Corporate Social Responsibility and Labour Productivity in the Context of Individualistic and Collectivistic Cultures: Evidence from the Chemical Industry in the United States, Japan, Germany, and South Korea

(1)

Corporate Social Responsibility and Labour

Productivity in the Context of Individualistic and

Collectivistic Cultures:

Evidence from the Chemical Industry in the United

States, Japan, Germany, and South Korea

Master Thesis

Felix F. Bender (S 2720493)

Supervisor: Paulo Marques Morgado

Study Program: International Business and Management

Research Question:

“Is the relationship between Corporate Social Responsibility and labour

productivity influenced by cultural context?”

(2)

Master International Business and Management (IB&M) 2014/2015

Title: Corporate Social Responsibility and Labour Productivity in the Context of

Individualistic and Collectivistic Cultures: Evidence from the Chemical Industry in Germany, the United States, South Korea and Japan

Author: Felix Frank Bender

Author’s e-mail address: Flx.Bender@gmail.com

Author’s address: Coendersweg 4a, 9722 GE Groningen, The Netherlands Department: Faculty of Economics and Business

Supervisor: Paulo Marques Morgado

(3)

1. Introduction

The concept of Corporate Social Responsibility (CSR) is not a new notion within business studies and business research. Nowadays, a growing number of companies are publishing their CSR efforts along with their financial report. This development was facilitated through reporting frameworks like the Global Reporting Initiative and the Integrated Reporting framework, which are both used by chemical companies like BASF (BASF, 2015). Through events like the Deepwater Horizon oil spill in the Gulf of Mexico in 2010 or the collapse of a garment-factory in 2013 in Savar, Bangladesh, CSR has further gained public attention. This development is supported by the growing media coverage of CSR related issues (Lee and Carroll, 2011).

Since Howard R. Bowen’s (1953) “Social Responsibility of the Businessmen” was published, a growing number of scholars are not only concerned with the definition of CSR, but also with the development of CSR frameworks (Caroll, 1999; Moser, 1986; Windsor, 2001). However, empirical research regarding CSR, namely the influence of CSR on competitiveness, is still scarce. This paper tries to contribute to this field of research by investigating the influence of CSR on labour productivity, focusing on possible differences of this influence in individualistic and collectivistic cultures. This relationship will be examined by studying companies in the chemical industry within the United States, Japan, Germany, and South Korea. The GLOBE study by House et al (2004) will be used throughout this paper to define the cultural characteristics of the investigated sample.

Thus, this paper aims to further contribute to the research examining the influence of informal institutions and culture on business practices. Therefore, the research problem that is investigated in this paper is the connection between CSR and labour productivity and the influence of cultural context on this relationship. The research question that derives from this problem is phrased as “Is the relationship between Corporate Social Responsibility and labour productivity influenced by cultural context?”.

(5)

The first part of this paper will discuss the theoretical foundation of this thesis by defining the term CSR as it is used in this paper. This will be done by reviewing existing literature that is concerned with the influence of CSR on firm performance, and by explaining the theoretical link between CSR and labour productivity, as well as possible influences of culture on this link. After reviewing the existing contributions, the second part of the paper will develop the hypotheses that are investigated within the present study. On the basis of these hypotheses, the framework that is used in this paper will be presented, by showing possible linkages and influential factors in a conceptual model. The methodology and sample that are used to test the conceptual model with its underlying hypotheses will be discussed in the subsequent methodology section. After the presentation of the results, the final chapters will summarize the value of this thesis with a concluding discussion, the limitations of this thesis, and suggestions for future research.

2. Literature Review

2. 1 Definition of CSR

Since the term CSR includes numerous aspects and there are a number of possible definitions of this term, the definitional boundaries of CSR are broad. Thus, it is necessary to discuss the definition of CSR as it will be used in this paper. A general accepted definition of CSR, which is used by numerous scholars that have empirically tested the relationship between CSR and competitiveness (Burchielle et al., 2009; Sánchez and Benito-Hernández, 2013; Sun and Stuebs, 2010, 2013; Wu and Shen, 2013), is provided by Vilanova et al. (2009), who group CSR issues into 5 dimensions, namely

• Vision, including issues like governance, corporate values and organizational structure, • Community relations, including issues like community measures and the partnership

with stakeholders,

• Workplace, including issues related to human rights and labour practices,

• Accountability, including issues related to transparency and communication, and • Marketplace, which is related to core business practices from a CSR perspective. Furthermore, the European Commission (2002) defines CSR as “a concept whereby companies decide voluntarily to contribute to a better society and a cleaner environment”.

(6)

provide further evidence for their applicability to the chemical industry. Further evidence for the applicability of these definitions can be derived from Verboven (2011), who investigates the communication of CSR within the chemical industry, and who also refers to both the European Commission (2002) and Vilanavo et al. (2009). It is important to mention that these definitions do not incorporate all possible aspects of CSR and are thus not exhaustive; however for the purpose of this paper, these definitions can be regarded as adequate.

2. 2 The business case for CSR

The previous definition has clarified the question of which aspects are incorporated into the term CSR. However, it is also important to clarify why companies engage in CSR activities. Within the discussion of CSR, scholars have made the distinction between a normative and a business case for CSR. The normative case suggests that engaging in CSR activities is a moral obligation. Thus, companies have to act as “good citizens” and have to do the “right things” (Branco and Rodrigues, 2006). This view was questioned by Milton Friedman (1970), who stated that “there is one and only one social responsibility of business - to use its resources and engage in activities designed to increase its profits […]”.

This objection to the normative case of CSR leads to the question, whether engaging in CSR activities can have a positive impact on profits and financial performance (Carroll and Shabana, 2010). To answer this question, scholars have investigated the business case for CSR. Kurucz et al. (2008) have grouped the possible benefits of CSR engagement into four categories:

1. Cost and risk reduction through proactive social and environmental engagement, 2. Gaining competitive advantage,

3. Developing reputation and legitimacy of business operations, 4. Creating win-win situations that are beneficial for all stakeholders.

