivating factor for the average firm is likely to be revenue/ profits. If adopting sustainable downstream measures, such as minimizing marketing/ contracting and product safety concerns, can help generate more revenue/ profits, many firms are going to be likely to adopt these kinds of measures. Hence, finding out what effect downstream sustainability concerns have on corporate financial performance can help in establishing clear causality between the different factors.
The integration of the two main constructs of this thesis, the I-R framework as well as Corporate Downstream Sustainability, is only a small first step in investigating the relationship between these two variables and CFP. Especially when considering the global extend of many firms value chains, it is important to understand the different factors that play into firms CSR considerations. As is discussed thoroughly in the limitations section, the data collection procedure for the I-R variable is not optimal. In most I-R related research papers, data that was collected on the Global Integration/ Local Responsiveness of companies was collected using survey/ interview methods. Unfortunately, due to time constraints the data in this thesis was collected by analysing the companies CSR reports. This thesis is therefore to be seen as a first step into investigating the effects between the different variables rather than a ‘perfect’ study. Kolk and Tulder (2010) argue that one of the reasons for scant attention may be the problem of linking CSR/ sustainable development to mainstream IB debates. It is essential to keep in mind the obstacles to combining the two fields, which are further discussed in the limitations section, when reading this thesis.
When making investment decisions, sustainability indices are becoming increasingly important in signalling sustainability conscious investors companies that perform exceptionally well in this regard. The Dow Jones Sustainability Index is one of the leading indices that ranks companies based on their Corporate Social Performance (CSP). There have already been several researchers that have investigated the effect of the inclusion in the DJSI on Corporate Financial Performance (CFP). Nevertheless, the DJSI membership variable can still give some insights into the companies included in this sample, and will also be used to see if the results obtained previously by various researches can be replicated.
The approach adopted here therefore disaggregates the components of Downstream Sustainability considerations as well as IB related sustainability considerations, enabling the researcher to make a more refined analysis on the individual variables and their contribution
to organizational performance. Thanks to being able to judge the importance of each individual factor, important insights into motivating factors for sustainability measures that are important to companies as well as consumers can be reached.
Literature review:
This section is structured in a way where all variables and frameworks used in this thesis are discussed separately at first, with the connection between the separate frameworks done later on in this study.
Arguably, the IR framework is the most popular framework for studying for studying international strategy in multinational firms (Devinney et al., 2000). As described by Devinney et al. (2000), the use of the IR framework in global strategy was initially proposed by Prahalad (1975) and subsequently developed and applied by a number of authors, including Doz et al. (1981) and Bartlett and Ghoshal (1989). The IR framework describes the trade-off that firms face with their international strategies. This trade-off is present in the way that MNEs’ face certain pressures for global integration as well as pressures for local responsiveness. Integration pressures include importance of multinational customers, the presence of multinational competitors, investment intensity, technology intensity, pressures for cost reduction, universal needs and access to raw materials and energy, while local responsiveness pressures include differences in customer needs and differences in distribution channels, availability of substitutes and the need to adapt, market structure and host government demands (Devinney et al., 2000). In order to develop successful international strategies MNEs’ need to find a balance between these two pressures. As becomes evident from the definition of the IR framework’s two forces, Integration forces are of an economic/ technological nature, while Responsiveness forces are of a market/ customer nature (Devinney et al., 2000). The objectives also differ to some extend, as Integration looks for efficiency, with the possible source of advantage being cost and productivity, while Responsiveness looks for customer satisfaction, with their source of advantage coming from differentiation and customer fit (Devinney et al. 2000). One can see a link between the IR framework and Porter’s generic strategies of differentiation and cost-leadership. The
relationship between the two strategies though is different than Porter predicts. Tian & Slocum (2014) for example show that subsidiaries that engage in technical differentiation and cost efficiency simultaneously achieved high performance and so did subsidiaries that engage in marketing differentiation and scale efficiency simultaneously.
Both Integration and Responsiveness forces play an important role for MNEs and can contribute to organizational goals in unique ways. Fan et al. (2008) argue that while existing literature argues that Integration helps MNE’s save costs and achieve global efficiencies (Chen & Cannice, 2006), Luo (2003, p.454) states that “to reap benefits from emerging market opportunities, heightened responsiveness is needed”. It has been argued in several pieces of the literature that there are differences in MNEs policy choices in regards to the IR framework in developed and emerging economies. This does not only have to do with emerging market consumers valuing different product attributes, but also with the fact that market activities etc. have to be adjusted to region context specific factors.
After introducing the IR framework in general, several links between the framework and sustainability can be made. Local Responsiveness and Global Integration forces can be seen in CSR strategies as well as general operational strategies. Locally Responsive strategies are likely to foster adaptation to local stakeholders as well as enabling firms to enhance their reputation in the local context. Globally Integrated CSR policies on the other hand are likely related to cost savings and the presence of multinational customers. Sustainability strategies are also likely related to the country-specific context of the MNE’s subsidiary. As argued in Meyer & Su (2011), at the level of subsidiaries, the MNE’s strategy influences, among other effects, the ways in which the subsidiary engages with its local environment in term of political activity (Blumentritt & Nigh, 2002) and CSR (Husted & Allen, 2006).
Sustainability considerations as such are important to firms for a number of reasons. Auger et al. (2010) argue that “rising social consciousness of consumers across the globe is
having a profound impact on the way organisations market their products and services both locally and globally. Consumers worldwide are exposed to a growing amount of information about and expanding market of international and local products and services (Auger et al., 2010), which is identified as one of the main pressures on MNE’s to adopt sustainable practices. The most important contribution that this paper makes to the topic is the combination of the International Business and CSR literature. As argued by Auger et al. (2010), “this combination brings new perspectives to the IB and CSR literatures and begins to fill gaps that have been mentioned by researchers in several areas such as a lack of research of research on CSR in emerging markets (Egri & Ralston, 2008) and more generally on consumers in emerging markets (Steenkamp, 2005)”.
