Three dimensions of corporate social responsibility and how
they affect financial performance: A quantitative research on
large organisations in the United States
By Nicolas A. Poolen (10646779) Supervisor: Pushpika Vishwanathan
Date: 22 April 2016
Abstract
Although profitability of corporate social responsibility (CSR) has been researched extensively, inconsistencies in approaches and outcomes continue to exist. In an attempt to explain these inconsistencies, previously adopted categorization methods are analysed. Subsequently, a new categorization model is developed to measure profitability of CSR sub-dimensions. The proposed dimensions are negative externality reduction, integrated public good creation and non-integrated public good creation. This research proposes CSR activities which are integrated with business process can foster more financial benefits than non-integrated CSR activities. Therefore, it is hypothesized negative externality reduction and integrated public good creation result in
increased financial performance. On the contrary, it also hypothesized non-integrated public good creation results in decreased financial performance. We test these hypotheses with a regression analyses on a sample of 2771 observations of 811 firms from 2005 till 2009. Only for integrated public good creation, a positive relation could be supported, suggesting firms who engage in CSR activities which are integrated with business processes and aimed at providing a public good, will result in increased financial performance.
Statement of Originality
This document is written by Nicolas Poolen, 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.
Introduction
Managers are under increasing pressure to consider firm’s social responsibility. However, firm’s social responsibility is a contested concept. The inconsistency in perspectives on corporate social responsibility (CSR), poses a challenge for research on CSR. Over the last decades, research has focussed on the relationship between CSR and corporate financial performance (CFP). A financial return for investing in CSR, would make the moral discussion obsolete. However, findings are mixed. While a meta-analysis of Orlitzky, Schmidt & Rynes (2003), finds an overall positive relationship, discussion on the subject continues.
A primary cause of the inconsistency of findings, may be due to the monolith regarding in past research. Barnett (2007) counters the perception of an overall consistent relationship, and argues for a firm-specific business case. In addition, Barnett and Salmon (2006) have been revolutionary to claim that different CSR-types may have diverse influences on CFP. As currently a variety of mediators, moderators and categorizations regarding the relationship between CSP and CFP have been proposed, the need for knowledge on the conditions affecting profitability of CSR becomes increasingly recognized by academics.
In current literature, the conditions affecting profitability of CSR, are often researched in distinguished dimensions of CSR. However, since CSR is a broad concept, there are many approaches to distinguish among CSR activities. This has resulted in a wide variety of CSR categorizations, often with unclarified overlap and interrelatedness among dimensions. Furthermore, for the dimensions for which interrelatedness are clarified, outcomes are often inconsistent. Therefore, usefulness of these researches has been limited.
This research attempt to summarize dimension differentiations within CSR-research to establish an overarching categorization, which would increase comparability. This results in the following dimensions: negative externality reduction, integrated public good creation and
non-integrated public good creation. Subsequently, this research tests these dimensions in terms of profitability. The following research question has been developed to address these issues: How do negative externality reductions, integrated public good creations and non-integrated public good creations affect CFP?
First, literature will be explored on CSR in general, return on CSR and different dimensions of CSR. Taken insights from observed concepts, a conceptual model will be developed. In the methods section, a measurement strategy for the conceptual model will be validated by experts. Then a regression analyses will be conducted on the different CSR dimensions and CFP to answer the research question. (conclusie)
Theory
The question of corporate social responsibility (CSR) originated with the moral discussion of the responsibility of business in society. As often mentioned (Van Oosterhout & Heugens, 2008), social responsibility is hard to define. A starting point for perspectives on social responsibility often regards the conceptualization of responsibility. Freeman (1984) and Friedman (1962) are rivalry pioneers on the concept. Whereas Friedman argues the only responsibility of the firm is to provide value for its owners (shareholders), Freeman, on the other hand argues the firm is responsible for all actors affected by the firm. Building upon this perspective, he develops a concept known as stakeholder theory. The theory holds, that every firm has specific groups with interests in the firm’s operations. The firm should actively manage interest of all these “stakeholders”. An interesting and convincing theory, according to many academics. Although Carroll & Buchholtz (1996) have provided a guideline to differentiate among legal, economical, ethical and philanthropic responsibilities of the firm, perspectives on the reach of the responsibility and to which stakeholders, are still divergent.
A widely adopted definition of CSR is provided by Davis (1974): “the firm’s considerations
of, and response to, issues beyond the narrow economic, technical and legal requirements of the firm” (P.312) to “accomplish social benefits along with the traditional economic gains which the firm seeks” (P. 313). Although this may sound noble, this description does not enable practical
operationalisation. It is important to consider multinationals operate primarily in multiple social dimensions. What is considered as social issue, or social benefit, may differentiate across nations and time. The European Commission (2011) provided a similar definition of CSR, which was required to be adopted across Europe for the purpose of building policy: “a concept whereby
companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis” (p.2). However, after 2011, the EU
broadened its definition to: “the responsibility of enterprises for their impacts on society” (p.2). The difference in the two definitions is a contested subject in literature. While the first definition requires activities addressing social concerns to be integrated within business operations, the second definition allows these activities to be not integrated within business operations.
Barnett (2007) attempts to narrow the definition of CSR and claims CSR only to be legitimate, when it adopts both a shareholder orientation and a social welfare orientation. Meanwhile, Van Oosterhout & Heugens (2008) claim no satisfactory definition of CSR has been adopted in previous research, which not has been victim to one of the following fallacies:(1) negative extensionality; (2) under inclusiveness; (3) category lumping; and (4) outcome fetishism.
While no consistent approach seems available, research on the subject has continued far beyond defining CSR, and researchers often develop their own approach to CSR.
Return on CSR
Partly as a result of the diverse perspective of CSR as a concept, the discussion on the economics of CSR exists. Previous research suggested investments in CSR could result in both negative or positive financial returns. A meta analyses of Orlitzky, Schmidt & Rynes (2003), found an overall positive relationship. However, a meta-analyses of Margolis & Walsh (2003) found mixed results.
