Juggling corporate social responsibility : how internal and external corporate social responsibility activities, and the balance between them, affect corporate reputation

(1)

Juggling corporate social responsibility

How internal and external corporate social responsibility activities, and the

balance between them, affect corporate reputation.

Student: Florieke van der Laan / Student number 11410558 M.Sc. Business Administration, Strategy Track

University of Amsterdam, Faculty of Economics and Business Supervisor: M.Sc. H. Fasaei

(2)

STATEMENT OF ORIGINALITY

This document is written by Florieke van der Laan 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)

ABSTRACT

The following thesis explores the answer to the question: how do internal and external

corporate social responsibility (CSR) activities affect corporate reputation? Firstly, the

important mediating effect of corporate reputation within the corporate social responsibility (CSR) and corporate financial performance (CFP) relationship has been assessed. Furthermore, the division between internal CSR activities and external CSR activities is made, in order to define the true effect of these two elements on corporate reputation. Moreover, empirical evidence is found that support the hypotheses through the use of mixed models regression analyses based on a sample of 426 firms from the United States of America between 2008 and 2014. The results of this study indicate that corporate reputation acts as a mediator within the CSR and CFP relationship. Furthermore, regarding the CSR division, external CSR activities are estimated to have a bigger positive effect on corporate reputation than internal CSR activities. Lastly, it was found that the bigger the imbalance between internal and external CSR activities the more negative the effect on corporate reputation. Hereby, this study contributes to the CSR literature as it has found additional empirical support for the mediated relationship of CSR and CFP through corporate reputation. Furthermore, to the knowledge of this author, this study is the first to define the effect of internal or external CSR activities on corporate reputation and determine the negative effect of an internal and external CSR gap on corporate reputation.

Keywords: Corporate reputation; corporate financial performance; corporate social responsibility; internal corporate social responsibility; external corporate social responsibility; greenwashing

(5)

INTRODUCTION

‘The Social Responsibility of Businesses is to Increase its Profits’ (Friedman, 1970).

Even though this statement is around fifty years old, it is proving to be more relevant than ever as the literature on corporate social responsibility (CSR) and its relationship with corporate financial performance (CFP) is growing at an exponential rate. Moreover, many have tried and succeeded in proving Friedman both right and wrong as the complicated relationship has been found to be considerably ambiguous. Even though the majority (59%) of scholars has presented a small positive relationship between CSR and financial outcomes, a concise explanation for this positive direction has not been found yet (Aguinis & Glavas, 2012). Therefore, many have expressed the need to further research the underlying mechanisms that moderate or mediate this relationship (Aguinis & Glavas, 2012; Bhattacharya & Sen, 2004; Lee & Jungbae Roh, 2012; Luo & Bhattacharya, 2006; Siegel & Vitaliano, 2007). Furthermore, some have argued that distinguishing between internal and external CSR activities may contribute to the demystification of this relationship and could aid in finding a reason for the inconsistent results (Hawn & Ioannou, 2016). Additionally, others have suggested that corporate reputation, as an intangible asset, could be a mediating factor that may help in clarifying the positive findings (Aguinis & Glavas, 2012; Luo & Bhattacharya, 2006; Saeidi, Sofian, Saeidi, Saeidi, & Saaeidi, 2015).

Moreover, when diving further into the topic of CSR it becomes clear that within previous research, CSR is often perceived as one activity instead of a complex network of activities that can be intertwined or independent from each other. Hereby, there is often little transparency and clarity around the topic. Therefore, some scholars suggest that firms should start to distinguish between people, planet, profit elements and/or internal and external CSR activities (Elkington, 2004; Hawn & Ioannou, 2016; Schons & Steinmeier, 2016). Furthermore, while this division is rarely made in the literature or in practice, Hawn & Ioannou (2016) found

(6)

that mapping out these activities and determining the balance between them will greatly affect a firm’s market value and is, therefore, a valuable exercise.

Additionally, as said, many have tried to find a conclusive argument pro or against Friedman’s statement, but it remains difficult to not only argue how CSR directly adds value to a firm but also to what extent this effect is present. Failing to demonstrate this added value will thereby add to the idea of irrelevance and dismissal of such activities by investors (Burke & Logsdon, 1996). More often than not, a firm solely focuses on the financial results of CSR, hereby only taking the demands of its shareholders into consideration (Margolis & Walsh, 2003). Moreover, the demands of other stakeholders such as customers and employees are hereby often overlooked while at the same time these stakeholder groups are increasingly expecting firms to invest in CSR initiatives. In addition to this, these forgotten groups largely influence the number of CSR activities of a firm and determine what type of activities should be considered as salient (Aguinis & Glavas, 2012). It is therefore important to not only consider the traditional performance measures of profitability but also to assess the environmental and social initiatives of a firm. By doing this, the diverging interests of stakeholders will be taken into account which will not only benefit the reputation of a firm, but it will also lead to higher investments and an increased attractiveness of the firm for future employees (Bhattarcharya & Korschun, 2006) Additionally, taking these demands into account willresult in more balanced CSR actions and hereby benefit the CFP (Hawn & Ioannou, 2016). Furthermore, by proactively implementing this way of working, firms will ensure their control over the situation and are thereby better able to nurture their relationships with salient stakeholder groups. These relationships are especially important as they are essential for building a positive firm reputation (Rupp, 2011; Waddock, 2008).

Moreover, this corporate reputation has often been considered to solely influence or be influenced by external stakeholders (Aguinis & Glavas, 2012; Brammer & Pavelin, 2006; Ellen,

(7)

2006; Orlitzky, Schmidt, & Rynes, 2003; Sen & Bhattacharya, 2001). However, a handful of scholars have addressed the internal effect of a positive corporate reputation and found contributions to the commitment and motivation of employees and thereby their productivity (Boyd, Bergh, & Ketchen, 2010; Godfrey, 2005; Maignan & Ferrell, 2004). Therefore, this thesis argues the added value of balancing organizational CSR activities, hereby considering the wishes of both internal and external stakeholders. Moreover, by overlooking the importance of this balance, and thereby underestimating the influence of CSR activities on corporate reputation, an imbalance could occur and negatively impact the financial performance (Hawn & Ioannou, 2016). Furthermore, as both CSR activities and corporate reputation are influenced by a firm’s stakeholders, these two elements are considered in this thesis to help enlighten the complex CSR and CFP situation (Aguinis & Glavas, 2012). Thus, within the next chapters, the gap in the empirical knowledge will be addressed by focusing on the mediating effect of reputation in the CSR and CFP relationship, hereby particularly addressing the influence of external and internal activities on corporate reputation. The defined gap has hereby led to the following research questions: How do internal and external CSR activities affect corporate

reputation? Moreover, as the effect of this division on corporate reputation has not been tested

before, to the knowledge of the author, the findings of this thesis are expected to produce novel contributions.