Thus, the business case for CSR suggests that CSR can be a source of competitiveness and a competitive advantage. In order to understand the causal relationships between CSR and competitiveness, it is important to gain a general understanding of how companies can create a competitive advantage and increase their competitiveness.

2. 3 Competition and Competitive advantage

(7)

(1979, 1985, 1990) has defined competition and competitiveness by considering a company’s home-market, the influential forces that shape a company’s industry competition, and the company’s value chain as a source of competitive advantage.

Porter (1990) argues, that “the only meaningful concept of competitiveness at the national level is productivity”. Thus, companies have to constantly improve productivity by increasing the quality of products, improving technology, and by increasing production efficiency. As a result, “companies achieve a competitive advantage through acts of innovation” (Porter, 1990). Consequently, a nation’s competitive advantage is determined by its ability to foster innovation within its companies. Porter (1990) describes 4 determinants of a nation’s competitive advantage:

1. Factor conditions: Insufficient availability of production factors in the home country forces companies to create these resources. Thus, it is not only important to deploy the existing resources, but rather to create resources efficiently. A lack of production factors can therefore lead to innovation, as well as innovative solutions for the creation of these production factors.

2. Demand conditions: Sophisticated and highly demanding customers in the home market provide a company with a clear picture of emerging buyer needs. If the company is forced to produce products that meet these high demands, it is also forced to be innovative. Buyer needs that are present and emerging within the home country are also likely to emerge in other countries. Consequently, a company that can meet buyer needs in a sophisticated home market is also likely to have a competitive advantage when expanding to other markets.

3. Related and supporting industries: Close proximity to suppliers can foster efficient communication and cooperation. However, it is important that companies work together with their suppliers in order to create a competitive advantage. It is also crucial that these suppliers are not only dependent on the home market, but are competing on a global level, since this global competition creates the need for innovative solutions. 4. Firm strategy, structure, and rivalry: Rivalry within a home country market supports

innovation because strong competitors force companies to come up with new solutions in order to stay in business. This can translate to a strong competitive advantage on the global level, where competition is likely to be even stronger.

(8)

sources of a competitive advantage. For this purpose, Porter (1985) introduces the value chain: “The value chain disaggregates a firm into its strategically relevant activities in order to understand the behavior of costs and the existing and potential sources of differentiation” (Porter, 1985). Within the company, Porter (1985) identifies cost reduction, focusing on a consumer segment, and differentiation as sources for competitive advantage. Companies that fail to identify these sources of competitive advantage or fail to pursue strategies that make use of this potential, have a disadvantage relative to their competitors, who are more successful in identifying these sources and translating them into a successful strategy.

As a conclusion, a company’s home market, as well as the activities within the company can be a source of a competitive advantage. Using these concepts, it is possible to explain how a successful CSR strategy can help companies to improve their performance and to create a competitive advantage.

2. 4 CSR as a source of Competitive Advantage

In their article “Strategy & Society”, Porter and Kramer (2006) show the link between CSR and competitive advantage, and suggest that “CSR can be much more than a cost, a constrain, or a charitable deed – it can be a source of opportunity, innovation, and competitive advantage”. In order to identify these sources of competitive advantage, the social dimensions of the competitive context, in which a company is embedded, as well as the social impact of its value chain have to be examined. The social issues that can be identified can then be prioritized according to their impact on the company’s business. Porter and Kramer (2006) group these issues into three categories:

1. Generic Social Issues are important social issues for society, but have no influence on a company’s business or its competitiveness.

2. Value Chain Social Impacts are social issues that are influenced by the company’s activities along the value chain.

3. Social Dimensions of Competitive Context are social issues that influence the competitive context, in which the company operates.

(9)

It is important to note that according to the business case for CSR, social activities can both, profit society and improve the financial performance of a company in the long-term. By investing in society, a company invests in, and improves its competitive context. For example, if a company identifies the lack of education as a social issue, it can invest to solve this issue by improving education through monetary means. In the long-term, this will benefit the company by providing necessary production factors in the form of human capital. Consequently, “the success of the company and the success of the community become mutually reinforcing” (Porter and Kramer, 2006). Engaging in CSR activities that are targeted at employees can also lead to a competitive advantage. A company that has employed a policy to promote the diversity of employees can ultimately have access to a greater talent pool than a company that does not promote and support diversity (Carrol and Shabana, 2010). According to a Fortune (2003) survey, business executives also agree that gaining a competitive advantage is one of the two most important benefits of CSR.

In the same way, identifying and solving social issues along the supply chain can translate into a competitive advantage. For example, inefficient consumption of energy translates into both cost disadvantages and environmental damage. By increasing the efficiency of energy consumption, a company can serve both its own business performance as well as the society in which it operates.

(10)

2. 5 CSR as a source of Competitive Advantage: Empiric Verification

The relationship between CSR and competitiveness has only been tested very rarely within business research. Vilanova et al. (2009) have constructed a framework that can be used to evaluate the connection between CSR and competitiveness. In their analysis of the European financial sector, they were able to find an indirect link between CSR and competitiveness: they state that “adopting a CSR strategy has an impact on identity and branding, which has a direct impact on competitiveness as it forces sustainable development and in corporate vision through corporate strategy” (Vilanova et al., 2009). In line with Porter and Kramer (2006), Vilanova et al. (2009) have identified the implementation of a CSR strategy itself as an indirect connection and source of competitiveness: in order to implement a CSR strategy, firms have to restructure the internal business processes and initiate sustainable and innovative thinking in their employees. Through “exploring, documenting, and determining success factors; understanding competencies and awareness and grasping the policy framework” (Vilanova et al., 2009), companies start innovative and dynamic thinking within their organization. Since “some of the key determinants to firm competitiveness center on issues such as brand equity, reputation or innovation” (Vilanova et al., 2009), this process ultimately increases firm competitiveness and leads to a competitive advantage.