A significant amount of research has decoupled the main arguments for a positive relationship, and performed empirical research to support the individual arguments. A first argument is that CSR activities increases reputation. Subsequently, increased reputation could result in increased market value and increased sales (Roberts & Dowling, 2002), for which Schwaiger (2004) observed empirical support. Second, firms engaging in CSR activities decrease risks (Roberts & Dowling, 2002). In addition, it often reasoned, engaging in CSR increases employee’s motivation and company engagement (Mirvis, 2012; Vlachos, Panagopoulos & Rap, 2013). Therefore, firms will decrease recruitment, training and retention costs, and increase employee performance.
However, proponents of the positive relationship, argue CSR activities often include allocating returns owned by shareholders, which do not provide a positive return (McWilliams & Siegel, 2000). This allocation can be viewed as a decrease in financial performance. Furthermore, getting involved in CSR-practices, may distract managers, from core business operations (Wang, & Bansal, 2012).
A heterogeneous relationship
These pro and con arguments are a response to the past monolith regarding of the CSR-CFP relationship. In this view, CSR either has a positive or a negative relationship with CFP. However, no investment provides a guarantee for a positive return. There may be infinite external conditions,
determining the return of every investment. The same holds for CSR investments. One could imagine, a large banking company sponsoring a small neighbourhood event, will not result in the same reputation increase, compared to when a local supermarket will do the same. Furthermore, firm, market and industry characteristics, have been observed to moderate the CSR-CFP relationship (Dixon-Fowler, Slater, Johnson, lstrand & Romi, 2013). Therefore, discussion on the CSR-CFP relationship has mostly acknowledge the heterogeneity (Godfrey, Hatch, & Hansen, 2008; Hillman & Keim, 2001)
Besides conditions external to the CSR activity, the internal factors of CSR could determine its return as well. Compared to sponsoring a neighbourhood event, hiring minorities seems subject to completely different conditions, regarding potential financial returns. Multiple conceptualizations of diverse CSR activities and their relationship to CFP, have been proposed. This research will establish a new conceptualization to differentiate among CSR activities.
Dimensions of CSR
To make a valid conceptualization, this research first analyses categorizations of CSR already utilized. By analysing which factors are often used to differentiate among CSR activities, consistent conclusions for distinguished dimensions can adopted for this research.
A first, categorization differentiates CSR by the type of stakeholder or social issue addressed by the CSR activity. However, the type of social issue can be expressed by multiple categorizations. An often utilized measurement and categorization of CSR is provided by the KLD-index (Inoue & Lee, 2010). This KLD-index categorizes environmental, social and governance performance indicators into community, environment, employee relations, corporate governance, diversity, human rights and product.
While the previous type of categorization defines clear subcategories, categorization based on the type of stakeholder relation addressed by a CSR activity, has been less consistent. Frederick (2006) provides definitions for three types of stakeholder management. The first is corporate philanthropy, in which a firm “has a policy and an active program of making voluntary
contributions, out of corporate profits, to various educational, charitable, religious, cultural or other community agencies” (p.30). In the second type, the manager is seen as a public trustee, who
“act as mediator among various groups that have claims on the corporation, seeking out strike an
equitable balance among them” (p.31). The third is direct social action, in which firms “enter directly into confrontation with social problems and seek a solution to them by utilizing the skills, the resources and the general influence of the corporation itself” (p.32). While the first category
seems to be an independent process, the other two categories can be considered as processes which are intertwined with other business processes.
Halme & Laurila (2009) also categorize based on the dependency of a CSR process, and propose divergent influences on financial performance for all CSR types. Their research includes the following similar variables and definitions: “(1) Philanthropy (emphasis on charity,
sponsor-ships, employee voluntarism etc.), (2) CR Integration (emphasis on conducting existing business operations more responsibly) and (3) CR Innovation (emphasis on developing new business models for solving social and environmental problems)” (p. 329) The distinction between the first and the
other two is a relatively clear; CSR activities distinct from the business model (philanthropy CSR) and CSR activities integrated within the business model. However, innovation CSR is less clearly differentiated from integrated CSR. Their definition causes innovation CSR to be similar to integration CSR, but instead has to be “new”. Their article lacks argumentation on what “new” models are, and why it should have a separate influence on CFP. While their research has not been
supported be empirical evidence, they argue for little or negative potential benefit of philanthropy CSR, large potential benefit for integration CSR and diverse potential benefit of innovation CSR.
The loosely defined innovation CSR category of Halme & Laurila, may be similar to systematic CSR out the “ages of CSR” as defined Visser (2010). According to him, CSR is expressed in the age of greed, philanthropy, marketing, management and finally, responsibility. He argues, only the latter age includes the proper CSR form: Systematic CSR, where firms “focus their
activities on identifying and tackling the root causes of our present unsustainability and irresponsibility, typically through innovating business models, revolutionising their processes, products and services and lobbying for progressive national and international policies” (p. 9).
According to him, other forms have multiple flaws.
Both Visser (2010) and Halme & Laurila (2009) include a philanthropy category. This category was first distinguished by Carrol & Buchholtz (1996). Wang & Berens (2015) build upon their categorization of legal, economic, moral and philanthropic CSR activities. They research the influence of the different CSR types on CFP, mediated by financial reputation and public reputation. While their typology provides interesting insights, there lacks further research acknowledging their findings. They do not find a significant relationship between the philanthropy CSR and both of the mediators. This would indicate, in their relatively small sample (n=231), the extent to which a firm distributes their wealth to charity, would neither affect the relationship with financial stakeholders nor public stakeholders, which seems rather illogical.