Concludingly, this thesis contributes to the literature in several ways. Firstly, by separating CSR activities into two types, this thesis will advance previous research regarding the specification of CSR activities (Hawn & Ioannou, 2016). Moreover, as the exercise of separating CSR into internal and external is still novel, little research has been done on the separation of these activities and the consequences of making such a division. Thereby, as addressed by Hawn & Ioannou (2016), investigating the effect of defining CSR into two separate activities on corporate reputation will also aid in providing an explanation for the

(8)

inconsistency within the CSR and CFP results. Moreover, by determining these actions, and further illustrating the value of doing this, the results will support in defining the quantitative benefits of CSR activities in the future (Porter & Kramer, 2006).

Secondly, by addressing the expressed need to further research the underlying mechanisms (i.e. mediators) of the CSR and CFP relationship, this thesis will provide additional empirical evidence to the current CSR literature regarding the mediating effect of corporate reputation (Aguinis & Glavas, 2012; Bhattacharya & Sen, 2004; Lee & Jungbae Roh, 2012; Luo & Bhattacharya, 2006; Siegel & Vitaliano, 2007). Additionally, as combining the two trains of thoughts (i.e. corporate reputation and internal/external activities) has, to the knowledge of the author, never been done before, and both these elements have been recognized to be able to aid in further demystifying the CSR and CFP relationship. Therefore, the theoretical framework, and outcomes of this thesis are considered to be novel contributions to the literature.

Thirdly, on a managerial level, this research will help in reducing the confusion around CSR activities. Creating clarity regarding the different CSR activities of a firm will help managers in structuring their future CSR plans as it will facilitate in creating future roadmaps. Moreover, by considering the demands of salient stakeholder groups and therefore investing in CSR actions, a firm will enhance its corporate reputation and increase the commitment among its stakeholders (Waddock, 2008). Furthermore, by illustrating the importance of defining your CSR actions and the effect on corporate reputation and CFP, the results of CSR will be made more tangible and visible (Hawn & Ioannou, 2016). Consequently, these results will help managers in their dialogue with shareholders when explaining the motivation behind the CSR initiatives and convince them of the necessity of making the financial investments that accompany these activities (Burke & Logsdon, 1996). Therefore, the results of this thesis will not only contribute to the current empirical research but will also help to emphasize the

(9)

importance of specifying the CSR actions and the (in)direct effect this has on corporate reputation and on CFP.

After this introduction, this thesis will continue with an in-depth overview of the current literature regarding internal and external CSR activities, the CSR and CFP relationship and the mediating effect of corporate reputation. Thereafter, the hypotheses will be introduced, and the methodology will be defined. Once the foundation of the thesis has been established, the results of the quantitative analysis will be described, and the most important conclusions will be shared. Moreover, the discussion section will place these conclusions into a broader context and emphasize the contributions of the thesis. Lastly, this last section will also determine the limitations of the thesis and the opportunities for future research.

(10)

THEORY DEVELOPMENT

Within this chapter, the current literature on CSR, different CSR activities, corporate reputation and corporate financial performance will be reviewed. Moreover, previous research regarding the CSR and CFP connection and the mediating effect of corporate reputation will be discussed and the gaps in the literature will be determined. Based on these defined gaps, the to be tested hypotheses are formulated.

Corporate social responsibility

Throughout the literature, there has been little agreement on a final definition of CSR. Within this thesis, the following definition has been chosen as it has been frequently cited by a multitude of authors (e.g. Aguinis & Glavas, 2012; Pisani, Kourula, Kolk, & Meijer, 2017; Rupp, Williams, & Aguilera, 2010). CSR is: ‘‘context-specific organizational actions and

policies that take into account stakeholders’ expectations and the triple bottom line of

economic, social, and environmental performance.” (Aguinis, 2011, p. 5)

As previously discussed, some argue that a firm only has one responsibility and that is to increase profits (Friedman, 1970). However, based on the definition presented above others contend that a firm should not base their responsibilities on one singular type of performance (i.e. financial) but managers should take three responsibilities into account when assessing the performance of a firm. This so-called triple bottom line was first introduced by John Elkington in 1994 (Elkington, 1998). Elkington argued that firms should consider the total costs of their operations because only if you incorporate environmental and social costs, besides your economic costs, you are able to build socially responsible firms and capture the total firm performance (Elkington, 1998; Hindle, 2009). Furthermore, this triple bottom line or ESG performance has been picked up and implemented by multiple companies, public and non-profit

(11)

organizations, to more comprehensively evaluate their performance based on the three pillars (Slaper & Hall, 2011).

However, today, the reality is that the success of CSR activities is often still based on financial measurements, hereby solely considering the demands of one stakeholder, the investor (Luo & Bhattacharya, 2006; Saeidi, Sofian, Saeidi, Saeidi, & Saaeidi, 2015). Nonetheless, as this presumably positive relationship has been difficult to explain there is a need to further consider the underlying predictors and outcomes of CSR (Aguinis & Glavas, 2012). Additionally, as previously mentioned, the increasing demand of consumers and employees regarding CSR should not be ignored, these stakeholder groups largely influence the number of CSR activities of a firm and determine what type of activities are considered to be salient (Aguinis & Glavas, 2012). Moreover, overlooking this growing need for CSR actions could lead to a decrease in customer loyalty and satisfaction and employee motivation and commitment, hereby harming the financial performance of a firm. However, by incorporating the triple bottom line approach, the total performance will be accurately determined, which entails that CSR activities and their influence are taken into account and thereby also the wishes of both external and internal stakeholders (Elkington, 1998). Therefore, by following the triple bottom line vision of Elkington, a firm will measure its performance based on these three pillars, and hereby the inclusion of environmental and social initiatives will become common practice. Furthermore, this vision enhances Friedman’s statement by extending the responsibilities of a firm from solely a financial responsibility towards the investors to a complete picture by taking social and environmental performance into account. Hereby, by pro-actively considering the demands of other stakeholders (i.e. employees and customers), firms will be able to take control over the situation and ensure a future positive relationship with these salient groups (Waddock, 2008). This, to not only satisfy investors but also additional internal and external stakeholders.

(12)

These stakeholder relationships are especially considered to be important as they contribute to a firm’s corporate reputation (Rupp, 2011; Waddock, 2008).