Based on the framework developed by Vilanova et al. (2009), Sun and Stuebs (2012) and Sánchez and Benito-Hernández (2013) have investigated the effects of CSR engagement on firm and labour productivity, respectively. Sun and Stuebs (2013) “results support a positive relationship between CSR performance and firm productivity”. They explain this relationship by the learning cycle that takes place when a company implements CSR practices. Thus, learning takes place, when integrating a CSR strategy into the business process. Through this learning practice, innovative ideas will be generated, which ultimately lead to increased competitiveness. Sánchez and Benito-Hernández (2013) found further evidence that “CSR actions derived from a commitment to quality in internal operational processes had a positive relationship with labour productivity”. However in their study of Spanish small and micro manufacturing firms, they could not find support for the hypothesized positive relationship between independent variables like external relationships, relationships with the community, environmental investment, and staff training and their dependent variable, labour productivity (Sánchez and Benito Hernández, 2013).

(11)

manufactures of consumer goods, Kim et al. (2010) focused their attention on the influence of CSR association and participation on employee commitment and employee company identification. The researchers found a positive relationship between CSR association and the perceived external prestige. Furthermore, they also found a positive relationship between perceived external prestige and employee-company identification. Thus, employees consider that their company is perceived positive by outsiders (measured through perceived external prestige) and therefore increase their identification and commitment with their company (Kim et al., 2010). These findings are in line with Porter and Kramer (2006), who remark that “employees derive justifiable pride form their company’s positive involvement in the community”. Moreover, Kim et al. (2010) find a positive relationship between CSR participation and employee-company identification. This underscores the “importance of planning CSR initiatives that can intensify employee-company identification” (Kim et al., 2010).

Using the framework of Vilanova et al. (2009), Sun and Stuebs (2010), investigated the relationship between reputation, labour efficiency and labour productivity. Comparing a sample of highly reputable U.S. firms to a matched sample of corresponding firms, the researchers can find a positive influence of reputation on both labour efficiency and labour productivity. They explain this connection by the increased popularity of reputable firms among talented and skilled employees. Furthermore, “these employees will work harder for firms with higher reputation” (Sun and Stuebs, 2010). These findings can be seen as complementary to the findings of Kim et al. (2010), who used perceived external prestige as a measure of reputation and who also found evidence for the connection to employee commitment.

Although the empirical verification of CSR as a source of competitive advantage is still rare, the connection between CSR and corporate financial performance has been tested more extensively. In their review of the empirical work that has investigated the business case for CSR, Carroll and Shabana (2010), come to the conclusion that “on the whole, CSP [Corporate Social Performance] – CFP [Corporate Financial Performance] research seems to indicate the existence of a positive relationship between CSP and CFP […]”. However, they also conclude that in order to understand the underlying process between this connection, the mediating and moderating effects that lead to this connection need further investigation (Carroll and Shabana, 2010).

(12)

labour productivity. It can be further concluded that reputational gains, as well as the implementation of a CSR strategy that includes the identification of shared value, have a mediating effect on the relationship between CSR and labour productivity.

2. 6 CSR in the Chemical Industry

As mentioned before, this paper will investigate the relationship between CSR and competitiveness by focusing on the chemical industry in the United States, Japan, Germany, and South Korea. In their study of the influence of corporate social responsibility on firm productivity, Sun and Stuebs (2013) suggest “that corporate social responsibility (CSR) has become an important part of business practice in the chemical industry”.

In order to assure the importance of the investigated CSR aspects within the chemical industry, the CSR reports of companies that are included in the sample are examined for statements regarding community and employee engagement:

• In their annual corporate report, which includes the financial report as well as the CSR report, BASF SE states: “As the world’s leading chemical company, we combine economic success with environmental protection and social responsibility” (BASF, 2015). BASE SE further states that the company cares about social responsibility: “We are involved in diverse projects worldwide, especially in the communities where our sites are located” (BASF, 2015).

• The Dow Chemical Co. publishes its sustainability reports every quarter. The company states that its “focus at every site was on the same objectives: addressing the most relevant needs of the community while developing and fostering new relationships to nurture community success” (Dow Chemical, 2015a). On its website, the company further states: “Dow people worldwide directly apply their passion and expertise to advance the well-being of people and the planet” (Dow Chemical, 2015b). • The Japanese chemical company Mitsui Chemicals underlines the importance of employee engagement in its CSR report and states that it runs “a range of support programs in order to suit the working environments and lifestyles of individual employees” (Mitsu Chemicals, 2015). The company further states that it is committed to work “in harmony with local communities on a range of initiatives” (Mitsu Chemicals, 2015).

(13)

creativity and autonomy, prioritize[s] each individual’s capabilities, and guarantee[s] employees the treatment suitable for their performance” (LG Chem, 2014). The company further states that “LG Chem is committed to becoming a socially responsible company that coexists and grows with society” (LG Chem, 2014).

It is important to mention that the purpose of this overview is not to provide an exhaustive analysis of the understanding of CSR in the chemical industry. However, it provides anecdotal evidence to support the notion that the understanding of CSR in the chemical industry that is used in this study is in line with the view of the businesses that are contained in the sample. Both community, as well as employee aspects of CSR play an important role for the companies that have been cited above.

Furthermore, the chemical industry is regarded as an important industry in all countries that are contained in the sample (Bug, 2014; Kikkawa, 2011). Since all of these countries are generally accepted as democratic industry countries, this selection also has the benefit of eliminating possible confounding factors like substantial differences in GDP or differences in formal institutions.

2. 7 CSR and Productivity: the Cultural Context

The possible influence of culture on the relationship between CSR and labour productivity has already been suggested by prior research: In their study that has been described earlier, Kim et al. (2010) indicate that culture could have an influence on the relationship between employees CSR associations and employee-company identification. They argue that “people in East Asia or collectivist cultures (e.g. China, Japan, and Korea) are highly motivated to conform to social relations”, whereas “in individualistic cultures (e.g., the United States), an independent self-construal usually occurs” (Kim et al., 2010). This is also in accordance with Roth et al. (2011), who state that idiocentrics “place higher emphasis on individual goals and needs”, whereas allocentrics are “placing higher emphasis on collective goals and obligations”.