A last well-known relevant categorization, is CSR activities towards primary stakeholders and secondary stakeholders. The difference is that primary stakeholders and the firm are mutually able to directly affect each other, while secondary stakeholders are not able to directly affect the firm. While the previous stated philanthropy category could deal with primary stakeholders, in practice the focus is on secondary stakeholders (Halme & Laurila, 2009). On the contrary, CSR
activities integrated within business operations, deals mostly with the core stakeholders, such as primary stakeholders. Hillman & Keim (2001) provide empirical insight in this categorization, and find a positive relationship between CSR towards primary stakeholders and CFP and a negative relationship between secondary stakeholders and CFP.
Public goods versus negative externalities
A first point to be made regarding the categorizations explained above, is the lack for differentiation in public good creation versus negative externalities reduction. To explain why this distinction is important, this paper first turns to literature on externalities. Externalities occur, when certain costs or benefits are not included in a transaction (Marshall & Guillebaud, 1961). For example, if a consumer buys a flight-ticket from Amsterdam Schiphol to London Airport, the transaction does not include the costs of noise pollution, residents surrounding the airport perceive as a result of the flight. When Schiphol airport voluntarily reduces such negative externality, such as building noise barriers, it can be viewed as Schiphol is considering their social responsibility.
The opposite of negative externality is positive externality. This occurs when a “product’s
value to the community is greater than their price” (Devarajan, Squire, & Suthiwart-Narueput,
1997, p. 37). However efficient firms will always attempt to account for positive community value of CSR activities in the price of products and services. Therefore, a differentiation between positive and negative externalities would not include all CSR activities. In addition, positive externality promotion is difficult to measure, as community value and product price are not directly comparable. Therefore, this research differentiates between negative externality reductions, and all forms of public good creation, which includes positive externalities.
It is difficult to examine public good creation, since previous literature has not yet deviated CSR in this dimension. Therefore, this research will build its analyses mostly upon examples.
Public good creation does not have to be a result of a transaction. For instance, Schiphol Airport, could implement policies to support career development for minorities within their organisation. Most arguments for a positive relationship between CSR and CFP are applicable to only public good creation or negative externality reduction. For example, increased financial performance as a result of eco-efficiently is not applicable to public good creation. Furthermore, one could argue increased financial performance due to reputational changes as a result of CSR activities may be diverse for increasing positive impact, relative to decreasing negative impact. These examples indicate the relationship of both negative externality reduction and public good creation with financial performance, are different in nature, and are required to be observed separately.
While some articles include both public good creation and negative externality reduction in their approach to CSR (disregarding the difference in economic return) others only include one of the two. For example, Bagnoli and Watts (2003) and Kotchen (2006), consider CSR to be the simultaneous production of private and public goods. This view does not account for the reduction of negative social impact. Visser’s (2010) argues CSR activities should solely include activities aimed at increasing positive impact, as these activities have the only potential to increase social wealth.
Two dimensions of public good creation
A second point regarding the literature on categorizations this research has observed, is the consistent need for a separation of CSR activities that are integrated within business operations, and CSR activities that are not. The reduction of negative externalities, always involves the adaption, or rethinking of existing business practises (Marshall & Guillebaud, 1961). Therefore, these CSR activities are integrated with normal business processes. On the contrary, public good creation, does not have to be integrated with business operations. For example, a public good
creation activity that is not integrated with business operations is a donation to a charity. However, no consistent delineating of integrated versus non-integrated public good creation has been provided.
Not-integrated public good creating activities are observed particularly in the form of philanthropic activities, such as donations. Nevertheless, similar to CSR as a whole, the definition of philanthropy CSR is inconsistent. Carroll & Buchholtz (1996) state philanthropy “includes
actively engaging in acts or programs to promote human welfare or goodwill “ (p.42). This paper
recognizes this definition is too broad to operationalize. Godfrey (2005) stated the FASB (1993) to very specifically define corporate philanthropy as “an unconditional transfer of cash or other
assets to an entity or a settlement or cancellation of its liabilities in a voluntary nonreciprocal transfer by another entity acting other than as an owner" (p. 28). Yet, this definition does not
include volunteering programs (Carroll & Buchholtz, 1996; Halme & Laurila, 2009). Therefore, this paper bases its definition of non-integrated public good creation on the most consistent elements in literature: Business activities, aimed at providing public goods and services, independent of processes aimed at the creation of goods and services creation.
For integrated public good creation this paper combines definitions of integrated CSR and innovation CSR from Halme & Laurila (2009). This results in a definition which states integrated public good creation are activities aimed at producing simultaneous public and private goods or services. The differentiation in dimensions is summarized in the table 1.
Negative externality reduction Integrated public good creation Non-integrated public good creation Social value creation area Integrated within business core Integrated within business core
Outside business core
Nature of activities Negative impact illumination
Positive impact provision
Positive impact provision Examples Green energy use,
waste management Employee engagement, opportunities for minorities, social product benefits
Donation of cash and non-cash assets, donation of workforce, cancelation of liabilities
Table 1; Dimensions of public good creation
Return on CSR-dimensions
To build argumentation towards relationships, past literature on the three categories, and related concepts are examined. The first concept, negative externality reduction, has independently been relatively unexplored. As negative externality reductions focus on voluntarily decreasing negative impact, it imputes unrequired costs, as do other forms of CSR. Hillman & Kiem (2001) observe a negative relationship between CSR activities aimed at secondary stakeholders, and firm market value. While the reduction of negative externalities could be focussed on primary stakeholders, in practise it largely deals with secondary stakeholders Halme & Laurila (2009). This would indicate a negative relationship between negative externality reduction and firm market value.
However, the meta analyses of Orlitzky, Schmidt & Rynes (2003), finds an overall positive relationship between general CSR and CFP, which could point at a positive influence of negative externality reductions. The main arguments described earlier, reputational benefits and risks reductions as a result of CSR activities, are especially applicable to negative externality reductions, since activities within this dimension decrease negative impact on society, of which consumers are dependent.