Elliot and Freeman (2001) produced further insight into the behaviour of consumers by “uncovering relatively high elasticity’s of demand for products made under bad conditions but low elasticity’s for products made under good conditions, implying that companies risk losses from having their products identified as being made under bad conditions but have little to gain from marketing their products as being made under good conditions” (Auger et al., 2010). Their sample of consumers indicated that they are willing to place value on social attributes, which may imply that opportunities for new market development and segmentation exist based upon these attributes separately from the more traditional approach based on functional attributes and product quality (Auger et al., 2010). They also showed that overall consumers from developed countries do seem to place more importance on social attributes than consumers from emerging markets (Auger et al., 2010). Auger et al., (2010) further found that consumers from more developed economies have slightly stringer concerns about environmental and labour issues as revealed by purchase choice.
As argued by Kolk and Tulder (2010), “the ‘modern era of globalisation’ entails a balancing act between the components that are part of MNE’s ‘regular’ internationalization
strategies and broader CSR considerations. As a result, “entry strategies, subsidiary relationships, and the choice of country, product and market portfolios, both upstream and downstream, involve complex decision-making processes in which a variety of trade-offs come to the fore simultaneously: economic, legal, ethical, environmental and social (Kolk and Tulder, 2010). In the paper (Kolk and Tulder, 2010), it is also argued that it is difficult for firms to operate in such a way where they can be ‘economically profitable, law abiding, ethical and socially supportive’ when operating in a large number of different contexts with often diverging views of the role of business in society (cf. Devinney, 2009). Devinney’s statement is closely related to the IR framework. When operating on a global scale, there is a trade-off between the four different factors mentioned by him. Depending on the strategy chosen (on the IR continuum), firms will be challenged to achieve being ‘economically profitable, law abiding, ethical and socially supportive’. Ideally there is a trade-off between GI and LR forces to balance between the ‘four forces’.
Due to firms facing the ‘four forces’ described in the previous paragraph, MNEs CSR activities are seen as becoming increasingly strategic, in the sense that they affect the core business of the firm and its growth, profitability and survival (Verberke, 2009), with CSR as a potential source of competitive advantage (Porter & Kramer, 2006). “Virtually every activity in a company’s value chain touches on the communities in which the firm operates, creating either positive or negative social consequences” as argued by Porter and Kramer (2006), clearly notes the deep embedded-ness of businesses into society.
Despite CSR becoming an increasingly important consideration for both firms and stakeholders, research in sustainability related fields is still relatively limited. Kolk and Tulder (2010) argue that, together with the small number of IB studies on CSR and sustainable development in general, a very unbalanced geographical distribution of empirical research is notable. The challenge for those interested in CSR and sustainable development is
to move to the mainstream, not only in terms of outlets but also when it comes to embedding and relating to IB theories and themes (Kolk and Tulder, 2010). They further propose that one of the main reasons for limited attention may be the problems associated with linking CSR/ sustainable development to mainstream IB debates. Downstream and upstream perspectives are relevant to add both for IB in general, and for CSR/ sustainable development in particular (Kolk and Tulder, 2010). ‘Compared to upstream activities, where MNEs’ corporate partners are involved and two-sided commitments are at play (cf. Rugman & Verbeke, 2008), when it comes to sales, these are one-sided without any guarantees from consumers’ (Kolk and Tulder, 2010). According to Kolk and Tulder (2010), this not only implies greater liabilities for MNEs but also a more complex situation for implementing CSR across the whole value chain, from beginning to end. They give the following direction for future research: To better understand in which situation and under which conditions MNEs’ can not only gain a sustainable competitive advantage, but also play a role in furthering sustainable development, consideration of institutional, industry, organisation and supply and demand drivers seems appropriate (Kolk and Tulder, 2010). Downstream activities therefore pose an especially interesting case for research in relation to MNEs’ choice on the IR spectrum and sustainability. Furthermore, the impact of MNE’s on sustainable development is largely unclear and needs further investigation (Dunning & Fortanier, 2007). The lack of research within the areas discussed by the authors in this section constitutes the research gap that this thesis will try to fill in the best possible way.
Schrettle et al. (2013) argue that there are two mechanisms that explain why firms take action towards more sustainability. They group drivers of sustainability into two groups: exogenous and endogenous drivers (Schrettle et al., 2013). The three main exogenous drivers that they identify are environmental regulation, societal values and norms and market drivers, while endogenous drivers on the other hand are strategy, culture and resource base (Schrettle
et al., 2013). The downstream activities investigated in this research are therefore related to the exogenous drivers, and can have the effect that firm’s have to take into consideration adopting more sustainable downstream activities.
The exogenous drivers that are closely related to downstream sustainability considerations are societal values/ norms as well as market drivers. Accordingly, several dimensions of ethical/ sustainable downstream value chain activities, which are closely related to these exogenous drivers, are taken into consideration in this thesis. When it comes to sustainable downstream activities, ethical practices in marketing are one of top issues for businesses. Issues that can arise from unethical marketing practices are often related the expectations of consumers not being adequately met because of false marketing. As argued by Laczniak and Murphy (1991), when a marketing decision is ethically troublesome, its highly visible outcomes can be a public embarrassment or sometimes worse. They also put forward that according to studies, the categories salespeople and advertising practitioners were ranked at the bottom of the honesty and ethical standards scale (Laczniak and Murphy, 1991). At the same time, many marketers are often faced with different kinds of ethical dilemma’s (65 to 75 percent), which are described as lowering one’s personal values in exchange for increased organisational or personal profits (Laczniak and Murphy, 1991). It is clear though, that marketers should be ethical, because not doing so will likely generate significant personal, organisational and societal costs (Laczniak and Murphy, 1991). Organizational costs, which are most important for this thesis, typically take the form of reduced sales and loss of goodwill (Laczniak and Murphy, 1991). The Nestle Company example where they falsely marketed their baby formula in third world countries, and as a result suffered a public relations nightmare as well as a balance sheet catastrophe, as the derogatory publicity along with a substantial loss of sales was due to various boycotts of Nestle products worldwide (Laczniak and Murphy, 1991), is only one of the many examples.