(14)

level, but rather on differences and similarities on a societal level. Nevertheless, it is vital to reassure that findings from a country study can be transferred to an organizational level. Since CSR includes a commitment to society and to others outside the company, this could mean that the possible correlation between CSR and labour productivity is different in individualistic and collectivistic cultures (Singelis et al., 1995). Triandis (1995) describes individuals in individualistic societies as competitive, focusing on their own performance, whereas individuals in collectivistic cultures tend to be more sociable and interdependent. While some aspects of CSR, like the labour practices or employee relations, focus more on the individual, and other aspects, like the relationship to the community focus more on the collective, it can be assumed that these aspects have a different influence on labour productivity depending on the cultural context. Labour practices or employee relations, or the “employee dimension”, as Schreck (2011) names this aspect, include the safeguarding of jobs, training and education of employees, as well as their payment. These aspects and measures are clearly aiming to benefit the individual employee. Aspects like relationship to the community, or the “society and community dimension”, according to Schreck (2011), are concerned with “the company’s external relations with the community or communities in which it operates, and concerns issues such as charitable contributions” (Schreck, 2011). It can be assumed that these aspects are more related to collectivistic societies.

(15)

the underlying items into account. However in their paper, Brewer and Venaik (2011) question the definition of In-Group Collectivism, which is supposed to measure individualistic traits, and rather name this dimension “Family Collectivism”, since the items, which are used to investigate this “do not ask at all about organizations, only about families” (Brewer and Venaik, 2011). Thus, In-Group Collectivism is not regarded as an applicable dimension for the purpose of this paper.

House et al. (2004) rate the countries that are used as a sample in this paper as follows: South Korea: 5.2, Japan: 5.19, the United States: 4.2, and Germany (west): 3.79, with higher ratings indicating higher degrees of collectivism. Thus, in compliance with Triandis (1995), South Korea and Japan tend towards a collectivistic culture, whereas the United States and Germany tend towards an individualistic culture.

In conclusion, scholars investigating the relationship between CSR and competitiveness, more specifically between CSR and labour productivity, have found evidence for a positive link between these two factors. In addition, there is a theoretical influence of cultural on this relationship. However this influence was not yet empirically verified and thus needs further investigation. The following part of this paper will incorporate these relationships into a number of hypotheses, which will then be incorporated into a conceptual model. The links within this conceptual model are embedded in the existing literature that was discussed in this section.

2. 8 Hypotheses and Conceptual Model

As discussed above, individuals in individualistic cultures tend to value the achievement of their own goals over the goals of others and individuals in collectivistic cultures value the collective performance over their own goals. Thus, it can be assumed that those aspects of CSR that focus on the individual, like the relationship to employees, and work practices have a greater influence in individualistic societies (Triandis, 1995). On the other hand, CSR aspects that focus on the collective, like relationship to the community, are assumed to have a greater influence on labour productivity in collectivistic cultures than in individualistic. Thus, the following hypotheses can be derived:

H1a: The influence of individualistic aspects of CSR on labour productivity is stronger in individualistic than in collectivistic cultures.

(16)

These relationships are illustrated in the conceptual model (Figure 1). This conceptual model is derived from the framework of Vilanova et al. (2009), which was used by Sánchez and Benito-Hernández’ (2013) and Sun and Stuebs (2010), to conceptualize the relationship between CSR and labour productivity. The CSR aspects that are used in the conceptual model of this paper are taken from Schreck (2011) and Vilanavo et al. (2009). The following chapter will develop a study, which will investigate these relationships and test the hypotheses.

(17)

3. Methodology

In order to test the hypotheses, cross sectional, longitudinal CSR and financial data from companies within the chemical industry will be collected, using the Datastream ASSET4 database. Through the use of different groups, namely companies from several countries and cultural backgrounds, and an analysis of longitudinal data, the research design is thus cross-sequential. This method is employed in order to investigate the development of the variables over time and to increase the number of observations. One more reason for the use of this method is the mitigation of economic effects of nonrecurring events that had a financial impact in a single year and would thus deter the overall results. As a result, the external validity of possible findings is increased through the use of a cross-sequential method.

After conducting an outlier analysis, an independent means t-test will be used to measure the influence of culture on the dependent variable. To investigate the relationship between each variable, a correlation analysis will be used, followed by a multiple regression analysis. This regression analysis will be conducted using 5 different models, with three of them containing interaction terms to measure a possible interaction effect between culture and CSR aspects. The remaining of this chapter will describe the collected sample, the dependent variable, the independent variables, and the regression model.

3. 1 Sample

(18)

3. 2 Dependent Variable

Generally, productivity can be measured as the quotient of an input and an output (Coelli et al., 2005). In CSR research, this was translated by Sánchez and Hérnandez (2013), as well as Sun and Stuebs (2010) into operating income as an output measure and number of employees as an input measure. Thus, Operating Income/Number of Employees, named labour productivity (LP) serves as the independent variable. Vilanova et al. (2009) also point out that “income statement-based methods” as a measurement method of a company’s competitiveness belong to the “most widely used valuation methods”. Although Sun and Stuebs (2013) employ a different method to measure firm productivity, namely a Data Envelopment Analysis, operating income is also included in their model, providing further validity to this method.

3. 3 Independent Variables

(19)

3. 4 Analysis and Model

Before the data is analyzed using a multiple regression model, an outlier analysis is carried out. After this, an independent samples t-test is conducted to assess whether or not the means of labour productivity differ significantly between the two groups of observation. Subsequently, a correlation analysis is used to analyze the correlation between variables.

To test the hypotheses, 5 models are introduced. In model 1, the relationship between the control variable and the dependent variable is established; thus, the first model only contains the control variable, lnAssets.