In addition, Halme & Laurila (2009) argue for a large potential benefit for integration and innovation CSR, which are related to negative externality reductions, as these concepts are also integrated within business processes. This supports the proposition that negative externality reductions could provide positive financial benefits.
Lastly, an often cited argument for a negative relationship of CSR to CFP due to distracted managers, is not applicable to negative externality reductions, as these activities are interrelated with core business activities. Especially eco-efficiency gains, could result in positive returns (Korhonen, 2003). Therefore, the following relationship is hypothesized:
H1: Negative externality reduction is positive related to corporate financial performance.
Empirical research on the relationship between integrated public good creation and CFP has been relatively unexplored as well, and builds upon similar arguments as H1. First, the meta analyses of Orlitzky, Schmidt & Rynes (2003) also could point at a positive influence for integrated public goods creation. According to attribution theory, reputational benefits as a result of CSR activity, is also applicable to integrated public good creation, since activities within this dimension are aimed at voluntarily providing a good to the public.
In addition, since the creation of public goods, simultaneous with private goods, is an integrated process, it relates to integration and innovation CSR, which according to Halme & Laura (2009) provide large potential benefit.
The difference between negative externality reduction and integrated public good creation is in the addressed stakeholders. Whereas negative externality reductions are mostly aimed at secondary stakeholders, integrated public good creation is integrated with core business activities and therefore deals with primary stakeholders. Hillman & Kiem (2001) observe a positive relationship between CSR activities aimed at primary stakeholders, and firm market value. This points at a positive relation of integrated public good creation and financial performance.
Lastly, integrated public good creation is also dependent upon core business activities. Therefore, the argument for a negative relationship of CSR to CFP due to distracted managers, is also not applicable to integrated public good creation. As most related concepts point at a positive relation, the hypotheses can be developed:
H2; Integrated public good creation is positive related to CFP
For the last concept, not-integrated public good creation, relationships have been only empirically observed for the related concept philanthropy. Argumentation for a positive relationship with CFP, evolves around similar argumentation for CSR as a whole; philanthropy increases reputation and stakeholder relations, which would increase CFP. Furthermore, Godfrey (2005) claims that, according to attribution theory, strategic corporate philanthropy increases firm’s moral capital and global reputation, which increases CFP. His argumentation is supported by Patten (2008), who finds a positive relation between market reputation and donations to support victims of a tsunami. However, they may not have recognized firms engaging in philanthropic activities, likely also engage in other CSR activities. Therefore, the positive effect observed, may be explained by the other CSR activities for which the research not has controlled.
Furthermore, on the contrary to integrated public good creation and integrated negative externality reduction, non-integrated negative externality reductions are not integral to core business. Sasse & Trahan (2007) argue ”corporations are less than competent in dealing with
products and services not integral to their business“ (p.35). Therefore, managers could be
incompetent to strategize for these non-integrated negative externality reductions.
Lastly, Halme & Laurila (2009) argued for little potential benefit for the related concept philanthropic CSR, in comparison to other CSR forms. Similarly, Hillman & Kiem (2001) observe a negative relationship between CSR towards secondary stakeholders and firm market value. As not integrated public good creation, concerns a process outside of core business, it largely deals
with stakeholders who are not interdependent on the firm (secondary). Therefore, Hillman & Kiem’s outcomes would indicate not-integrated public good creation would result in negative financial performance. For this reason, the following relationship is hypothesized:
H3; Not-integrated public good creation is negative related to CFP
Figure 1; Research model
Research design
Procedure
To test the hypotheses, this research aims to conduct a quantitative databases research. Data for financial performance will be extracted from COMPUSAT. COMPUSAT is a database, which includes most information from firm’s financial reports of US publicly listed firms.
Data for CSR will be extracted from KLD index, which is developed and updated by MSCI and derives positive and negative environmental, social, and governance performance indicators from governmental and non-governmental databases and self-reported company disclosure. These indicators are grouped according to the social issues targeted by the indicator. In addition, the index includes multiple controversial concerns indicators. However, this research will disregard these indicators, since they do not fit within any of the dimensions targeted. The sample is formed by the combination of the two data bases (COMPUSAT & KLD), for
which observations could be coupled. Subsequently, only companies engaging in one or more of the relevant indicators are included.
Measurements
In previous research, multiple strategies have been adopted to measure financial performance. Both accounting-based measures, market-based indicators, and combinations are widely utilized (Dalton, Daily, Ellstrand & Johnson, 1998). As there seems no consensus on which indicators are more accurate presenters of financial performance, this paper prefers financial measures, as they are most frequently utilized when determining firm’s performance in terms of return to
shareholders (Shrader, Blackburn & Iles, 1997). The accounting based indicator is return on assets (ROA) (Saunders & Jones, 1992), which was calculated by dividing the total net income by the total assets.
To separate between negative externality reduction, integrated public good creation and non-integrated public good creation, one could attempt to find fitting categorizations within the KLD-index. For example, non-integrated public good creation largely includes philanthropy, and for research on philanthropy past research has utilized community KLD category (Marquis, Glynn & Davis, 2007). However, the KLD categories and indicators are dynamic concepts, changing over time. Therefore, such attempt to find equivalent categories, should be revalidated continuously. This paper will attempt such validation. Eight experts with interest and experience in CSR research allocated each KLD item to a category. These experts were approached via an email, including the first and last name of the expert, to increase interest. This resulted in the following group of experts: an alderman, two social entrepreneurs, two master students, two PhD students and one CSR facilitating organisation employee. Subsequently Anderson & Gerbing’s (1991) substantive validity coefficient is utilized to assess whether the highest numbers of allocations of a specific
item to a category is the valid assignment. The substantive validity coefficient is calculated by the following formula:
In this formula Csv is the substantive validity coefficient, nc is number of assignments to the
category, no is the number of assignments to the other category, and N is total number of
correspondents. Following the argumentation of Anderson & Gerbing (1991), eight correspondents requires the Csv value to be 6/8 or higher to be significant. This solely occurs if seven of eight
correspondents assign an indicator to the same category. Subsequently, this will result in three sets of KLD-items. When the individual items in each set are summed up, they will represent negative externality reduction, integrated public good creation and non-integrated public good creation. This resulted in the indicators per dimension illustrated in table 2.