Therefore, for the organization that takes its ethical duties seriously, the provision of mechanisms to help managers better morally reason through ethical problems and the establishment of a corporate culture which will help direct managerial actions toward beneficial ends goes far in the establishment of an ethically enlightened marketing organization (Laczniak and Murphy, 1991). These ethical marketing organizations in are the ones that have little to no marketing concerns as described in this thesis.
Companies that portray themselves as being sustainable are often the very opposite. Powell (2011) describes corporations, which provide mission and value statements with an ethical/ sustainable orientation, or which promote themselves as being such through their corporate brand, do not necessarily have an underpinning ethical identity, and may be found to be lacking in their ethical and sustainability approaches to corporate marketing in reality. The Deepwater Horizon catastrophe during 2010 is a perfect example for this. BP was operating under an environmentally friendly corporate brand, banner and logo, and faced a major revolt from a wide range of stakeholders who attacked the perceived lack of ethical identity underpinning the corporation and its corporate brand (Powell, 2011). This was particularly the case via environmentalist groups such as Greenpeace who effectively marshalled many stakeholders together against BP mainly via the Internet and social media, both during and after the disaster (Powell, 2011). Therefore, when it comes to marketing, firms can actually deceive consumers and the wider public in several ways. No matter in what way consumers are deceived, companies that act in such an opportunistic way are likely to face ramifications when their actions come to light.
Naturally, as pressure for becoming more socially responsible is increasing for firms, corporate managers are developing CRM (cause-related marketing) strategies that exploit the firm’s social initiatives to its advantage (Liston-Heyes and Liu, 2013). A big part of these CRM strategies is corporate sustainability reporting. From an ethical viewpoint, the practice
of extensive corporate social reporting is questionable. As explained by Liston-Heyes and Liu (2013), motivating factors for the extensive use of CRM are often sales/ revenue related rather than stemming from a genuine interest into these issues. Nevertheless, the question of the extend to which unethical considerations are partially motivating factors for CRM are far beyond the scope of this thesis and will therefore not be discussed in detail.
Furthermore, product safety issues are also important downstream sustainability consideration that can affect the consumer in several ways. In terms of importance, product safety issues are likely even more important than unethical marketing issues, as faulty products can cause serious harm to consumers. A series of high profile product recalls in recent years has shaken public confidence in the ability of manufacturers and governments to assure the safety of food and other products used by consumers (Gallup, 2008). Examples of high profile recalls include Toyota and General Motors. Thanks to this, Toyota incurred huge litigation fees, as well as estimated losses in the billions of dollars, due to lost sales, reduced manufacturing output, and enhanced incentive campaigns (Marucheck et al., 2011). Product safety problems are magnified in terms of scope and scale due to the global reach of today’s economy (Marucheck et al., 2011). A partial reason for this is that in the long and complex supply networks, many entities, including outsourcers and subcontractors located in emerging economies such as China handle the product as it moves across geographical and national borders, thereby creating many physical and temporal threats that pose a risk to product safety (Marucheck et al., 2011). Marucheck et al. (2011) identify primary safety challenges in that arise in five highly regulated industries: food, pharmaceuticals, medical devices, consumer products and automotive. One common denominator of these 5 industries is that they are rapidly globalizing, meaning that the supply chain is becoming longer and more complex (Marucheck et al., 2011). Product safety issues therefore occur in a number of key industries, and are an important factor of downstream sustainability research.
The effect of product safety, which is one of the downstream variables, on CFP can also be predicted fairly well using results from previous studies. Kaynak (2003) shows in his study that a positive relationship exists between the extend to which companies implement Total Quality Management and firm performance. TQM practices include the continuous improvement of a company’s product, and therefore can be seen as improving products in terms of safety features. Important factors that determine the overall effectiveness of a firm’s TQM practices depends on management leadership, which in turn is directly related to training, employee relations, supplier quality management and product design, and indirectly related to quality data and reporting and process management (Kaynak, 2003). Therefore, even though Product Safety can be seen as a downstream sustainability variable, it consists of several upstream and organizational factors. Supplier quality management for example emerges as an important component of TQM, directly and positively affecting product design, process management and inventory management performance (Kaynak, 2003). As companies increasingly outsource parts of their value chain activities with the benefit of accruing cost savings, supplier quality management is likely to be the most important factor in relation to product safety. Overall though, the most important mediating factor of TQM success is management leadership, as it drives the firm’s performance in the separate contributors to TQM (Kaynak, 2003).
The overall relationship that is being investigated in this thesis is that between several dimensions of CSR performance as well as Corporate Financial Performance (CFP). This relationship between a firm’s CSR performance and its CFP has been widely discussed in recent literature. Even though a wide range of research has been done in regards to this topic, there is no uniform insight that can be identified across a selection of these readings. There have been authors that identified a positive relationship between CSR performance and CFP (Wood, 2010; Margolis et al., 2007), while some showed a negative relationship, and some
no relationship at all. From reading the literature it is therefore very difficult to tell whether, and in what way, CSR performance is related to CFP. The papers discussed above also give no consideration to the kind of factors that are going to be discussed in this thesis. CSR performance is not divided into upstream/ downstream considerations, and the international factors of the CSR policies are also not given much consideration. The absence of any conclusive evidence regarding the relationship between CSR performance and CFP, as well as the lack of research into the effects of downstream sustainability and IB considerations when it comes to research on CSR are therefore the core motivations for the research carried out in this thesis.