LabourProd = β₀ + β₁ lnAssets + ε (1)

Model 2 will additionally include all independent variables that have been described above, in order to measure in particular the relationship between the group variable and the CSR ratings with labour productivity. This gives a first insight into the separate influence of group and CSR on labour productivity. However, this does not allow for an assessment of the

hypotheses. Consequently, model 3, 4 and 5 are introduced.

Labour Productivity = β₀ + β₁Group + β₂Community + β₃Employee + β₄lnAssets + ε (2)

In model 3 and 4, the interaction variables CommunityXGroup and EmployeeXGroup are added respectively. This allows for the identification of the individual effects between group and CSR ratings and the identification of possible interaction effects between group and CSR ratings on labour productivity.

Labour Productivity = β₀ + β₁Group + β₂Community + β₃Employee + β₄lnAssets + β₆

CommunityXGroup + ε (3)

Labour Productivity = β₀ + β₁Group + β₂Community + β₃Employee + β₄lnAssets + β₆

EmployeeXGroup + ε (4)

Finally, model 5 contains both interaction terms:

LabourProd = β₀ + β₁Group + β₂Community + β₃Employee + β₄lnAssets + β₆

CommunityXGroup + β₇ EmployeeXGroup y + ε (5)

(20)

4. Results

4. 1 Outliers and Missing Values

Missing values for employees for the companies LG Chem and Hanwha Chemical for the year 2012 were estimated using the values of 2011 and 2013. This procedure was also employed for the missing employee value in 2009 of one U.S. company, Mosaic, using the values for 2007 and 2010. Furthermore, CSR data was missing for some companies in certain years. These companies have been excluded in the respective years.

A scatterplot analysis showed that some values for the independent variable were relatively high, compared to the rest of the sample (see Appendix 4). As a result, two companies (Lotte Chemicals and CF Industries) were excluded from the sample and one value (OCI in 2010) was corrected, using the previous and preceding values. As it can be seen in appendix 5, a number of companies, besides the two companies that have been excluded from the sample, showed high labour productivity values, which were, in some cases, nearly seven standard deviations above the mean. Transforming the labour productivity values into their natural logarithm would solve this problem; however, since a number of values were negative, this procedure is not advisable. Since methods like a t-test and OLS regression are used in this paper, these outliers would have limited any further analysis by violating the Gauss-Markov assumptions of normality and linearity:

• Using the standardized predicted values and standardized residuals of the baseline regression model (Model 2), it can be observed that the residuals are not normally distributed. Thus the assumption of normality is not satisfied (Appendix 6).

• By using the dependent variable, labour productivity and one of the independent variables, the community rating, it can be observed that the observations are not linearly distributed (Appendix 7).

• A correlation analysis reveals that there is a low correlation between the independent and the dependent variables (Appendix 8).

(21)

Although the method of excluding outliers is controversial, this method was employed in order to increase the statistical accuracy of the following analysis. Using the standardized predicted values and standardized residuals of the baseline regression model, it can be observed that the residuals are now more normal distributed (Appendix 9).

4. 2 Descriptive Statistics and T-Test

Table 2 shows the mean and standard deviation for the sample. The results are listed per group. It can be observed that the number of observations between the two groups are nearly equal; 146 observations from individualistic cultures and 147 observations from collectivistic cultures.

Group N Mean Std. Deviation Std. Error Mean

LP Individualistic 146 $62.84 $33.61 $2.78

Collectivistic 147 $38.83 $33.04 $2.73

Table 1: Descriptive Group Statistics

An independent t-test was conducted to compare the means of the independent variable, labour productivity, for the two groups. As shown in table 3, there was a significant difference between companies from individualistic (M= $62.84, SD= $33.61) and collectivistic (M= $38.83 SD= $33.04) cultures; t (291) =6.167, p= 0.000.

T-Test for Equality of Means

t df Sig. (2-tailed) Mean Differenc e Std. Error Differenc e 95% Confidence Interval of the Difference Lower Upper LP Equal variances assumed 6.167 291 0.00 $24.02 $3.89 $16.35 $31.68 Equal variances not assumed 6.167 290.83 0.00 $24.02 $3.89 $16.35 $31.68

(22)

4. 3 Correlation

LP ComRat EmpRat Group lnASS

Labour Productivity 1

Community Rating 0.165** 1

Employee Rating 0.160** 0.715** 1

Group -0.340** -0.128* -0.249** 1

Ln(Assets) -0.020 0.507** 0.514** -0.075 1

Table 3: Correlation Matrix (N=293, ** p<0.01, * p<0.05)

A Pearson product-moment correlation coefficient has been computed to assess the relationship between the dependent variable, labour productivity (M = 50.79 SD = 35.38) and the independent variables. As it can be seen in table 4, the results of the correlation analysis show a significant, positive correlation between the two CSR ratings and labour productivity. There is a significant, positive correlation between ln(Assets) (M = 15.77 SD = 0.93) and both CSR ratings, which can be explained by evidence that suggests that bigger firms engage in more CSR activities (Chih et al., 2010). It is also worth noting that the two CSR ratings, the Community Rating (M = 0.6 SD = 0.19) and the Employee Rating (M = 0.51 SD = 0.18), have a highly positive and significant correlation; r = 0.165, p = 0.005, n = 293. When companies decide to engage in CSR, they usually try to cover the full range of CSR dimensions (Vilanova et al., 2009). In the United States for example, the government encourages chemical companies to participate in the Responsible Care program, which also promotes a holistic CSR approach (Sun and Stuebs, 2013). This can be seen as a possible explanation of this correlation.