Corporate reputation

As the positive link between CSR and CFP has proven to be difficult to explain, there is a need to further define the context and circumstances in which CSR can positively influence a firm and its strategy (Bhattacharya & Sen, 2004; Luo & Bhattacharya, 2006). Therefore, to place this matter into a strategic context, the mediating effect of corporate reputation will be considered. Here, reputation is referred to as ‘‘a stakeholder's overall evaluation of a company

over time, based on the stakeholder's direct experiences with the company, any other form of

communication and symbolism that provides information about the firm's actions and/or a

comparison with the actions of other leading rivals’’ (Gotsi & Wilson, 2001, p. 29).

Moreover, in the case of CSR, reputation is often used as an argument to invest in these initiatives, as CSR engagement is believed to contribute to a positive image of a firm and brands (Porter & Kramer, 2006). Furthermore, corporate reputation is considered an essential intangible asset that managers should take into account when reviewing their CSR initiatives (Orlitzky et al., 2003). Although many have raised questions regarding the concrete quantitative contributions, intangible assets such as corporate reputation are more and more often associated with a sustainable superior performance (Porter & Kramer, 2006; Roberts & Dowling, 2002). Besides reputation, R&D and human capital are examples of other intangible assets that are possible mediators within this relationship (Branco & Rodrigues, 2006). However, compared to these other assets, reputation is not only influenced by CSR but it also influences both internal and external stakeholders. Moreover, besides this wide-spread influence, corporate reputation as an intangible asset is often mentioned as a possible element that could help clear up the current confusion around the tie between CSR and CFP (Margolis & Walsh,

(13)

2003; Orlitzky et al., 2003; Surroca, Tribo, & Waddock, 2010). Therefore, corporate reputation is in this case considered as the intangible asset that could aid in closing the literature gap with regards to questions raised concerning the complex CSR and CFP relationship.

To continue with the internal and external influences, there are numerous benefits of a positive corporate reputation that contribute to the financial performance of a firm. More specifically, a positive corporate reputation can shape stakeholder behavior as it not only convinces consumers to pay a higher price, increases their satisfaction and loyalty but it also increases the employer attractiveness in the eyes of prospective employees. Additionally, with regards to the financial performance, when a firm has a respected position within the industry it convinces future investors to acquire shares and enhances the loyalty and satisfaction of current investors (Creyer, 1997; Galbreath, 2009; McWilliams & Siegel, 2011; Pfarrer, Pollock, & Rindova, 2010; Siegel & Vitaliano, 2007). Additionally, reputation is perceived to be an important competitive advantage that is difficult to imitate and/or substitute (Siltaoja, 2006). Therefore, using corporate reputation as a non-traditional way of marketing makes it easier to reach the entire stakeholder network of a firm. Furthermore, the discussed benefits of having a positive reputation may also be the reason why corporate reputation is often considered to be a mediator within the CSR and CFP relationship. Therefore, the following paragraph will introduce the first hypothesis with regards to this mediating effect.

The mediating role of corporate reputation in the CSR and CFP relationship

With regards to the introduction statement of this thesis, a positive corporate financial performance is considered to be an important result to highlight the benefits of CSR to skeptic shareholders, especially in a time when firms’ CSR initiatives are faced with cynicism and criticism by investors. Moreover, shareholders are doubtful and follow Friedman’s statement by questioning the financial benefits and contributions of CSR activities to the firm’s reputation and competitive advantage (Arendt & Brettel, 2010). Additionally, the direct benefits of CSR

(14)

initiatives are often not easily visible or measurable which makes it difficult to substantiate the pro-CSR arguments to unconvinced shareholders (Aguinis & Glavas, 2012; McWilliams & Siegel, 2011). Furthermore, the competitive pressure and an unclear relationship between CSR and corporate firm performance (CFP) further contribute to the idea of irrelevance and dismissal of such activities (Burke & Logsdon, 1996). Visualizing an (indirect) positive relation between CSR activities and an increase in the return on assets (ROA) will, therefore, help in showing that Friedman’s way is not the only way of doing business and will demonstrate the relevance and value of corporate social responsibility.

Furthermore, utilizing CFP as the dependent variable is not only done because it has often been used as an important indicator of the success of CSR investments, but also as it is still very much valued by investors and as it is considered to be an important indicator of a firm’s long-term success (Aguinis & Glavas, 2012).

Moreover, previous research has indicated that corporate reputation is highly likely to be a mediator of the CSR and CFP relationship. More specifically, according to a meta-analysis performed in 2003 by Orlitzky et al., the majority of the studies show a positive relationship between corporate social performance and the financial performance of a firm, within this relationship, corporate reputation is often considered to be a key mediator.

Furthermore, Barnett & Salomon (2012) found a u-shaped relationship between CSR and ROA. They explain this relationship through the stakeholder influence capacity (SIC). This capacity determines a firm’s ability to influence the relationships with its stakeholders through CSR. Thus, if there is an adequate SIC, a firm is able to capitalize on their CSR activities which will result in a positive ROA, hereby emphasizing the importance of incorporating the needs of all stakeholders in strategic decisions.

Moreover, Dowling (2006) and Roberts & Dowling (2002) both argued that successful firms in possession of a relatively good reputation are more likely to sustain an exceptional

(15)

performance over time. This argument is supported by Iwu-Egwuonwu & Chibuike (2010) who found that, in the long term, reputation is not only a measurement used to determine the current value of a firm but due to its strategic nature it can also determine the future quality of the firm.

More specifically, as suggested by Luo & Bhattacharya (2006) and Saeidi et al. (2015) corporate reputation can act as a mediator between corporate social responsibility and corporate firm performance. a result that has also been found by Galbreath & Shum (2012). Moreover, Fombrun (2005) described a growing trend in CSR initiatives and argues that being an early adopter of the trend can benefit the reputation of the firm. Thus, reputation has proven to be, both theoretically and empirically, a large determinant of the success or failure of a firm (Dowling, 2006; Russo & Fouts, 1997; Worcester, 2009). Therefore, considering the internal and external influence of corporate reputation, the mediating effect it has on the CSR and CFP relationship and the strategic dimension it offers, corporate reputation has been chosen to serve as the mediator within this thesis. This conclusion has led to the following hypothesis:

Hypothesis 1: Corporate reputation mediates the relationship among

corporate social responsibility and corporate financial performance.

The effect of a CSR imbalance on corporate reputation

Moreover, this thesis investigates not only similarities but also differences to previous research done. Whereas other authors (e.g. Saeidi et al., 2015; Waddock & Graves, 1997) tried to explain the variation in outcomes through the use of a mediator (e.g. corporate reputation, firm size, competitive advantage, customer satisfaction) this research takes an additional step by further defining CSR into two types of activities (i.e. internal and external). Furthermore, making this distinction may result in finding other explanations for the variation in the above-mentioned relationship.