One major conclusion that Wood (2010) could draw from his research is that a positive CSP – FP connection is now reasonably well established, despite the remaining measurement, methodological and theoretical issues. Waddock and Graves (1997) found that CSP does depend on financial performance and that the sign of the relationship is positive. They argue that this is in support of the slack resource theory, which basically states that firms with slack resources potentially available from strong financial performance may have greater freedom to invest in positive CSP (Waddock and Graves, 1997). They therefore take a different angle on the issue, arguing that the relationship between the two variables is reversed compared to the previously discussed authors. They also further found that financial performance also depends on good social performance. The effect may therefore lie in both directions, which is something to consider in the discussion section.
Lourenco et al. (2012) describes that some studies, such as Curran and Moran (2007), Consolandi et al. (2009), Cheung (2011), Doh et al. (2010), and Robinson et al. (2011) test whether inclusion in or deletion from, sustainability indexes (such as the Dow Jones Sustainability Stoxx Index) results in a positive or negative impact, and that results do suggest that investors do value CSP. He further shows that Garcia-Catsro et al. (2010) show
that KLD does not impact on financial performance, arguing that the positive relationship found in most of the previous research on the link between CSP and CFP becomes non-significant or even negative when endogeneity is properly taken into account (Lorenco et al., 2012). Lo and Sheu (2007) on the other hand, using the listing in the DJSI USA as a proxy for CS, show in their key findings that sustainable firms are rewarded with higher valuations in the market place. Loureco et al.’s (2012) findings are consistent with those of Lo and Sheu (2007), as they also find a positive relation between CSP and CFP, as investors value CSP. They also find that investors penalize large, profitable firms with low level of CSP, which face greater public scrutiny and pressures from stakeholders (Lourenco et al., 2012). The reason for this is that these firm’s are expected to signal sustainability leadership, and if they don’t they are penalized by the market (Lourenco et al., 2012). Artiach et al. (2010) found that leading CSP firms are significantly larger and profitable when compared with conventional firms. The key findings of Lourenco et al. (2012) were therefore that CSP is positively associated with the financial performance of large and profitable firms, which are able to signal their sustainability performance, and has a negative association with the performance of large and profitable firms that are not able to signal their sustainability performance. They also argue that these findings, which were obtained in the North American institutional setting, are not susceptible of generalization to other countries, especially those with very different characteristics (Lourenco et al., 2012).
On the other hand, Ziegler (2012) finds an overall weak or neutral effect of inclusion in the DJSI on CFP. A few potential reasons for this are given. For Example, while the inclusion in a sustainability stock index as a positive signal for higher corporate sustainability performance can lead to a higher firm reputation with positive consequences for financial success, proactive activities that are necessary for the inclusion in the DJSI World could also be a compulsory response to market pressure with negative consequences for corporate
financial performance (Ziegler, 2012). The results however do not support the pessimistic view of a negative impact of corporate environmental and social activities, which are necessary for the inclusion in sustainability stock indexes and which can lead to additional not directly productive costs (Ziegler, 2012).
Other studies however show that there are positive/ negative relationships between the two variables. With the results of their meta-analysis, Orlitzky et al. (2003) show that there is a positive association between CSP and CFP across industries and across study contexts. They further demonstrate that the universally positive relationship varies (from highly positive to modestly positive) because of contingencies, such as reputation effects, market measures of CFP, or CSP disclosures (Orlitzky et al., 2003). Furthermore, the causation seems to be that CSP and CFP mutually affect each other through a virtuous cycle: financially successful companies spend more because they can afford it, but CSP also helps them become a bit more successful (Orlitzky et al., 2003). Reputation seems to be an important mediator of the relationship (Orlitzky et al., 2003). This means that reputation plays an important role in determining a firm’s position in this cycle. Lopez et al. (2007) on the other hand found that the link between performance indicators and CSR is negative, which indicates that the effect of sustainability practices on performance indicators is negative during the first years in which they are applied. They further argue that it will be necessary to examine a longer time frame to see whether these practices acquire continuity and begin to influence corporate performance positively (Lopez et al., 2007).
Several studies therefore came to very different conclusions about the effect of CSR on CFP, it will therefore be interesting to see what the effect of Downstream Sustainability, as well as the firm sustainability performance on the IR spectrum will have on CFP. In order to investigate these relationships, several hypotheses are developed.
Hypothesis development:
In this thesis, the relationship between corporate financial performance and downstream sustainability performance will be investigated. The Corporate Financial Performance (CFP) is measured as the EBITDA margin (EBITDA/revenue). The CFP is therefore the dependent variable.
Downstream Sustainability Performance (DSP) is characterised by several independent variables. Firstly, there are four variables for downstream sustainability related concerns. These four concerns are Product Safety (PS), Marketing Communication (MC), Antitrust (AT) and Other. A variable for the extend to which CSR policies are locally responsive or globally integrated is added to the equation next. The final independent variable relates to the firm’s inclusion in the Dow Jones Sustainability Index. The control variables that have been chosen for this research are consistent with previous approaches (Lopez et al., 2007) and are the size of the firm measure by total assets, and a variable capturing the risk level of the firm as measured by the debt ratio (total debt/ total assets).
In order to be able to create a theoretical model for the variables presented in this thesis, it is essential to acknowledge the trade off that is present with an increase in CSP in general, and also the separate variables. This trade off is between the benefits that a firm can potentially obtain from increasing it’s CSP, such as a positive effect on the reputation of the firm and therefore an increase in sales etc., as well as negative effects such as an increase in costs as well as coordination issues between the different CSR practices in separate locations. The effect of the variables on CFP is therefore likely to be closely related to the extend that this trade off is present, and to what extend the positive effects of an increase in CSP-related variables can off-set the higher costs of implementing these practices.
Hypothesis 1a: There is a negative relationship between Product Safety Concerns and Corporate Financial Performance.
Hypothesis 1b: There is a negative relationship between Marketing/ Contracting Concerns and Corporate Financial Performance.
Hypothesis 1c: There is a negative relationship between Antitrust Concerns and Corporate Financial Performance.