Individualistic Collectivistic

LP ComRat EmpRat LP ComRat EmpRat

Labour Productivity 1 1

Community Rating 0.273** 1 -0.026 1

Employee Rating 0.186* 0.703** 1 -0.019 0.723** 1

Table 4: Correlation Matrix split by Group (N=293, ** p<0.01, * p<0.05)

(23)

4. 4 Regression

Model 1 Model 2 Model 3 Model 4 Model 5

Step and variables B SE B SE B SE B SE B SE

Intercept 62.79 (35.06) 129.18** (35.53) 128.16** (35.07) 133.68** (35.31) 129.37** (35.26) Control Ln(Assets) -0.76 (2.22) -5.7* (2.47) -6.87** (2.48) -7.06** (2.52) -7.04** (2.51) Main Effects Group -22.676** (3.99) 13.45 (12.93) 4.22 (12.18) 15.03 (13.53) Community Rating 33.43* (14.92) 60.28** (17.33) 33.8* (14.8) 56.74** (19.48) Employee Rating 5.67 (16.14) 10.96 (16.03) 35.88 (20.59) 17.5 (22.91) CommunityXGroup -60.28** (20.55) -52.13 (28.94) EmployeeXGroup -52.3* (22.39) -12.57 (31.37) R Square 0.00 0.147 0.171 0.172 0.172 Δ R Square 0.146 0.025 0.000 0.009

Table 5: Output Multiple Regression (** p<0.01, * p<0.05)

In order to test the hypotheses, a multiple regression analysis has been carried out using 5 models. The first model only uses the control variable, ln(Assets), the following four models were subsequently extended by the main independent variables (see table 6). A test of collinearity revealed that model 5 has problems regarding multicollinearity (CommunityXGroup, Tolerance = 0.044, VIF = 22.74; EmployeeXGroup, Tolerance = 0.52, VIF = 12.675, Group, Tolerance = 0.079, VIF = 12.67). This effect can be explained by the inclusion of both interaction variables, which both include the Group variable. Consequently, this model will be excluded from the analysis and model 3 and 4 will be used to assess the hypotheses. It is also important to mention that there are substantial differences between the labour productivity means of the two groups. However, as already mentioned in the methodology description, this problem was accounted for by introducing a fixed group variable.

(24)

evidence that CSR activities that are targeted at the community have a stronger positive influence in individualistic than in collectivistic cultures. This leads to the rejection of hypothesis 1b. A similar pattern can be observed using model 4. In this model, group 1 has a significant negative moderating influence on the relationship between the employee rating and labour productivity (Beta = -52.3, p < 0.05). Similar to model 3, the positive effect of the employee rating that is observed in model 4 (Beta = 35.88, ns) turns negative in individualistic cultures. Although the effect of the employee rating on labour productivity did not reach significance, the interaction effect between group and the employee rating did. Thus, hypothesis 1a is not rejected.

(25)

5. Discussion

As Porter and Kramer (2006) argue, CSR can be more than a simple marketing tool to boost reputation. With the right implementation, CSR can become a source of competitive advantage. Despite the limitations of this study, which will be discussed in chapter 7, it was possible to observe a positive relationship between CSR efforts and labour productivity. As it was shown in this paper, the moderating effect of culture has to be taken into account, when investigating this relationship.

The results of the correlation analysis and regression models 2, 3 and 4 showed that both CSR activities that are targeted at the community, as well as CSR activities that are undertaken for the benefit of the individual employee, have a positive influence on labour productivity for companies within the chemical industry in individualistic cultures like the U.S. and Germany. These findings are in line with evidence found by Sun and Stuebs (2013), who suggest that “CSR can lead to higher [firm] productivity in the chemical industry” and add the employee perspective to the firm level perspective that Sun and Stuebs (2013) employed.

This effect could not be observed for companies in collectivistic cultures like Japan and South Korea. This leads to the question whether substantial differences in labour productivity that were observed using a t-test cause the different relationships between the CSR ratings and labour productivity in the two groups of observation. Splitting up the results of the correlation analysis by group showed a small, non-significant, but negative relationship between both CSR ratings and labour productivity in a collectivistic context. Although this correlation did not reach significance, it can be seen as preliminary evidence for the negative relationship between CSR ratings and labour productivity.

One reason for the obtained results in collectivistic cultures could be the aftermath of the Tōhoku earthquake and tsunami, which had a lasting impact on the Japanese economy (Elliot, 2013). As a result of this impact, the labour productivity values in the years after 2011 have probably been diminished (Elliot, 2013). Furthermore, the earthquake also started a discussion regarding the impact of companies on environment and society, which might have resulted in increasing CSR public relation efforts of companies in order to regain the trust of society. A detailed analysis of the impact of this event is beyond the scope of this paper and will thus not be further discussed.

(26)

to collectivistic cultures. Thus, hypothesis 1b is not rejected, whereas hypothesis 1a was not supported by the results. Linking these findings to the research question it can be concluded that culture has an effect on the relationship between CSR and labour productivity. As already mentioned, there are some limitations to this study. These will be further discussed in chapter 7.

6. Conclusion

As discussed in the previous section, the results of the present study show a positive relationship between CSR engagement and labour productivity in individualistic cultures. These findings further contribute to the business case of CSR and provide evidence that the community and employee engagement of companies can also have financial benefits. These findings are in line with previous research and the increasing trend of financial investors to look for stock that scores high on CSR indicators (Carroll and Schabana, 2010).

In addition to contributing to empirical CSR research, the present study also provides further reasons, besides reputational benefits, for companies to engage in social activities that benefit employees and communities. Thus, chemical companies should systematically analyze their supply chain and competitive environment for CSR related challenges and opportunities for the creation of shared value (Porter and Kramer, 2011). In order to increase the benefits of this process in the form of an increased commitment of their employees, firms have to include their employees in the communication, as well as the creation of this shared value (Kim et al., 2010). As the present study shows, this can ultimately lead to a competitive advantage in the form of increased labour productivity.

The present paper can further be used by NGOs that try to convince businesses to pursue a more sustainable business practice. Using these findings, NGOs and affected communities can argue that social and environmental friendly business practices do not have to come with a financial trade off. They can use these arguments to convince chemical companies to cooperate with them and use their expertise for the creation of shared value and a more sustainable form of business.