(16)

Moreover, following the example of Hawn & Ioannou (2016), the division in CSR activities is made by looking at the stakeholders of a firm. Internal stakeholders (i.e. employees) are confronted with internal CSR activities such as training, diversity, equal opportunity and all other initiatives relating to the overall well-being (Hameed, Riaz, Arain, & Farooq, 2016; Hawn & Ioannou, 2016). Moreover, external CSR activities such as corporate philanthropy, environmental research, and CSR reporting relate to external stakeholders (i.e. consumers, governments, shareholders, and suppliers) (Hawn & Ioannou, 2016).

Furthermore, making this division has proven to be meaningful, however, it is rarely done, let alone tested (Hawn & Ioannou, 2016; Schons & Steinmeier, 2016). However, recently, Hawn & Ioannou (2016), have managed to illustrate a small part of the performance heterogeneity of firms relating to CSR activities. In their article, they define the consequences of an imbalance between the two CSR types on the market value and public evaluation of the firm. As found, such a discrepancy could lead to a market value of 26,5% below the average, while a balance will result in a market value that is 36,5% higher than the average value (Hawn & Ioannou, 2016). Additionally, a firm risks to be accused of greenwashing in a situation where the number of external activities largely outweigh the internal CSR actions (Hawn & Ioannou, 2016).

Moreover, accusing a company of greenwashing suggests that the organization only engages in CSR to benefit their reputation and/or hide any missteps or indiscretions (Laufer, 2003). Especially as active communication with regards to green or social initiatives is increasing, the public more often questions the authenticity of these statements (Lyon & Montgomery, 2015). Furthermore, the phenomenon of greenwashing is mainly concerned with the communications around environmentally responsible activities that are executed by the related firm, thereby creating a highly positive image of the firm. These communications are

(17)

often misleading as they misrepresent the actual environmental and/or social performance of the company (Lyon & Montgomery, 2015).

In addition to results found by Hawn & Ioannou (2016), Walker & Wan (2012) also determined based on a sample of 103 top Canadian firms that when there is a discrepancy between the ‘green talk’ and ‘green walk’ it will negatively affect the financial performance. Furthermore, such an imbalance can be explained by the fact that only reporting on a firm’s CSR activities (green talk) without actively implementing them (green walk) is easily achieved and provides a certain level of flexibility. The attractiveness of greenwashing practices is therefore not surprising, as actual substantive CSR actions (internal activities) require high financial investments, while symbolic actions (external activities) are considered to quickly result in a boost of the corporate reputation, therefore it is an easy decision for a firm to make (Schons & Steinmeier, 2016).

Furthermore, an increasing gap between internal and external activities should be a growing concern for organizations as (perceived) greenwashing will be punished by both external and internal stakeholders. More specifically, when external stakeholders become aware of claims made externally that are not supported internally, the support for these actions will disappear and their loyalty will decrease. This course of events will then inevitably lead to a lower financial performance (Walker & Wan, 2012).

Furthermore, the negative effect of such a gap has an exponentially negative effect on financial performance as not only external but also internal stakeholders will respond to an asymmetry in CSR activities (Schons & Steinmeier, 2016). These internal stakeholders represent the party that is faced with the internal CSR actions of a firm and can, therefore, better determine which activities a company claims to implement, and which are actually implemented. This group is also highly involved in the actual implementation of these actions and has, therefore, more knowledge regarding the processes of an organization and the resulting

(18)

insights (Schons & Steinmeier, 2016). Therefore, in the case of a difference between the CSR claims and the execution, this may mean that employees decide to leave the firm and/or that productivity will be negatively affected due to the fact that employees do not support this way of working and are thereby unsatisfied with the CSR engagement of their firm (Schons & Steinmeier, 2016). These employee actions will directly impact the financial performance, this entails that companies should be aware of the consequences of their CSR activities. Thus, ‘talking without walking’ (i.e. external activities > internal activities) is highly discouraged based on previous literature.

Thus far only consequences of a CSR imbalance on CFP have been measured, however, in the case of corporate reputation, a similar result is also expected to take place, resulting in the next hypothesis. Even though, in the short term, external actions (i.e. claims/reports on CSR activities) may positively benefit the corporate reputation, in the long term, once the gap between internal and external activities will become too large it is no longer expected to positively contribute to the corporate reputation.

More specifically, external stakeholders will perceive publications on sustainable activities of organizations as something positive, however, if this position is not echoed internally and the external parties become aware of the imbalance, confidence and loyalty will be lost and the corporate reputation will be damaged. Moreover, a positive reputation was argued to contribute to the commitment and motivation of employees and thus their productivity (Boyd et al., 2010; Maignan & Ferrell, 2004). However, if employees become aware of a possible disbalance, this commitment and motivation will suffer, and a vicious cycle of lower productivity, employee turnover, and financial performance is expected to start and thereby affecting the corporate reputation. Therefore, the next hypothesis has been formulated:

Hypothesis 2: The bigger the imbalance between internal and external

(19)

The positive effect of external CSR activities on corporate reputation

As discussed in the paper of Holt & Barkemeyer (2012) and Aguinis & Glavas (2012) the media and research coverage of sustainability-related topics has increased throughout the last two decades. This growth in awareness and demand for more sustainable solutions has put pressure on corporations and has required them to provide an answer to these developed needs. Furthermore, as tested by Hawn & Ioannou (2012), when an organization increasingly reports on CSR investments, the public assumes that, from a trust point of view, actual internal CSR activities will also be executed. This belief exists even if organizations do not implement such internal actions. Additionally, this result is not surprising as the more a company reports and makes claims with regards to their CSR activities, the more the public feels confident in the organization that this way of working continues internally. Even if this is not the case, it is difficult for the public to find out what the actual internal CSR actions are.

Furthermore, as described previously, internal stakeholders are in a better position to evaluate the actual CSR situation of their firm as they have more insights regarding the internal and external activities of a firm. Schons & Steinmeier (2016), therefore conclude that it is difficult for external stakeholders to distinguish between external activities and internal activities, while internal stakeholders are better able to make this distinction. Furthermore, external CSR actions, therefore, show a positive relationship with financial performance when they are directed towards external stakeholders, while internal actions directed at these stakeholders do not impact the financial performance (Schons & Steinmeier, 2016). Moreover, the lack of awareness of external stakeholders regarding the internal activities of a company explains the absent effect on financial performance.