Hypothesis 1d: There is a negative relationship between all Other Concerns and Corporate Financial Performance.
Hypothesis 2: The more Globally Integrated a firm’s CSR policies are, the better is its
Corporate Financial Performance (CFP).
Hypothesis 3: There is no relationship between a firm’s membership of the DJSI and it’s
Corporate Financial Performance (CFP).
Hypothesis 4: Risk will have a negative relationship to Corporate Financial Performance
(CFP).
Hypothesis 5: The Size of the firm will have a positive relationship on Corporate Financial
Performance (CFP).
Theoretical model:
The theoretical model that is at the core of this thesis relates to the Hypotheses stated in the previous section and will hope to explore the relationship between the independent and dependant variables. This thesis looks at several different variables and at the way that they relate to CFP. The combination of variables is unique, as no previous research paper has specifically sought to integrate the position of a firms CSR policies on the I-R scale to
downstream sustainability issues. The model developed here is therefore a combination of the different variables and their effects on CFP.
When relating the I-R model to sustainability practices, many of the I-R arguments can be used to predict a likely relationship between the CSR policy’s position on the I-R scale and the effect this is likely to have on CFP. It has been previously argued by authors that it is necessary for MNE’s to find a balance between factors of Local Responsiveness as well as Global Integration. The question though is, if this “balancing-act” has the same effect for CSR policies as it does for a MNE’s ‘normal’ business. One of the main drivers for firm’s adopting globally integrated CSR policies for example are likely to be cost reduction considerations. Cost reduction advantages that stem from having globally integrated CSR policies are therefore likely to reflect positively on CFP. The issue here is that the motivations of firms to have a balanced approach to their CSR policies are likely to be different from the company’s ‘normal’ business.
As there have been several studies on the effect of DJSI membership in CFP, there are several papers that can be used as a reference point for the development of the theoretical model. First and foremost, as Ziegler & Schroeder (2010) state that the inclusion of a company in sustainability stock indexes can be treated as an appropriate indicator for the level of CSP, it becomes clear that it is appropriate to be used as a “proxy” for CSP in our equation. Further studies that use the same approach when comparing the effect of the inclusion in the DJSI on CFP are McWilliams & Siegel (2000) and Becchetti et al. (2008). Most importantly for the development of the DJSI inclusion effect on CFP hypothesis, Ziegler (2012) found an overall weak or neutral effect of inclusion in the DJSI on CFP. Previously, Ziegler & Schroeder (2010) found that the inclusion of European firms in the DJSI implies a positive relationship to CFP when using restricted econometric analysis, but in more flexible panel models the impact become ambiguous. Their results therefore stress
the importance of the use of panel data. It is therefore predicted that the membership of the DJSI will not have a direct impact on the CFP.
When it comes to the Downstream Sustainability variables, the relationship between the four variables and CFP is a tricky one to predict. The main reason for there being difficulties in making a good prediction is the lack of studies that investigate the relationship between these two variables. Even though, as discussed before (Wood, 2010), the KLD database is one of the most frequently used in research regarding CSP, there has not been any research, which focuses on only the downstream sustainability activities. As shown in the hypotheses though, several predictions can still be made. As discussed previously, both Marketing/ Contracting Concerns and Product Safety Concerns are both critical in influencing the reputation of a company and its products, which in turn can lead to a decrease in sales and can therefore directly affect revenues/ profits (Laczniak and Murphy, 1991). Therefore, when making predictions about the likely effect of downstream sustainability concerns on CFP, downstream sustainability concerns are likely to affect CFP in a negative way. This closely correlates with the main hypothesis as well as the sub-hypotheses 1a to 1d.
Some of the general features of the different dimensions in the I-R framework can also be used to predict the relationship between the extend to which corporate CSR policies are globally integrated or locally responsive and CFP. To date, there has been no relevant literature that looks at CSR policies from this perspective. As seen in the literature review section, firm’s that operate in a globally integrated manner are often doing so due to cost efficiency considerations. The relationship between cost savings and globally integrated CSR policies is likely to be of a similar nature. Having a CSR standard that is being implemented on a global scale is likely to be much more cost efficient than having to adapt CSR policies to local circumstances. Globally integrated CSR policies are therefore likely to be related to
better CFP, and having more locally responsive CSR policies is likely to be negatively related to CFP. This prediction can be seen in the hypotheses.
The effect of the control variables on CFP can be easily predicted by looking at some of the previously discussed literature. Risk for example is very likely to be negatively related to CFP, while the size of a firm is likely positively related to CFP. Through controlling for these two variables, the results that are obtained from the other variables are likely to be more reliable.
Sample construction and data collection:
The research approach adopted in this thesis is of a quantitative nature. Due to the fact that the data was collected for a set of 124 companies over the time period of five years, panel data analysis with the standard multiple regression function was used to analyse the data in this thesis.
Data and Method
Sample Selection:
The data used in this thesis was obtained from secondary sources, namely Databases as well as Company’s CSR reports. In order to obtain all the data necessary for the variables in this thesis, data had to be collected from several sources and combined afterwards. The databases that were used are as follows: CRSP/ Compustat merged, KLD and Dow Jones Sustainability Indices. Furthermore, one of the variables was obtained from analysing the CSR reports of company’s in the sample.