(27)

7. Limitations

As mentioned above, one of the limitations of this study was the effects of the Tōhoku earthquake on the Japanese economy. This probably caused a decrease in operating income for Japanese chemical firms and subsequently to a decrease in labour productivity, while the CSR activities of companies were increased as a result of the earthquake. This could have led to the negative correlation between labour productivity and CSR activities that was observed in this study.

A further limitation is the number of companies that are contained in the sample. After deletion of 2 companies because of extreme values, the sample contained 65 companies from both cultures. Although the companies within the sample can be seen as representative of the chemical industry in terms of number of employees and revenue, they are limited in terms of number. This reduces the statistical accuracy of this study, which manifests itself in the low significance ratings that had been observed. However, the Datastream database did not have more CSR data available at the time of data collection and the author of this paper could not get access to a different CSR database, like the KLD database that was also used by other scholars investigating the financial effects of CSR (Scholtens, 2009; Sun and Stuebs, 2013). An additional limitation is the trimming of the outliers that has been carried out. Although this trimming was explained using statistical accuracy as an argument, it was not supported for other reasons than that.

The cultural factor that was included in this study also adds limitations. Although the GLOBE study contains an impressive sample and uses an advanced methodology that is in line with the theoretical background of this paper, it was not conducted using solely companies and individuals that are active in the chemical industry. A more reliable approach would have been a systematic questioning of employees within the industry, asking for attitudes about both topics, culture and social responsibility. However, given the boundaries of this paper, this procedure was not possible.

(28)

8. Future Research

As discussed in the previous section, the limitations of this study are also opportunities for future research. A combination of a qualitative and quantitative approach that examines the underlying processes that connect CSR activities with labour productivity could provide a better understanding of this relationship. As discussed in the literature review, scholars argue that the implementation of a CSR strategy leads to increased productivity as a result of increased innovativeness, quality and reputation (Vilanova et al., 2009). Especially the mediating aspect of reputation could be of interest in a cultural setting. The conceptual model in appendix 12 can serve as a foundation for future research investigating this mediation in the context of culture. Answering the questions of how employees perceive the CSR efforts of their company and how much they care about these efforts could help to further understand this effect and the business case for CSR. Kim et al. (2010) and Sun and Stuebs (2010) have already laid the theoretical foundation and found evidence for this effect.

(29)

9. References

Ambastha, A., Momaya, K. 2004. Competitiveness of firms: review of theory, frameworks, and models. Singapore Management Review, 26(1): 45-61.

BASF. 2015. BASF report 2014, economic, environmental and social performance. BASF Se: Ludwigshafen.

Bowen, H. R. 1953. Social responsibility of the businessman. New York: Harper & Row. Branco, M.C., Rodrigues, L. L. 2006. Corporate social responsibility and resource-based

perspectives. Journal of Business Ethics, 69: 111-132.

Breusch, T. S., A. R. Pagan. 1979. A simple test for heteroscedasticity and random coefficient variation. Econometrica, 47: 1287-1294.

Brewer, P., Venaik, S. 2011. Individualism-collectivism in Hofstede and globe. Journal of International Business Studies, 42(3): 436-445.

Bug, T. 2014. The German chemical industry. Berlin: German Trade & Invest.

Burchielli, R., Delaney, A., Tate, J., Coventry, K. 2009. A fairwear campaign: an ethical network in the Australian garment industry. Journal of Business Ethics, 90(4): 575-588.

Carroll, A. B. 1999. Corporate social responsibilty. Evolution of a definitional construct. Business and Society, 38(3): 268-295.

Carroll, A. B., Shabana, K. M. 2010. The business case for corporate social responsibility: a review of concepts, research and practice. International Journal of Management Review, 12(1): 85-105.

Chih, H.L., Chih, H.H, Chih, T.Y. 2010. On the determinants of corporate social responsibility: international evidence on the financial industry. Journal of Business Ethics, 93: 115-135.

(30)

Cook, R. D., S. Weisberg. 1983. Diagnostics for heteroscedasticity in regression. Biometrika, 70: 1-10.

Dow Chemicals. 2015a. 2015 Sustainability goals: 4Q 2014 update. The Dow Chemical Company: Midland.

Dow Chemicals. 2015b. Goals & reporting. The Dow Chemical Company. http://www.dow.com/en-us/science-and-sustainability/sustainability-reporting. Accessed 30.04.2015.

Elliot, D. 2013. Fukushima: Impacts and implications. Basingstoke, Hampshire: Palgrave Macmillan.

European Commission. 2002. Green book: promoting a European framework for corporate social responsibility. Commission of the European Communities. Brussels.

Fortune. 2003. Corporate America’s social conscience. Fortune Magazine. 8.

Friedman, M. 1970. The social responsibility of business is to increase its profits. New York Times Magazine.

House, R.J., Hanges, P.J., Javidan, M., Dorfman, P.W., Gupta, V. 2004. Culture, leadership and organizations. The globe study of 62 societies. Thousand Oaks, CO: SAGE. Kikkawa, T. 2011. How Japan’s chemical industry can overtake the world. The Chemical

Society of Japan, 64(7): 559-560.

Kim, H. R., Lee, M., Lee, H. T., Kim, N. M. 2010. Corporate social responsibility and employee-company identification. Journal of Business Ethics, 95(4): 557-569. Kurucz, E., Colbert, B., Wheeler, D. 2008. The business case for corporate social responsibility.

The Oxford Handbook of Corporate Social Responsibility. Oxford: Oxford University Press.

(31)

Mitsui Chemicals. 2015. Mitsui chemicals group: CSR communication 2014, Mitsui Chemicals, Tokyo.

Moser, M. R. 1986. A framework for analyzing corporate social responsibility. Journal of Business Ethics, 5(1): 69-72.

Payaud, M. A. 2014. Marketing strategies at the bottom of the pyramid: examples from Nestlé, Danone, and Proctor & Gamble. Global Business and Organizational Excellence, January/February 2014.

Porter, M. E. 1979. How competitive forces shape strategy. Harvard Business Review, 57(2): 137-145.

Porter, M. E. 1985. The competitive advantage: creating and sustaining superior performance. New York: Free Press.