Moreover, this lack of sharing of internal activities could lead to the underestimation of a firm’s CSR activities. Additionally, inconsistency in the shared information will also add to the perception of greenwashing (Delmas & Burbano, 2011; Hawn & Ioannou, 2016). Therefore,

(20)

if a firm undertakes many internal CSR activities and does not externally communicate or report these initiatives it will damage the market value, thus if a company is seeking to get credit for their internal activities, in the form of reputational benefits, the organization should create awareness through active external communication of their CSR actions (Hawn & Ioannou, 2016).

Furthermore, considering previous research that has been done, most scholars define corporate reputation as something that has been build and formed by external stakeholders and external CSR activities (Fombrun & Shanley, 1990; Orlitzky et al., 2003). This implies that external CSR activities such as CSR reporting will have a bigger influence on a corporate reputation than internal CSR actions (i.e. the existence of a CSR committee or tangible policies on energy or water efficiency). Moreover, as the external stakeholder groups are significantly larger than the internal stakeholder group (i.e. employees) the impact of the former is expected to outweigh that of the latter. Therefore, as external CSR activities are often actively published and address the external stakeholders who largely influence a corporate reputation, the following hypothesis has been formulated:

Hypothesis 3: External CSR activities will have a bigger positive effect

on corporate reputation than internal CSR activities.

Moreover, based on previous literature, the literature gap has been defined. This gap forms the basis of the research question and the above-stated hypotheses. Furthermore, these have led to the following framework, presented in Figure 1. This framework acts as a visual representation of the relationships that will be tested between internal and external CSR activities, corporate reputation and corporate financial performance.

(21)

(22)

METHODOLOGY

Sample and data collection

To test these hypotheses, this thesis makes use of secondary quantitative data, gathered from DataStream. The sample consists of North American firms that are in the Standard & Poor’s index, the 500 companies on this list are all registered on the stock market of the United States of America. Moreover, this specific list gives an indication of the performance of the current U.S. economy and stock market (Ribeiro, 2017). Furthermore, the reliability of the list is illustrated by the fact that annually, an average of 23 changes are made to the list (4,6%) (Ribeiro, 2017).

Moreover, considering the availability of data on North-American firms within DataStream and as the majority of the firms in the reputation list of Fortune are from North-America, this sample facilitates in testing the relationships between CSR activities, corporate reputation, and financial performance. Additionally, the list of companies is designed in such a way that it represents all the big industries and therefore gives a good indication of the national economy (Bloomberg, 2018). Furthermore, with the use of the DataStream database, longitudinal data of these companies (2008 – 2014) could be accessed across a wide variety of subjects which will be discussed next.

This specific timeframe has been chosen as a significant amount of data has not been updated after 2014 and therefore, any data after this data point could not be considered as reliable. Moreover, data before 2008 was often incomplete and thus also excluded. Additionally, the list consists of 500 companies of which 74 are excluded because of inactivity. Therefore, based on the exclusion of invalid data and the size of the sample, the remaining 426 firms are believed to be an appropriate representation of the population.

(23)

Independent variables

Internal and external CSR activities. The independent variable in this research is corporate

social responsibility and data has been collected from the validated Asset 4 dataset (Hawn & Ioannou, 2016). This database includes data on different CSR attributes, such as corporate governance, social and environmental initiatives (Hawn & Ioannou, 2016). In here, data is combined by research analytics who gather data from sources that are publicly available such as company reports, websites, NGOs and news articles. Even though some of the elements have not been updated regularly, the majority of the information is updated on a continuous basis and key events (i.e. political, economic, environmental) are hereby taken into consideration (Hawn & Ioannou, 2016; Reuters, 2018; Schons & Steinmeier, 2016).

Moreover, the data in Asset 4 consists of a list of questions which are split into four pillars (i.e. corporate governance, economic, social or environmental). Based on this diverse set of questions, Hawn & Ioannou (2016) have selected 120 questions that represent both internal and external activities. Thereafter, these questions were discussed and labeled based on the external or internal orientation. This way of working led to a list of externally focused questions, mainly aimed to define the communication pattern of a firm, while the internally focused questions deal more with corporate governance and the implementation of CSR activities (Hawn & Ioannou, 2016). More specifically, examples of internally focused questions are: ‘What is the percentage of women on the boards of directors?’ or ‘Does the company develop products or technologies that are used for water treatment, purification, or that improve water-use efficiency?’. Moreover, examples of externally focwater-used questions are: ‘Does the company report about environmentally friendly or green sites or offices?’ or ‘Does the company claim to provide its employees with a pension fund, health care, or other insurance?’ (Hawn & Ioannou, 2016; Thomson Reuters, 2013)

(24)

After the list had been defined, the reliability was tested through Cronbach’s alpha and some questions were excluded based on this test. This, in the end, led to a final list of 21 internal CSR questions and 24 external CSR questions. In the end, the list of in total 45 questions has shown a good internal consistency and reliability (Cronbach’s alpha of 0,75 and 0,74 and inter-item covariance of 0,06 and 0,09) (Hawn & Ioannou, 2016).

Within the data collection of this thesis, this final list provided an excellent start off point. Therefore, the 45 questions on the list were extracted from Asset 4. Once gathered, it became clear that five questions relating to external actions needed to be excluded since there was a significant amount of data missing (>40%). Therefore, this research considers 19 questions that make up the external CSR activities and 21 questions relating to internal CSR activities which have led to a total number of 40 questions that are considered. This final list can be found in Appendix 1.

Moreover, the scores of these activities are collected over a period of seven years and are based on a Yes (= 1) or No (= 0) or defined by a percentage which is described as Low (= 0), at the cut-off point of < 50%, hereby representing the cumulative percentage, or High (= 1) if > 50%. This way of coding follows the example of Hawn & Ioannou (2016) who define it as a ‘robust and valid way’ to determine the exact value of the different CSR actions.

Once recoded, due to the size of the final list, a factor analysis was performed which will be discussed in more detail in the results section. Hereafter, the remaining questions and their relevant scores (i.e. 1 or 0), labeled as internal CSR, were added up. The sum that resulted from this calculation represents the total internal CSR score per year for each firm that is included in the sample. Moreover, for external CSR, the process was executed the same way and the sum of the scores for all remaining questions represents the total external CSR score per company. Therefore, after these steps, the list and results of 40 questions were narrowed

(25)

down to two separate scores, one representing internal CSR and the other representing external CSR.

Furthermore, this way of calculating the final variables also helped in determining the imbalance between the activities. In order to define the difference between internal and external CSR, a new variable needed to be created, Therefore, following the example of Hawn & Ioannou (2016), the internal CSR scores were subtracted from the external CSR scores, resulting in the gap between internal and external activities (Gap CSR).

Lastly, with regards to Hypothesis 1, the total CSR score of firms was required. Therefore, in order to be able to test the mediating effect of corporate reputation, the total CSR score was calculated by adding up the internal CSR and external CSR scores, resulting in another new variable (Total CSR).