First and foremost, the sample contains 124 US firms and the time span for the data collected on these firms is from 2009 until and including 2013 (5 years). Several criteria were used to end up with the final sample of 125 firms, for which there are 625 observations for the time frame of this research. Most importantly, the firms in our sample had to be represented in the KLD database. This is due to the fact that the downstream sustainability variables were obtained from this database. At present, the approximate total number of companies covered in the KLD database is 3000. This group of 3000 companies includes the 500 largest US companies, the MSCI KLD 400 Social Index, and the 3000 Largest US companies. In order to obtain a sample of firms that operate on a global scale, so as to be able to use the I-R variable, the initial sample of firms was obtained by choosing companies that were represented in the S&P 500 Index throughout all 5 years of the time frame that was used
for the analysis. The second step in obtaining the sample of companies was to control for specific industries that were determined to be most applicable to this research. Companies from the following industries were omitted from the research: Energy (Industry Group 1010), Materials (1510), Commercial & Professional Services (2020), Transportation (2030), Consumer services (2530), Media (2540), Retailing (2550), Financials (4010 - 4040) and Software and Services (4510). The remaining companies in our sample were therefore: Capital Goods (2010), Automobiles & Components (2510), Consumer Durables & Apparel (2520), Consumer Staples (3010 – 3030), Health care (3510 – 3520), Technology Hardware & Equipment (4520) and Semiconductors & Equipment (4530). The final 125 companies in the sample therefore had the following attributes: They were represented in the S&P 500 Index throughout all 5 years of the tested time horizon and they were members of the industries noted above. Thanks to the filtering of companies, the data for all variables was available for all companies in the sample.
Measurement of dependant variable:
The dependant variable and the control variables were computed with the help of data obtained from the CRSP/ Compustat merged database. The database was accessed through Wharton Research Data Services (WRDS). This database was chosen over the Copustat database because it made the process of combining the data obtained from the KLD database with financial data that was needed easier. CRSP/ Compustat merged and KLD both use Ticker’s as company identifiers, whereas Compustat does not use tickers. Choosing CRSP/ Compustat merged was therefore the logical step to take. Data for the following variables was obtained from this database. Firstly, the Dependant Variable (EBITDA margin), was calculated through dividing the EBITDA by Total Revenue. Return on Equity was also calculated by dividing the Net Earnings (In the form of the EBITDA) through Total Equity.
Because of reasons described in the methodology section, we did not end up using the ROE as a ‘checking mechanism’.
The EBITDA profit margin was chosen over other alternatives because it provides a clearer view of a company’s core profitability due to excluding depreciation and amortization (investopedia.com). Due to this it was chosen over the more commonly used EBIT margin.
Measurement of independent variables:
In order to obtain the data for the independent variables related to Corporate Downstream Sustainability (CDS), the KLD database was used. This database is also accessible through WRDS. KLD stats is a data set with annual snap-shots of the environmental, social and governance performance rated by KLD Research & Analytics, Inc. (Getting Started with KLD). Wood (2010) also showed that the KLD database is actually one of the most frequently used CSP variables in empirical papers. The database encompasses several categories of variables. These Qualitative issue areas are the following: Community, Corporate Governance, Diversity, Employee Relations, Environment, Human Rights, and Product. Furthermore, data on controversial business issues, such as Alcohol, Gambling, Tobacco, Firearms, Military and Nuclear Power is also available. In order to investigate into downstream sustainability issues, the qualitative issues area ‘Product’ was used to obtain data for the variables. Within this area, a number of strengths and concerns are discussed. The problem is that a number of these strengths and weaknesses were not available for the time frame that is being discussed in this thesis. The only data that was consistently available, were concerns that relate to Product Safety, Marketing/ Contracting, Antitrust and Other. The definitions of the variables are described in the next paragraph. The data available on KLD for these product concerns is a dummy variable of 0 or 1, 1 meaning that there is a concern, and 0 that there is none.
A Product Safety Concern is related to the company having recently paid substantial fines or civil penalties, or is involved in major recent controversies or regulatory actions, relating to the safety of its products and services (Getting started with KLD). Marketing/ Contracting Concerns mean that the company has recently been involved in major marketing or contracting controversies, or has paid substantial fines or civil penalties relating to advertising practices, consumer fraud, or government contracting (Getting started with KLD). Antitrust Concerns are related to the company recently having paid substantial fines or civil penalties for antitrust violations such as price fixing, collusion, or predatory pricing, or is involved in recent major controversies or regulatory actions relating to antitrust allegations (Getting started with KLD). Other Concerns mean that the company has major controversies with its franchise, is an electric utility with nuclear safety problems, defective product issues, or is involved in other product-related controversies not covered by other KLD ratings (Getting started with KLD).
The DJSI membership variable of this research was obtained from the Dow Jones Sustainability Indices. The DJSI focuses on the triple bottom line of sustainability in terms of economic, environmental and social criteria, with a strong focus on long-term shareholder value (Lopez et al., 2007). This database is not readily accessible for UvA students, and therefore a signed request to obtain access to the database has to be filled out. Upon handing in this request, a DJSI representative sent an email with PDF’s containing lists of companies that were included in the DJSI North America in the time frame that is being investigated in this thesis. The DJSI membership variable is therefore a dummy variable, where 0 stands for non-membership of the DJSI North America, and 1 stands for membership of said index in the respective year. In similar fashion to prior literatures, membership of the DJSI will be used to distinguish between high and low performance in terms of CSP. Inclusion in the DJSI indicates that the firm is a high performer, while non-inclusion in the index is used as an
indicator for low CSP performance. Ziegler and Schroeder (2010) for example argue that a firm’s inclusion in sustainability indexes can be used as an appropriate indicator for their CSP, which will be tested in this study for the sample companies.
Lastly, the I-R variable used in this thesis was obtained through thorough analysis of company CSR reports. This is a somewhat simplistic approach to obtaining this variable, but due to reasons of time and availability, this was seen as the only appropriate approach to obtaining this data. A scale of 1-7 was used for the purpose of coding the CSR practices of the companies in the sample based on their Integration vs. Responsiveness. A 1 was assigned when the company was displaying fully Locally Responsive CSR practices. In its essence, this means that CSR practices are specially created/ adapted for the different locations of a firm and cater to local needs and stakeholders. When a company’s CSR practices were fully Globally Integrated, a 7 was assigned to them. Having fully Globally Integrated CSR practices means that there is one global standard for all CSR Practices that is being implemented on a global scale. As there are only very few cases that can be classified as being either fully Locally Responsive or fully Globally Integrated, scores between 1 and 7 were given based on the firm’s performance on the scale. In order to guarantee some kind of consistency, the CSR reports were analysed and coded by myself, and two other classmates that performed similar research and with which I therefore worked together for some of the data collection part of the thesis. For the scores that were different between the three of us, we went through a thorough process of going through the CSR reports together and justifying each other’s choice of score, and only thereafter gave a final consolidated score. All CSR reports were given the same thorough process of analysis guaranteeing that the scores given to the different companies are consistent across the whole sample. This approach and its limitations were the only viable option to collect the IR variable given the issues of time as well as availability of data. This is further discussed in the limitations section of this study.