Negative externality reduction
Integrated Public good creation
Non-Integrated public good creation Pollution Prevention Recycling Clean Energy Management Systems Strength (environment) Limited Compensation Plant, Property &
Equipment
Cash Profit Sharing Retirement Benefits Strength Employment of the Disabled Quality R+D-Innovation Benefits to Economically Disadvantaged Employee Involvement Beneficial products &
Services
Charitable Giving Innovative Giving Support for Housing Support for Education Volunteer Programs
Control variables
In addition to the variables described as measurements, this research includes control variables. One of these control variable is firm size, measured by the log of total assets (Lioui & Sharma, 2012). for two main reasons. First, firm size increases the visibility and stakeholder relations of the firm. Subsequently, increased visibility and stakeholder relations result in increased stakeholder management, as CSR (Orlitzky, 2001). Second, larger firms are able to operationalize economies of scale, which results in positive financial performance (Thompson, 1967).
A second control variable is R&D intensity, measured by R&D expense divided by total assets (Lioui & Sharma, 2012). McWilliams & Siegel (2000) emphasise the relationship between R&D intensity and financial performance, suggesting innovative firms have more opportunities for positive returns. Furthermore, they argue CSR is highly related to R&D, as CSR largely includes innovating process or products.
The third control variable is industry. McWilliams & Siegel (2001) argue the relationship between CSR and CFP is mediated by industry type. Industry conditions may vary significantly. For multiple industry factors can be argued, they influence the CSR-CFP relationship. For example, labour market conditions can influence the relation. High employee availability, will decrease the recruitment and retention cost, which make CSR activities inefficient as a recruitment strategy. Second, visibility, also considered relevant in the firm size control variable, may vary per industry. In addition, consumer market conditions, are considered relevant. If the consumers have high income, they value additional factors more, such as CSR aspects of a product. A fourth important influencing factor, stated by McWilliams & Siegal is industry lifecycle stage. In a maturity stage, competition largely becomes cost-based. Therefore, firms in this stage are less likely to engage in CSR activities. COMPUSAT includes the measure SIC code, which is used to differentiate among 10 main industry types in one dummy variable.
A last control variable is time. The data is from companies measured in the years between 2005 and 2009. During this time, macro-economic conditions has evolved dynamically. One particular event, the financial crisis of 2008, could have significant influence on the financial performance of the dataset. Therefore, using year as a control variable, will compare firms only in the same year. Therefore, the variance due to these type of macro-economic changes is accounted for.
Analyses and expectation
To test the hypothesis, a regression analyses will be conducted. This analyses will be conducted using SPSS for windows (version 20).
First, the regression offers insight on the relationship between the KLD-items combination representing negative externality reduction and CFP. Here, this paper expects to find a positive relationship. For instance, this would indicate companies engaging in eco-efficiency gains, will have higher financial performance.
Second, the regression will offer insight in the relationship of the KLD-items combination representing integrated public good creation and CFP. Here this paper expects to find a positive relationship. For instance, this will indicate firms providing opportunities for economically disadvantaged by a product characteristic, will have higher financial performance.
Third, the regression will offer insight in the relationship of the KLD-items combination representing non-integrated public good creation and CFP. Here this paper expects to find a negative relationship. For instance, this will indicate firms donating to charities, will have lower financial performance.
Results
Sample
The combination of the COMPUSAT and KLD database resulted in data from 811 firms, forming 2771 total observations from 2005 till 2009. The amount of employees working at the firm’s ranges from 0 till 2100 with an average of 30,54 (SD=103,83). The firms operated in different industries, with most in manufacturing industry (44,4%), finance, insurance and real-estate (25%) transport and public utilities (10,3%). Data was collected from all states in North America, with the largest groups of firms from California (18,3%), New York (9,7%) and Texas (8,5%).
Correlations
The variables used in this study are shown in Table 3. This table includes the means, standard deviations and correlations. This table helps to identify if something interesting is occurring for every variable. The mean and standard deviation of industry and year are disregarded, as they are categorical variables.
M SD 1 2 3 4 5 6 7
1. ROA ,020741 ,163788
2. Negative externality reduction ,4316 ,76086 ,110**
3. Integrated public good creation 1,05 ,759 ,041* ,264** 4. Non-integrated public good
creation ,44 ,698 ,081
** ,221** -,146**
5. Relative R&D ,0774 ,12262 -,649** -,120** ,020 -,135**
6. Firm Size 3,6060 ,83706 ,212** ,232** ,010 ,486** -,403**
7. Industry yes** yes** yes** yes** yes** yes**
8. Year yes** no No yes** ,yes* no yes**
The table provides useful information as most correlations observed are significant to some level. A notable observation is the highly significant relation of non-integrated and integrated public good creation. However, this does not indicate a direct relationship, because multicollinearity has been disregarded. In addition, year has no significant correlation with negative externality reduction and integrated public good creation. On the contrary, it is significantly correlated to non-integrated public good creation. This could indicate in this sample popularity for investments in non-integrated public good creation has changed between 2005 and 2009. Another notable outcome, is the lack of significant correlation between relative R&D and integrated public good creation. Especially since in the conceptual validation, the R&D indicator is found to measure integrated public good creation. To gain more insights, all these variables are put into four regression models, of which the results are illustrated in Table 4. The first model only includes the dependent variable ROA and the control variables. In the following models each time one proposed CSR dimensions is added. The numbers displayed here are the unstandardized regression coefficients, as well as the beta weights.