Mediator

Reputation. The data for this mediator or dependent variable, depending on the hypothesis, has

been gathered from the validated and reliable Fortune’s America’s Most Admired Companies list (FAMA), in which fifty companies with the highest reputation are ranked (Cho & Pucik, 2005; C. Fombrun & Shanley, 1990; Fortune, 2018; Luo & Bhattacharya, 2006).

Annually, this is list is conducted by Fortune, together with the management consulting firm Korn Ferry, by defining the 1,000 biggest US companies based on revenue (>$10 billion). This number is then further reduced by determining the highest-revenues per industry (Fortune, 2018). Once this list is prepared, the remaining companies are rated by directors, executives, and analysts based on nine different criteria (i.e. CSR, ability to attract talent). Thereafter, each respondent chooses ten companies he/she admires the most and these responses constitute the final list. Within this thesis, the annual reputation lists of Fortune were collected over the years

(26)

2008 – 2014. Hereafter, the presence of the companies on the FAMA lists was compared to the companies in the sample. Then, presence (=1) or absence (=0) was marked and entered in SPSS.

Dependent variable

Corporate financial performance. The dependent variable in this research is measured based

on the return on assets (ROA) data of the 426 firms in the sample and is extracted from DataStream over a period of seven years. The ROA is a percentage that is often used in previous literature as an indicator of the financial performance of a firm (Barnett & Salomon, 2012; Waddock & Graves, 1997). The return on assets is calculated by dividing the net operating income through the average total assets and hereby indicates the profitability of the firm (Simpson & Kohers, 2002). Moreover, there is little agreement on what constitutes a good ROA, as what is considered to be a healthy ROA differs per industry and firm (Gallo, 2016). However, the percentage is expected to be positive in order for it to be labeled as healthy. Additionally, the higher the ROA the better, but as indicated, there is no specific benchmark.

Control variables

Firm size. In this case, the research will consider two control variables, the first being firm size.

Data has been gathered from DataStream and the variable is measured based on the number of employees. As defined by the government of the United States, a firm with > 1,000 employees is considered to be a large firm and is therefore coded with a 1 (<1.000 employees = 0) (SBA, 2018).

Moreover, firm size is often utilized as a control variable when assessing corporate financial performance (Aguinis & Glavas, 2012; Waddock & Graves, 1997). In this case, firm size matters because large firms have the possibility to make use of economies of scope and/or

(27)

scale, have a higher efficiency and easier access to capital than smaller firms (Stierwald, 2009). Therefore, firm size is an important variable to control for as it could have an effect on CFP.

Furthermore, when considering corporate reputation as a dependent variable, previous research found that bigger firms have better reputations (Brammer & Pavelin, 2006; Delgado García, De Quevedo-Puente, & De La Fuente-Sabaté, 2010; Fombrun & Shanley, 1990). More specifically, the explanation given for this is that the bigger the company the more they will be in the public eye and more stakeholders will pay attention to them. Therefore, exhibiting (im)proper behavior will more quickly lead to (un)favorable corporate reputation. On the other hand, when a company is small, the audience (i.e. stakeholder groups) will also be smaller and less critical of their actions and therefore significant positive or negative changes in corporate reputation are less likely (Delgado García et al., 2010).

Industry. With regards to industry as a control variable, this thesis follows the definitions of the

S&P index (Kennon, 2018). This index specifies ten business areas which were all dummy coded in order to facilitate the inclusion in the SPSS dataset.

In the case of corporate financial performance, industry is an important control variable and has shown to explain a significant percentage of the profit variation across firms (Margolis & Walsh, 2003; McWilliams & Siegel, 2000; Russo & Fouts, 1997). This is explained by Porter’s (1980) five forces model, which states that the industry in which a firm operates affects the profitability. The forces within each industry differ and thereby the intensity of competition. When the forces and thus competition is intense the chances of earning high returns are low. Furthermore, if the forces are mild, the chances of a higher firm profitability increase.

Moreover, each industry has different stakeholders, therefore, firms experience different kinds of pressures in each industry, the response to these pressures differs and the effect of these pressures on corporate reputation will also differ per industry (Brammer & Pavelin, 2006; Melo

(28)

RESULTS

Within this section, the results of this thesis are shared. Firstly, the model specification and data preparation steps are reported, thereafter, the descriptive statistics and the corresponding correlations are discussed. Lastly, the regression analyses are addressed which have tested the before stated hypotheses.

Statistical methods

Moreover, as the sample consists of observations made over a period of seven years, these repeated measures needed to be taken into account when defining the statistical steps. Therefore, in order to test the first hypotheses, a regression analysis needed to be conducted. More specifically, with the use of multilevel modeling or mixed models in SPSS these analyses could be performed while also considering the fact that the data was collected over multiple years (Field, 2009). However, in order to properly execute these models, the data structure needed to be adjusted. Hereby, the data points were transformed from a wide format to a long format and years, Gap CSR and Total CSR were added as variables.

Additionally, two variables, namely ROA and corporate reputation, needed to be lagged. Lagging these variables was necessary for this because CSR activities happening in year 1 are expected to affect the inclusion or exclusion of a firm on the corporate reputation list. Additionally, for ROA the same effect was expected as a change in corporate reputation in year 1 will impact the corporate financial performance of the year after. Hereby, the true measurement of the effect of CSR on corporate reputation and the effect of corporate reputation on CFP is ensured. Hereafter, the years were considered as the repeated measures in SPSS and the hypotheses were tested, of which the results will be shared later on in this chapter.

For the first hypothesis, due to the dichotomous nature of the corporate reputation variable, it was not possible to use the PROCESS program of Hayes (2013). Therefore, this

(29)

mediation was tested with the use of a regression analysis. In this case, multiple regression analyses were performed in steps, and within each of these analyses, the significance of the coefficients was tested. Through this process, the mediating role of corporate reputation within the CSR and CFP relationship could be determined. The detailed process of these analyses can be found in the results chapter.

Furthermore, hypothesis 2 and 3 were tested with the mixed models as well. The regression model belonging to hypothesis 2 demonstrates what the effect is of an imbalance between CSR activities on the lagged corporate reputation when controlling for firm size and industry. The results of the model of hypothesis 3 indicate if external CSR activities have a bigger positive effect on corporate reputation than internal CSR activities.