Control variables:
Finally, control variables were used in order to control for certain factors. We used the CRSP/ Compustat merged database to obtain data for our control variables. These control variables are the Risk level of the firm as measured by Total Liabilities divided by Total Assets, and the Size of the firm as measured by Total Assets. The control variables that are being used in this thesis have frequently been chosen in studies, such as Lopez et al. (2007), that are comparable to this one. All financial data that was required for this thesis was therefore obtained from CRSP/ Compustat merged.
Due to the fact that the data analysed consists of 124 companies data measured over a timeframe of 5 years, a variable was also included to control for the year effect. This was done by creating a dummy variable for each respective year for SPSS to take into consideration when performing the regression analysis.
Method:
In order to test this model, a standard multiple regression approach was used in order to investigate the effect of the independent variables and the dependant variable. This approach was chosen, as there is a set of independent variables that is likely to cause variance in the dependant variable. On top of that, the standard multiple regression approach will also give an indication to the extend that each individual variable contributes to the variance in the dependant variable. Thanks to this, all of the hypotheses stated previously can be tested.
After transferring my data from excel to SPSS, several tests were run in order to find any errors in the data. After running the frequencies, one of the dependant variables that was to be used as a further “checking” mechanism, ROE, proved to have too big a standard deviation due to having many extreme outliers. It was therefore decided to continue the
analysis with only the EBITDA margin as the dependant variable. Furthermore there were several missing values in the data obtained from KLD (Downstream sustainability indicators). These missing cases were excluded pairwise.
Next up, I further ran descriptive and explore procedures. Due to the fact that “With reasonably large samples, skeweness will not make a substantive difference in the analysis” (Tabachnick & Fidell, 2007), and “Kurtosis can result in the underestimate of the variance, but this risk is also reduced with a large sample” (Tabachnick & Fidell, 2007), the results for both skeweness and kurtosis values were not deemed to play too big a role. The reason for this is that a sample of 620 observations is can be categorized as a large sample (Tabachnick & Fidell, 2007). Nevertheless, the results for skeweness and kurtosis can be seen in Table 1 below.
Table 1:
Descriptive Statistics
N Minimum Maximum Mean
Std.
Deviation Skewness Kurtosis
EBITDAmargin 620 -0.28 0.55 0.20 0.11 0.39 0.24
RISK 620 0.20 1.19 0.58 0.18 0.18 -0.33
SIZE 620 1422.00 781818.00 34438.59 72486.89 7.22 64.01
IRRating 620 1.00 7.00 4.08 1.72 0.27 -0.99
As seen in Table 2, Tolerance and VIF values were also checked. Tolerance is an indicator of how much the variability of the specified independent is not explained by the other independent variables in the model, while VIF acts as an indicator for multicollinearity (Pallant, 2011). The collinearity for the variables in in the range of .820 to .934, indicating that there is no high collinearity between the variables. Furthermore the VIF values are between 1.076 and 1.220, which is also well below the cut of point of 10.
Table 2: Tolerance VIF (Constant) PS 0.819 1.221 MCC 0.812 1.232 AT 0.911 1.097 Other 0.917 1.091 IRRating 0.919 1.088 DJSImembership 0.916 1.092 RISK 0.929 1.076 SIZE 0.844 1.184 Year 0.907 1.102
On the Normal P-P Plot, the points lie in a reasonably straight diagonal line from bottom left to top right, suggesting that there are no major deviations from normality. In the Scatterplot of the standardised residuals, the residuals are roughly rectangular distributed, with most of the scores concentrated in the centre. This is another positive indicator as it adds, as deviations from a centralised rectangle suggest some violation of the assumptions.
Table 3: Correlations Variable 2 3 4 5 6 7 8 9 10 1.EBITDA 0.030 0.230 0.162 -0.106 0.088 0.050 -0.387 0.074 0.042 2.PS 1 0.287 0.188 0.117 -0.187 0.138 0.140 0.288 -0.080 3.MCC 1 0.205 0.171 -0.103 0.153 -0.054 0.141 -0.256 4.AT 1 0.154 -0.097 0.085 -0.027 0.158 -0.085 5.Other 1 -0.057 0.006 0.119 0.104 -0.173 6.IRRating 1 -0.165 -0.133 -0.190 0 7.DJSImembership 1 0.050 0.225 0.022 8.RISK 1 0.149 0.031 9.SIZE 1 0.033 10. Year 1
Hypothesis testing – Multiple regression analysis
The standard multiple regression analysis was used in order to explain the extend to which the independent (PS, MC, AT, Other, IRRating, DJSImembership and Time) and control variables (SIZE and RISK) affect variance in the dependant variable (CFP). It is also useful in explaining the extend to which each of the independent variables causes unique variance in the dependant variable. Standard multiple regression analysis was chosen over hierarchical multiple regression due to the fact that there has been a very small amount of research done into the relationship between the variables that are being tested in this thesis, meaning that no adequate theoretical support could be given to entering the variables in any specific order. Without any theoretical groundwork having been done before, using the hierarchical multiple regression method is unsuitable.
Downstream sustainability concern variables (Product Safety, Marketing Contracting Concerns, Antitrust Concerns and Other Concerns), I-R Rating, Membership of the DJSI, Time, SIZE and RISK were used in standard regression analysis to predict Corporate Financial Performance.