Model 1 Model 2 Model 3 Model 4
Dependent Variable ROA ROA ROA ROA
Coefficient SE Beta Coefficient SE Beta Coefficient SE Beta Coefficient SE Beta
Constant 31,595 5,589 33,278 8,229 33,227 8,214 32,954 8,230
Negative externality
reduction ,006 ,006 ,026 ,003 ,006 ,012 ,002 ,006 ,009
Integrated public goods
creation ,014
* ,006 ,052* ,014* ,006 ,052* Non Integrated public
goods creation ,005 ,008 ,014
Relative R&D -,979*** ,035 -,600*** -,991*** ,038 -,610*** -,998*** ,038 -,614*** -1,000*** ,039 -,615*** Firm Size ,026*** ,005 ,105*** ,028*** ,007 ,107*** ,027*** ,007 ,102*** ,025** ,007 ,096**
Industry no no no no no no no no no no no no
Year yes*** yes yes*** yes*** yes yes*** yes*** yes yes*** yes*** yes yes***
R² .442 .456 .459 .459
Outcomes
The first regression model, with an explained variance of 44,2%, partially supports are arguments for control variables. However, apparently industry does not have a significant relationship to return on investment in our sample.
The second regression model explains 45,6 % of the variance in ROA. There was hypothesized that investments in negative externality reduction would result in positive returns. However, H1 is not supported, as in the regression model no significant relation can be observed between negative externality reduction and CFP (β =.026, p > 0.05).
The third regression model explains 45,9 % of the variance in ROA. There was hypothesized investments in public good creation activities, which are integrated with private good creations, would result in a positive return. H2 is confirmed, as a weak positive relation can be observed between integrated public good creation and CFP (β =.052, p < .05). This effect indicates investments in CSR activities, which provide a positive impact and are integrated within other business operations, result in positive economic returns overall.
Lastly, the fourth regression model explains 45,9% of the variance in ROA. There was hypothesized investments in public good creation activities, that are not integrated with private good creations, would result in a negative return. However, H3 is also not supported, as in the regression model no significant relationship can be observed between non-integrated public good creation and CFP ((β =.041, p > .05).
Figure 2; Outcomes of research model.
Discussion
Implications
By observing one significant relationship in multiple CSR dimension, this research offers some implications. First, Hillman & Kiem (2001) found a positive relation between CSR activities towards primary stakeholders and firm market value. This research has already pointed at the relatedness of CSR activities towards primary stakeholders and integrated public good creation. The expected positive relation found in this research between integrated public good creation and CFP supports the notion of a positive return on CSR activities aimed at primary stakeholders. However, Hillman & Kiem (2001) also find a negative relation between CSR activities aimed at secondary stakeholders and firm market value. This research has already explained the
relatedness of CSR activities towards secondary stakeholders and negative externality reduction. The notion of a negative return on CSR activities aimed at secondary stakeholders could not be
CSR
Negative Externality Reduction Public Good Creation
Integrated Non-Integrated
CFP
β=ns β=ns β=0,052*supported, since unexpectedly there was no significant relation found for negative externality reduction.
Second, the fact that only integrated public good creation is significant in this sample, supports the argument that CSR-activities which are integrated with business processes and CSR activities which are not, have diverse influence on CFP. While some conceptual research has acknowledged this differentiation (Halme & Laurila, 2009) most empirical research seems to ignore it. CSR is mostly differentiated in terms of type of social issue addressed. Furthermore, although only a part of the KLD-indicators could be categorized by the experts in this research, a coherent line can be observed. This supports the notion that the proposed dimensions are distinct forms of CSR.
Third, this research adds to the inconsistency in the relation of CSR to CFP. The meta analyses of Orlitzky, Schmidt & Rynes (2003), found an overall positive relationship and a meta-analyses of Margolis & Walsh (2003) found mixed results. Both non-integrated public good creation and negative externality reduction are found to have no significant relationship to CFP in the sample. A relatively large sample, which only includes a relation between integrated public good creation and CFP, may suggest the conceptual model used, is not extensive enough. The non-significant dimensions may contain contradicting indicators in terms of financial return, and may require to be further differentiated.
Lastly, Godfrey (2005) argues attribution theory indicates philanthropic activities would result in increased moral capital, and thus financial returns. This research already explained the relatedness of philanthropic activities and negative externality reductions. Though it may be true attribution theory indicates some financial benefits of philanthropic activities, these financial benefits may in practise not rise above the initial investment in the philanthropic activities. Unexpectedly, the outcomes of this research could not reject the argument of Godfrey.
Limitations
Regarding limitations, especially the categorization method requires some attention. The experts used in this sample required to have some relevant experience regarding CSR. Therefore, master students were adequate experts. However, these students could have too less experience in CSR related research. In addition, also non-academics were allowed to be respondents, such as social entrepreneurs. However, academic definitions and perspectives can be radically different from non-academic approaches. Furthermore, the number of experts is limited. Eight is the minimum amount of experts to find significant results. Lastly, the experts were approached by emails and the survey was unsupervised. This could have decreased effort experts put into the survey. As a result of these limitations, this research was able to allocate 19 indicators in categories. However, an additional high amount of indicators was allocated to the same dimension by five and six of the eight experts. In this research these allocations were not significant allocations. However, if the limitations described above were accounted for, these indicators could have been categorized significantly. An increased indicator collection per CSR-dimensions may have resulted in significant relationships for not-integrated public good creation and negative externality reduction.
Another limitation that requires attention is the sample. The combination of COMPUSAT and KLD, was filtered to companies who scored on one of the relevant indicators of this research. However, this excludes a significant amount of companies who are involved in CSR-activities, which are not targeted by these indicators. This decreases comparability of this research to other research, in which all KLD indicators were included. In addition, our sample only included US-based firms. As CSR environment across the globe differs, the outcomes of our research is not directly generalizable.