Analytical strategy

Moreover, once the data had been collected with the use of Asset 4 and DataStream, it was entered in SPSS. Thereafter, the following preparatory steps and statistical analyses were performed in this program. Firstly, the frequency tables were created to determine if there were any errors in the data. As this was not the case, missing values were considered. Furthermore, as the sample consists of data points of 426 firms gathered over multiple years, some values were missing and these have been excluded pairwise. Excluding these values pairwise ensures that only the missing values are excluded and that no firm is excluded from other analyses in case it has missing values in one variable.

Factor analysis. As discussed in the methodology chapter, after collecting the questions of the

pre-defined list of Hawn & Ioannou (2016), there were still 40 different CSR questions included. Therefore, the next step was to conduct a factor analysis to help in the reduction and simplification of the dataset. Moreover, for internal CSR activities, based on the Keiser's

(30)

criteria, a scree plot and KMO (> 0.6) and Bartlett’s test (p < 0.5), seven factors were extracted. Then, for external activities the same steps were executed, where the requirements for the KMO and Bartlett’s test were met as well, which led to the extraction of six factors. These results can also be found in appendix 2 and 3.

Assumptions

With regards to the regression analysis, there are certain assumptions that need to be taken into account to ensure that conclusions can be drawn about the population (de Vocht, 2017; Field, 2009).

Firstly, all variables within the data set have either an interval or ratio scale. Moreover, they are all quantitative and as allowed by the assumptions, the independent variables have been recoded into dummies. Moreover, all predictor variables meet the non-zero variance assumption which dictates that the values of these variables should show some variations and cannot all have the same value.

Secondly, there should not be an exact linear relationship between any of the independent variables within the research. In order to test this assumption, a multicollinearity test was executed and following the rule of thumb, the VIF scores were well below 10, thereby indicating that there is no multicollinearity. Additionally, as can be seen in table 1, none of the independent variable correlations are above .90, except those of internal/external CSR and Total CSR, which is not surprising as these first two are part of the total CSR variable. Therefore, this assumption is met as well.

Thirdly, no clear signs of heteroscedasticity have been found based on scatterplots of the independent and dependent variables. Therefore, the homoscedasticity assumption is also met which means that the variance of the residuals is constant.

(31)

Lastly, as mentioned before, once the variables had been recoded, the descriptive statistics such as mean, standard deviation, kurtosis, and skewness were set up. Based on the skewness, Q-Q plots, kurtosis, and histograms, most of the variables appeared to be normally distributed, with the exception of corporate reputation and firm size. The absence of a normal distribution of corporate reputation and firm size can be explained by the fact that these variables have been recoded into a 0 or 1, therefore a normal distribution can no longer appear.

Descriptive statistics and correlations

The following table (1) shows the means, standard deviations and the correlations of the variables. The correlation analysis tests the strength and the direction of the relationship between variables. Firstly, table 1 tells us that all the variables except for industry are significantly correlated with corporate reputation. When the rules of thumb by Cohen (1988) are taken into account we see a large positive effect (r = .72) between firm size and corporate reputation, indicating that the bigger a firm, the better the corporate reputation. Moreover, the correlation between corporate reputation and CSR gap shows to be negative (r = -.07), and even though it is a small effect, it does indicate that an increase in the CSR gap has a negative effect on corporate reputation. Moreover, when we look at the ROA, firm size and industry show small correlations with the dependent variable. Even though the effects are small, it does indicate the importance of controlling for these variables.

Furthermore, the correlations with a large effect are also interesting, which is the relationship between Gap CSR and internal CSR (r = -.69), which indicates that an increase in internal CSR activities will lead to a decrease in the CSR gap, which is considered to be a good thing as it leads to a smaller imbalance. Moreover, external CSR also shows this negative relationship, however with a smaller effect (r= -.17). Two other positive correlations are those

(32)

between internal (r= 0.13) or external CSR (r = 0.06) and ROA, which shows that an increase in CSR activities will lead to a better return on assets (i.e. corporate financial performance).

Lastly, internal and external CSR activities are correlated, the effect (r=.08) is small, however, it does indicate that when a firm takes action in one of the activities, it will positively affect the activities in the other type.

Table 1. Means, standard deviations and correlations

Variable M S.D. 1 2 3 4 5 6 7 8 1. Reputation 0.08 0.27 - 2. Industry 4.63 2.04 -.02 - 3. Firm size 0.94 0.23 .72** _.03 _- 4. Internal CSR 3.95 2.54 .25** _-.03 _.22** _- 5. External CSR 2.57 1.86 .27 ** _.02 _.25** _.8** _- 6. Total CSR 6.53 4.18 .27**_{-.005 .24}** _.95** _.93** _- 7. CSR gap -1.38 1.55 -.07** _.04*_-.10**_-.69**_-.17**_-.48** _- 8. ROA 7.10 7.29 .07** _-.04*_.10** _.13** _.06** _.11**_-.11** _-

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

Regression results

When testing the first hypothesis, regression analysis was used to take the first steps toward determining the mediating effect of corporate reputation. In total, four steps were taken, these are visualized below, in figure 2, and the results are presented in table 2.

(33)

Firstly, in all the steps, firm size and industry were controlled for, this led in the first step to a significant positive relationship between total CSR (i.e. internal plus external CSR) and corporate financial performance (a = .0745, p = .018), indicating that an increase in total CSR has a positive effect on CFP. Moreover, total CSR also shows to have a positive significant effect on corporate reputation (b = .0153, p = .000). Additionally, if there is a positive increase in corporate reputation, the CFP of a firm will positively benefit, in this case, the coefficient is large (c = 1,35, p = .004).

Furthermore, the fourth step shows the mediating effect of corporate reputation within this framework, as when corporate reputation is added to the analysis, total CSR is no longer significantly related to corporate financial performance. Hereby, the results in table 2 show support for hypothesis 1. Lastly, with the help of an online calculator, the confidence interval of the indirect effect was calculated (Biesanz, Falk, & Savalei, 2010). This resulted in a confidence interval of CI = 0.010; 0.231, as these values are both above zero, this also indicates that the CSR activities a firm undertakes, positively influence the corporate reputation, which in turn highly contributes to the corporate financial performance.

(34)

Table 2. Mediating effect of corporate reputation in the CSR and CFP relationship

Coefficient S.E. Sign 95% C.I.