The prediction model was statistically significant at a significance value of .000. The R Square value is .244. This means that the model explains 24 % of the variance in the dependant variable. The regression model therefore predicts CFP significantly well.
Table 4: STd. Beta Sig. PS 0.001 0.978 MCC 0.222 0.000 AT 0.123 0.001 Other -0.108 0.004 IRRating 0.089 0.018 DJSImembership 0.015 0.690 RISK -0.366 0.000 SIZE 0.099 0.012 Year 0.099 0.009
The strongest predictor of the EBIT margin was the Risk variable. Marketing/ Contracting Conern was also a strong predictor of the dependant variable. Apart from the Product Safety and the DJSI membership variable, all other variables also predicted the EBIT margin to a lesser extend. Out of the variables in the model, the PS as well as the DJSI membership variable show no significance in explaining the variance in the dependant variable. This is further supported by the low standardized coefficient Beta values.
This ranking of importance can be taken from the standardized Beta values (table 4), as they provide an insight into the ‘importance’ of a predictor in the model. The Risk variable has a standardized beta value of -.366, which is by far the highest value for any variable in the model. Marketing/ Contracting Concerns have the second biggest beta value of .222, being followed by Antitrust Concerns (.123), Other Concerns (-.108) and Size (.099), the Year (.099) and the IR Rating (.089). DJSI membership (.015) and Product Safety Concerns (.001) were by far the least important predictors of the model. By looking at the Sig. values it becomes evident that only some of our variables have made a significant contribution to predicting the outcome. The ones that are significant are Risk (.000), Marketing/ Contracting Concern (.000), Antitrust concerns (.001), Other Concerns (.004) as well as the IR Rating variable (.012) and Size (.012).
Several of the variables used in this research are therefore related to the CFP, with some of the other variables showing no statistical significance in predicting the model. As shown in the previous paragraph, if not taking into account the control variables, Marketing/ Contracting Concerns, Antitrust concerns and Other Concerns, in that order, made the biggest strongest unique contribution to explaining the dependant variable, when the variance explained by all other variables in the model is controlled for (SPSS survival guide). However, even though these variables show significance, their effect on CFP is often not as expected. For example, the results for the Marketing/ Contracting Concerns indicate that
being involved in a concern of this nature is positively related to CFP. This is because initially, Marketing/ Contracting Concerns were expected to be negatively related to CFP. The same was the case for Antitrust Concerns. Other Concerns on the other hand were negatively related to CFP, meaning that they had a negative effect. For the IR Rating, there is also a positive relationship between having a higher score on the scale (which means that the corporate CSR policies are more globally integrated) to CFP. DJSI membership and Product Safety Concerns on the other hand were not able to predict CFP. The results obtained are further discussed in the following section.
Discussion:
After interpreting the results of the regression analyses performed on the model, making certain inferences about the results is possible. Firstly, discussing the hypotheses made earlier will set the groundwork for the discussion.
H1, which states that there is a positive relationship between Corporate Downstream Sustainability and Corporate Financial Performance was only proven to some extend. As discussed previously, Corporate Downstream Sustainability is composed of the four concerns described previously (marketing/ contracting, antitrust, product safety and other), and there are four sub-hypotheses, one each for every downstream sustainability variable. A ‘perfect’ score for a company would be one where there are 0 concerns, and the ‘worst’ would be if the company has 4 concerns. A company that is a high performer in terms of CDS would therefore have a low score in terms of the 4 variables, while a firm that is a low performer in terms of CDS would have a high score. Out of these four concerns, ‘Product Safety’ showed no significance in predicting CFP. The other three concerns on the other hand do show significance in predicting CFP. The standardized beta values show that both ‘Marketing/ Contracting Concerns’ and ‘Antitrust Concerns’ are positively related to CFP. Only ‘Other Concerns’ are negatively related to CFP. As there have been very mixed results with regard to the different components of Corporate Downstream Sustainability and their effect on CFP, H1 is not proven. Out of the four sub-hypotheses, only H1d was proven. H1 in general was therefore not proven.
The results for H2 are easier to interpret than those for the previous hypothesis. It states that the more Globally Integrated a firm’s CSR policies are, the better is its Corporate Financial Performance. It is therefore clearly related to the score on the IR scale. The results from the regression analysis are statistically significant, and also show that the higher a firm’s score on the IR scale is (meaning the more globally integrated a firm’s CSR policies are), the
better it’s CFP. The results therefore provide enough evidence to give credibility to and thereby proving H2.
H3, by predicting that there is no relationship between membership in the DJSI and Corporate Financial Performance, is very much based on the findings of previous literature when it comes to the effect of a firm’s membership in the Dow Jones Sustainability Index on CFP with panel data. The hypothesis was proven, as the DJSI variable had no statistical significance in predicting CFP.
Lastly, H4 and H5 relate to the two control variables used in the regression model. They correctly predict that Risk has a negative relationship to Corporate Financial Performance, while the size of the firm has a positive relationship to CFP.
The effect of the I-R variable on CFP is interesting. Devinney et al. (2010) argue that empirical studies of the IR framework show no significant differences in performances according to firm strategy as defined by that framework (transnational, global etc.). In the case of Devinney et al. (2010), one possible reason for this lack of a clear relationship is the fact that firm performance is dependant on factors not accounted for in the I-R framework. They further argue that Prahalad and Doz identify economic, technological, political and competitive factors that influence the varying need for GI and LR, but omit considerations of managerial creativity or organizational form (Devinney et al., 2010). Certain factors have to be taken into consideration when applying this thinking to Sustainability related IR forces.
It is likely that there is much less to gain from having LR sustainability practices as compared to having LR business practices in general. Being locally responsive can benefit the company through better relationships with the customers and local governments, leverage valuable information about the local business environment and let this knowledge feed back into the company’s products. In terms of locally responsive sustainability practises on the other hand, potential gains are limited. For instance, when CSR practices are locally