Furthermore, this research adopted accounting indicators to measure firm performance. However, as mentioned before, there seems no consensus whether accounting based indicators are better representatives of financial performance in comparison to market based indicators. Therefore, if market-based indicators are better representatives of financial return, then the indicator ROA, used in this research, is not the adequate representative.
Lastly, this research included the control variables industry, year, relative R&D and firm size. Although these are the most frequently utilized control variables in research regarding the CSR-CFP relationship, there may be other unaccounted for variables affecting the relationship. Furthermore, multiple mediators and moderators, affecting the relationship have been proposed, which may have different effects in the different CSR-dimensions. As this research attempted to find a bases for the dimension differentiation, it goes beyond the scope of this research to speculate on those variables for each dimension
Practical guidance
First, this research observes firms invested in CSR-activities aimed at the creation of a public good, simultaneous with a private good, will generally have increased financial performance. Therefore, firms should identify such opportunities within their organisation. Evidently, a general effect does not indicate a positive return for each investment. Therefore, each identified opportunity within this dimension, should be evaluated individually, taken into account each factor that may influence the relationship with financial return. For the identification of CSR opportunities within this dimension, the indicators identified for the integrated public good creation dimension can be utilized as a starting point. For instance, firms can directly start “employment of disabled”. For other indicators firms have to be creative. For example, to pursue CSR activities targeted by the
indicator “beneficial products and services”, different activities can be implemented, dependent on the type of product and service.
Perhaps the most practical guidance this research provides, results from the conceptual model. The differentiation made by experts, and the difference found in the result, both support the view to approach the proposed dimensions separately. Firms interested in implementing a CSR-policy could use the dimensions to structurally brainstorm on and organize opportunities.
As this research only observed one of the three significant relationships, practical usability remains limited. However, non-significant relation in this sample, does not prove there is no relation between the dimensions. If other research finds the expectations to be true, then practical guidance would be more elaborate. For instance, this would indicate, it is not profitable to invest in an activity to create public goods, independent of private good creation activities. Therefore, managers would be less likely to find positive returns, and require to approach these opportunities with more caution. Furthermore, if all hypotheses found true, it would also indicate investments in the reduction of negative externalities would result in positive returns. Therefore, for every manager interested in increasing profit, especially by cost reductions, it would be wise to evaluate these opportunities for their company.
Conclusion
This research has attempted to answer the following research question: “How do negative
externality reductions, integrated public good creations and non-integrated public good creations affect CFP?”. Therefore, general CSR concepts and their proposed financial returns were analysed.
Subsequently, the proposed dimensions in this research were delimited, and compared to the CSR approaches of other research. Then, the measurements of the dimensions were determined and
validated using experts’ ratings. Lastly, the relation between these separated dimensions were analysed by correlation and regression analyses. This resulted in the following expectations and answers on the research question.
First, this research has argued for a positive relation of externality reduction and financial performance, because these activities cause reputational gains, risks reductions and eco-efficiency gains. However, this relationship could not be proven by the sample of this research.
Second, this research has argued for a positive relation of integrated public good creation and financial performance, because these activities also cause reputational gains, and are aimed at primary stakeholders, which are found to be profitable. This relationship was supported by this research.
Lastly, this research has argued for a negative relation of non-integrated public good creation, as these activities are independent of core business practices, and therefore outside of the expertise of managers. In addition, these activities are focussed on secondary stakeholders, which are found to be unprofitable. However, this relationship was not supported by the outcomes of this sample.
Future research
The outcomes of this research poses multiple opportunities for future research. A primarily opportunity is to improve the current research. First, the validation method to establish a measurement model for the individual CSR dimensions could be improved and elaborated. For instance, the requirements to pose as experts could be stricter. If experts would be only academic researchers, with more than five year experience, the indicator allocations would probably be more consistent. Also, the number of experts could be increased to enhance significance level. In
addition, it would be optimal if the experts would be supervised during the survey, to increase effort.
Second, the method used to gather data may be improved. If all KLD indicators can be categorized, the sample does not have to be filtered. Then, the sample size could be larger, because firms engaging in other types of CSR-activities could be included. Furthermore, future research could attempt to include small firms in the sample, or firms outside of the United States to increase applicability of outcomes.
Third, future research could include market based financial indicators, instead of only accounting based financial indicators. Other measurements for financial performance may provide new insights.
Lastly, besides improvement research, this research also poses opportunities for new research. Primarily the conceptual model proposed, may provide some insights in the inconsistency of the relationship between CSR and CFP. Future researches attempting to identify new moderators or mediators in the relationship, could use the proposed conceptual model to approach CSR dimensions separately, to increase usability of the results. For instance, Arora, & Dharwadkar (2011) found attainment discrepancy to be moderating the relationship of CSR and CFP. However, this moderating effect may be different for the different CSR dimensions proposed in this research. Therefore, when deviated in dimensions, managers and research can evaluate the potential of CSR activities more individually.
Literature
Anderson, J. C., & Gerbing, D. W. (1991). Predicting the performance of measures in a confirmatory factor analysis with a pretest assessment of their substantive validities. Journal of Applied Psychology, 76(5), 732–740
Arora, P., & Dharwadkar, R. (2011). Corporate governance and corporate social responsibility (CSR): The moderating roles of attainment discrepancy and organization slack.
Corporate governance: an international review, 19(2), 136-152.
Bagnoli, M., & Watts, S. G. (2003). Selling to socially responsible consumers: Competition and the private provision of public goods. Journal of Economics & Management Strategy, 12(3), 419-445.
Barnett M.L. (2007). Stakeholder influence capacity and the variability of financial returns to corporate social responsibility. Academy of Management Review 32(3): 794-816.
Barnett, M. L., & Salomon, R. M. (2006). Beyond dichotomy: The curvilinear relationship between social responsibility and financial performance. Strategic Management Journal, 27(11), 1101-1122.
Carroll, A. B., & Buchholtz, A. K. (1996). Ethics and stakeholder management. Cincinnati:
South-Western.