Porter, M. E. 1990. The competitive advantage of nations. New York: Free Press.

Porter, M. E., Kramer, M. R. 2006. Strategy & society: The link between competitive advantage and corporate social responsibility.Harvard Business Review, 84(12).

Porter, M. E., Kramer, M. R. 2011. Creating shared value. Harvard Business Review, 89: 62-77.

Roth, K., Kostova, T., Dakhli, M. 2011. Exploring cultural misfit: Causes and consequences. International Business Review, 20: 15-26.

Sánchez, P. E., Benito-Hernández, S. 2013. CSR policies: Effects on labour productivity in Spanish micro and small manufacturing companies. Journal of Business Ethics. Scholtens, B. 2009. Corporate social responsibility in the international banking industry.

Journal of Business Ethics, 86(2): 159-175.

(32)

Singelis, T. M., Triandis, H. C., Bhawuk, D., Gelfand, M. J. 1995. Horizontal and vertical dimensions of individualism and collectivism: A theoretical and measurement refinement. The Journal of Comparative Social Science, 29: 240-275.

Sun, L., Stuebs, M. 2010. Business reputation and labor efficiency, productivity, and cost. Journal of Business Ethics, 96(2): 265-283.

Sun, L., Stuebs, M. 2013. Corporate social responsibility and firm productivity: Evidence from the chemical industry in the United States. Journal of Business Ethics, 118(3): 251-263.

Thomas Reuters. 2010. Environmental, social & governance (ESG) data. http://extranet.datastream.com/data/ASSET4%20ESG/Index.htm. Accessed on 02.04.2015.

Triandis, H. C. 1995. Individualism and Collectivism. Boulder, CO: Westview.

Verboven, H. 2011. Communicating csr and business identity in the chemical industry through mission slogans. Business Communication Quarterly, 74(4): 415-431.

Vilanova, M., Lonzano, J.M., Arenas, D. 2009. Exploring the nature of the relationship between csr and competitiveness. Journal of Business Ethics, 87(1): 57-69.

Windsor, D. 2001. The future of corporate social responsibility. The International Journal of Organizational Analysis, 9(3): 225-256.

(33)

10. Appendix

(34)

(35)

Appendix 3: Datastream Asset4 Datapoints with Description (Thomas Reuters, 2010)

Datapoints Individual Score Datastream Code Description

Diversity and Opportunity Policy Elements/Work

Life Balance SODODP0012

Does the company have a work-life balance policy?

Vacations and Sabbaticals SODODP025 Does the company claim to provide generous vacations, career breaks or sabbaticals? Day Care Services SODODP027 Does the company claim to provide day care services for its employees?

Value - Employment Quality/Implementation SOEQD02V Does the company describe the implementation of its employment quality policy? Employee Relations Policy Elements/Culture of

Trust SOEQDP0012

Does the company have a job security policy?

Employment Quality Monitoring SOEQDP010 Does the company monitor or measure its performance on employment quality? Bonus Plan for Employees/Employees SOEQDP0201 Does the company claim to provide a bonus plan to most employees?

Generous Fringe Benefits SOEQDP025 Does the company claim to provide its employees with a pension fund, health care or other insurances?

Employee Health & Safety Policy Elements/Health

& Safety SOHSDP0011

Does the company have a policy to improve employee health & safety? Employee Health & Safety Processes/Health &

Safety SOHSDP0121

Does the company describe, claim to have or mention processes in place to improve employee health & safety?

Value - Training and Development/Policy SOTDD01V Does the company have a policy to support the skills training or career development of its employees?

Value - Training and Development/Improvements SOTDD04V Does the company set specific objectives to be achieved on the employee training and career development?

Training and Career Development Policy

Elements/Skills Training SOTDDP0011

(36)

Datapoints Community Score Datastream Code Description

University Partnerships SOTDDP025 Does the company claim to cooperate with schools or universities?

Community Reputation Monitoring SOCODP021 Does the company monitor its reputation or its relations with communities? Company Foundation SOCODP038 Does the company provide funding of community-related projects through a

corporate foundation?

Employee Engagement Voluntary Work SOCODP037 Does the company foster employee engagement in voluntary work? Community Reputation Processes/Indirect

Economic Impact SOCODP0125

Does the company describe, claim to have or mention processes in place to improve the indirect economic impact it has on local communities?

Community Reputation Policy Elements/Indirect

Does the company have a policy to strive to increase the indirect economic impact it has on local communities?

Value - Community/Policy SOCOD01V

Does the company have a policy to strive to be a good corporate citizen or endorse the Global Sullivan Principles? AND Does the company have a policy to respect business ethics or has the company signed the UN Global Compact or follow the OECD guidelines?

Community Reputation Commitment/Good Citizen

/Community Involvement SOCODP0031

Has there been a public commitment from a senior management or board member to good corporate citizenship?

Community Reputation Commitment/Indirect

Has there been a public commitment from a senior management or board member to strive to increase the indirect economic impact it has on local communities?

Community Reputation Processes/Good Citizen

/Community Involvement SOCODP0121

Does the company describe, claim to have or mention processes in place to improve its good corporate citizenship?

Value - Community/Donations in General SOCOO02V

(37)

(38)

(39)

(40)

(41)

Appendix 8: Correlation Matrix – Outliers corrected, Lotte Chemical and CF Industries excluded (** p<0.01, * p<0.05, N=310)

LP ComRat EmpRat IC lnASS

(42)

(43)

Appendix 10: Histogram of Distribution of the Dependent Variable

(44)

Appendix 11: Regression results for all models, Outliers corrected, Lotte Chemical and CF Industries excluded (** p<0.01, * p<0.05, N=310)

Model 1 Model 2 Model 3 Model 4 Model 5

Step and variables B SE B SE B SE B SE B SE

(45)

Corporate Social Responsibility and Labour Productivity in the Context of Individualistic and Collectivistic Cultures: Evidence from the Chemical Industry in the United States, Japan, Germany, and South Korea