Step 1 - a Firm size 1,318607 0,607907 <.05 0,126551 2,510662 Industry -0,220282 0,063154 - -0,344122 -0,096442 Total CSR 0,074598 0,031622 <.05 0,012591 0,136606 Step 2 - b Firm size -0,001233 0,023742 - -0,047786 0,045321 Industry -0,002905 0,002471 - -0,007749 0,00194 Total CSR 0,015361 0,001235 <.001 0,012939 0,017782 Step 3 - c Firm size 1,561406 0,591675 <.05 0,40118 2,721632 Industry -0,213276 0,063126 <.05 -0,337059 -0,089492 Corporate Reputation 1,352723 0,465977 <.05 0,438976 2,266469 Step 4 Firm size 1,320359 0,607303 <.05 0,129487 2,511231 Industry -0,215198 0,063127 <.05 -0,338984 -0,091411 Total CSR 0,05761 0,032324 - -0,005775 0,120996 Corporate reputation 1,172456 0,47675 <.05 0,23758 2,107331

Moreover, when testing hypothesis 2, the internal CSR activities were subtracted from the external CSR activities to determine the discrepancy between the activities. This new variable, CSR gap, was then placed in the regression model and shows a significant relationship with corporate reputation. Furthermore, as can be found in table 3, the coefficient is small and negative (b = -.0103, p < .01), indicating that there is a negative relationship between an imbalance in CSR activities and the corporate reputation. This result indicates that hypothesis 2 is supported when firm size and industry are controlled for.

(35)

With regards to hypothesis 3, the independent variables of internal CSR and external CSR were entered into the regression model. In table 3, the results indicate that, although a small effect, external CSR activities is highly significant with corporate reputation (b= .0290, p < .001). This implies that the more a firm engages in external CSR activities, the more corporate reputation will positively increase. On the other hand, the relation between internal CSR and corporate reputation is non-significant and therefore, internal activities do not affect corporate reputation. These results thus indicate that hypothesis 3 is supported and external CSR activities have a bigger positive effect on corporate reputation than internal CSR.

Table 3. Regression analysis for CSR activities and corporate reputation Independent and control

variables Coefficients (standard errors)

Industry -.0033 -.0021 (.0024) (,0025) Firm Size -.0028 .0596* (.0237) (.0237) Internal CSR .0059 (.00324) External CSR .0290*** (.0045) CSR Gap -.0103** (.0033) * p<0.05, ** p<0.01, *** p<0.001.

(36)

DISCUSSION AND CONCLUSIONS

This thesis has been created to define how internal and external CSR activities affect the corporate reputation. After the statistical analyses, it has become clear that the findings have found support for the suggested effects. Moreover, the following paragraphs will place these findings in a theoretical and practical context. Hereby, the limitations will also be addressed, some of which create opportunities for future research.

Main findings

Until today, Friedman’s statement can still count on support, especially from investors who highly value the creation of profits, but the question is whether it is still a valid statement to make? Should firms solely contribute something to their stakeholders in a financial way? Are profits really the only social responsibility of a firm? Following the results of this thesis, the answer to these questions is no. More specifically, based on previous research, recent trends and the findings of this research, stakeholders of firms are increasingly looking for additional benefits of a firm’s operations, namely, in an environmental and socially friendly manner. Moreover, as consumers and employees become more aware of greenwashing practice, this growing attention towards a firm’s overall performance puts pressure on the organizational activities but it also puts the topic higher on the strategic agenda (Aguinis & Glavas, 2012; Lyon & Montgomery, 2015). However, entering CSR into the strategy domain and perceiving it as a possible competitive advantage, implies that investing in these activities will lead to positive financial results (Arendt & Brettel, 2010; Siltaoja, 2006).

Moreover, this is where the ambiguous relationship begins, as the reason for the positive relationship between CSR and CFP remains uncertain (Aguinis & Glavas, 2012). Furthermore, due to this complex relationship, previous literature suggests to research underlying mechanisms that mediate the relationship, therefore corporate reputation has been included in

(37)

this research as the mediator of this relationship (Aguinis & Glavas, 2012; Luo & Bhattacharya, 2006). Hereby, this thesis has found additional empirical evidence for the positive effect of CSR on CFP and the positive mediating effect of corporate reputation within this relationship. As expected, this relationship shows an indirect effect and CSR contributes to the financial performance of a firm through corporate reputation. Therefore, support has been found for the first hypothesis.

Furthermore, as established by Hawn & Ioannou (2016) and Schons & Steinmeier (2016), going one step further into these steps may further help in the demystification of this relationship. Within their studies, the positive effect of external CSR activities on CFP has been established. Additionally, these scholars both emphasized the importance of maintaining balanced CSR activities. While these articles focused on the effect on financial performance, this thesis builds on these results by replacing CFP for corporate reputation.

Hereby, the results show a similar effect as external CSR activities have a positive and significant effect on corporate reputation, while internal activities do not show such an effect. Therefore, support was also found for the second hypothesis.

Continuing on the division of CSR activities into internal and external, the remaining results of this study are also similar to those of Hawn & Ioannou (2016) and Schons & Steinmeier (2016). In their case, a bigger gap in CSR activities negatively affects the corporate financial performance. Moreover, based on the findings of the regression analyses in this study, a gap in CSR activities has a negative effect on corporate reputation. This result implies that the bigger the imbalance, the worse the effect on corporate reputation will be. Therefore, when a firm invests more resources in its external activities than in its internal activities, the corporate reputation will suffer in the long run.

Following these results, it can be concluded that all predications made in the beginning of this thesis have been supported based on the empirical evidence that has been found. Hereby,

(38)

a final answer to the research question can be formulated, which will be done after the following contributions have been shared.

Theoretical contributions

When diving into these main findings and the ambiguous relationship between CSR and CFP, it is important to go further than the first level of analysis. However, up until today, many scholars have solely considered this top layer of the complex initiative by perceiving CSR as one activity. Moreover, through dissecting this relationship by defining and specifying CSR into two types of activities, the role that CSR plays will be better understood.

Hereby, this thesis advances the current literature on this topic and follows the example of Hawn & Ioannou (2016) and Schons & Steinmeier (2016) who have illustrated the valuable effect of taking this additional step in a firm’s CSR analysis. Also, as the current literature lacks a decisive answer to the question of whether CSR does or does not contribute to CFP, this field of research has proven to contribute to this inconclusive discussion.

Moreover, the found results again indicate the positive significant relationship between CSR and CFP and the mediating effect of corporate reputation, this thesis therefore provides support for the argument that claims that CSR activities can contribute to the financial performance of a firm. More specifically, these conclusions add to the existing evidence that investing in CSR activities is indeed a profitable and strategically smart decision to make. Moreover, now that this result has been established, from a firm perspective, it is interesting to go one level deeper and specify how CSR exactly influences the reputation of a company. Furthermore, this positive effect of CSR on corporate reputation and the creation of a positive image of a firm was already established, however, which specific part of CSR ensured this positive relationship remained unclear (Orlitzky, Schmidt, & Rynes, 2003; Porter & Kramer, 2006).

Juggling corporate social responsibility : how internal and external corporate social responsibility activities, and the balance between them, affect corporate reputation