Dalton, D. R., Daily, C. M., Ellstrand, A. E., & Johnson, J. L. (1998). Meta-analytic reviews of board composition, leadership structure, and financial performance. Strategic management
journal, 19(3), 269-290.
Davis, K. (1973). The case for and against business assumption of social responsibilities.
Academy of Management Journal. 16, 312-322
Devarajan, S., Squire, L., & Suthiwart-Narueput, S. (1997). Beyond rate of return: reorienting project appraisal. The World Bank Research Observer, 12(1), 35-46.
Dixon-Fowler, H. R., Slater, D. J., Johnson, J. L., Ellstrand, A. E., & Romi, A. M. (2013). Beyond “does it pay to be green?” A meta-analysis of moderators of the CEP–CFP relationship.
European Commision (2011) Corporate Social Responsibility: a new definition, a new
agenda for action. Retrieved at 25 May 2016 from http://europa.eu/rapid/press-release_MEMO-11-730_en.htm
FASB (1993) Statement of Financial Accounting Standards No. 116. Retrieved at 25 May 2016 from http://www.fasb.org/resources/ccurl/770/425/fas116.pdf
Frederick, W. C. (2006). Corporation, be good!: the story of corporate social
responsibility. Dog Ear Publishing.
Freeman, R. E. (1984). Strategic management: A stakeholder approach. Boston: Pitman.
Friedman, M. (1962). Capitalism and freedom. Chicago: University of Chicago Press. Godfrey, P. C. (2005). The relationship between corporate philanthropy and shareholder wealth: A risk management perspective. Academy of management review, 30(4), 777-798.
Godfrey, P. C., Hatch, N. W., & Hansen, J. M. (2008). Toward a general theory of CSRs: The roles of beneficence, profitability, insurance, and industry heterogeneity. Business & Society. Halme, M., & Laurila, J. (2009). Philanthropy, integration or innovation? Exploring the financial and societal outcomes of different types of corporate responsibility. Journal of business
ethics, 84(3), 325-339.
Hillman, A. J., & Keim, G. D. (2001). Shareholder value, stakeholder management, and social issues: what's the bottom line?. Strategic management journal, 22(2), 125-139.
Inoue, Y., & Lee, S. (2011). Effects of different dimensions of corporate social responsibility on corporate financial performance in tourism-related industries. Tourism
Management, 32(4), 790-804.
Korhonen, J. (2003). Should we measure corporate social responsibility?. Corporate Social
Kotchen, M. J. (2006). Green markets and private provision of public goods. Journal of
Political Economy, 114(4), 816-834.
Lioui, A., & Sharma, Z. (2012). Environmental corporate social responsibility and financial performance: Disentangling direct and indirect effects. Ecological Economics, 78, 100-111.
Margolis, J. D., & Walsh, J. P. (2003). Misery loves companies: Rethinking social initiatives by business. Administrative science quarterly, 48(2), 268-305.
Marquis, C., Glynn, M. A., & Davis, G. F. (2007). Community isomorphism and corporate social action. Academy of management review, 32(3), 925-945.
Marshall, A., & Guillebaud, C. W. (1961). Principles of economics. London: Macmillan for the Royal Economic Society.
McWilliams, A., & Siegel, D. (2001). Corporate social responsibility: A theory of the firm perspective. Academy of management review, 26(1), 117-127.
McWilliams, A., Siegel, D. (2000). Corporate social responsibility and financial performance: correlation or misspecification? Strategic Management Journal 21(5), 603–609
Mirvis, P. (2012). Employee engagement and CSR. California Management Review, 54(4), 93-117.
Orlitzky M, Schmidt FL, Rynes SL. 2003. Corporate social and financial performance: A meta-analysis. Organizational Studies 24(3): 403-441.
Orlitzky, M. (2001). Does firm size comfound the relationship between corporate social performance and firm financial performance?. Journal of Business Ethics, 33(2), 167-180.
Patten, D. M. (2008). Does the market value corporate philanthropy? Evidence from the response to the 2004 tsunami relief effort. Journal of Business Ethics, 81(3), 599-607.
Roberts, P. W., & Dowling, G. R. (2002). Corporate reputation and sustained superior financial performance. Strategic management journal, 23(12), 1077-1093.
Sasse, C. M., & Trahan, R. T. (2007). Rethinking the new corporate philanthropy. Business
Horizons, 50(1), 29-38.
Saunders, C. S., & Jones, J. W. (1992). Measuring performance of the information systems function. Journal of Management Information Systems, 8(4), 63-82.
Schwaiger, M. (2004). Components and parameters of corporate reputation-an empirical study. Schmalenbach business review, 56, 46-71.
Shrader, C. B., Blackburn, V. B., & Iles, P. (1997). Women in management and firm financial performance: An exploratory study. Journal of managerial issues, 355-372.
Thompson, J. D. (1967). Organizations in action: Social science bases of administrative
theory. Transaction publishers.
Van Oosterhout, J., & Heugens, P. P. (2006). Much ado about nothing: A conceptual critique of CSR. ERIM Report Series Reference No. ERS-2006-040-ORG.
Visser, W. (2010). The age of responsibility: CSR 2.0 and the New DNA of business.
Journal of business systems, governance and ethics, 5(3), 7.
Vlachos, P. A., Panagopoulos, N. G., & Rapp, A. A. (2013). Feeling good by doing good: Employee CSR-induced attributions, job satisfaction, and the role of charismatic leadership.
Journal of Business Ethics, 118(3), 577-588.
Wang, T., & Bansal, P. (2012). Social responsibility in new ventures: profiting from a long‐ term orientation. Strategic Management Journal, 33(10), 1135-1153.
Wang, Y., & Berens, G. (2015). The impact of four types of corporate social performance on reputation and financial performance. Journal of Business Ethics, 131(2), 337-359.
