What do we know about corporate social responsibility research? a content analysis

There has been an increasing use of content analysis as a rigorous way of exploring many important but difficultto- study issues (Duriau et al., 2007). Early definitions emphasise the quantitative approach by stating that ‘content analysis is a research technique for the objective, systematic and quantitative description of the manifest content of communication’ (Bereson, 1952:55). The more recent definitions are broader in scope and view content analysis as a class of methods within empirical social science that can be applied both in a quantitative and a qualitative way (Seuring and Gold, 2012). Therefore, a particular strength of content analysis is that this technique can combine quantitative approaches (e.g., frequency counts, correlations, trends and differences over time), while retaining rich meaning with powerful qualitative analysis (e.g., theme identification, theory elaboration) (Duriau et al., 2007; Laplume et al., 2008). The present study adapts this broader approach for several reasons. Firstly, content analysis is repeatable (Cascio and Aguinis, 2008) and, therefore, helps to summarise the intrinsic value of the information, without the disadvantage of bias. Secondly, content analysis allows an assessment of the effects of corporate context (Kolbe and Burnett, 1991), because CSR strategy and the consequences of CSR activities have significantly different motivations (e.g., due to different institutional, industrial and organisational effects). Thirdly, this method is highly flexible (Cascio and Aguinis, 2008), which makes it more efficient. CSR involves various fields (e.g., marketing, human resource management and stakeholder management) that are difficult to categorise without the use of this approach. Finally, content analysis has been previously used successfully to accumulate knowledge and conduct literature reviews in management and organisation studies (Duriau et al., 2007; Cascio and Aguinis, 2008; Laplume et al., 2008, Seuring and Gold, 2012).

The aim of the journal selection process was to identify leading journals that publish CSR articles and make contributions to the development of CSR research. Following Laplume et al. (2008), we chose the same seven leading journals in management (Academy of Management Journal; Academy of Management Review, Administrative Science Quarterly, Strategic Management journal, Organization Studies, Journal of Management Studies and Journal of Management). We used Journal of International Business Studies (JIBS) instead of Organization Science as no direct CSR-related articles were found in Organization Science during the defined period and JIBS has the same ranking as the seven leading management journals (Harzing, 2010; ABS, 2010). Following Laplume et al. (2008), we also selected the three leading specialist journals devoted to business ethics (Journal of Business Ethics; Business Ethics Quarterly; and Business & Society). In addition, we included another specialist journal Business & Society Review, as a significant number of CSR articles (80) were found in this journal. Because of the close connection between CSR and marketing, some CSR papers are likely to be found in marketing journals. Therefore, three important marketing journals were also included (Journal of Marketing; Journal of the Academy of Marketing Science; and International Journal of Research in Marketing). Finally, three mainstream managerial practice journals were also selected: California Management Review, Harvard Business Review and Sloan Management Review.

In our selection of papers, we reviewed definitions of the key term ‘corporate social responsibility’ and of terms that resemble CSR, such as ‘corporate social performance’, ‘corporate citizenship’ and ‘corporate philanthropy’. We selected articles with these key words in their titles, abstracts or contents. For thoroughness, we also referred to bibliographies of numerous reviews of CSR, such as a review of previous studies regarding the linkage between CSP and CFP (Orlitzky et al., 2003), a review of CSP measurement (Wood, 2010) and a review of CSR (Lee, 2008). The final CSR sample database (Appendix A) consisted of 774 articles, including 122 articles from leading academic journals, 68 from leading managerial practice journals and 584 from specialist journals.

We used these 774 articles to analyse the CSR publication trends (Figure 1 and Appendix A). As it would be excessively time-consuming and arguably unnecessary to conduct detailed analysis of all the 774 articles, we considered it reasonable to take the 122 articles from leading academic journals, as being representative of the main trends in CSR research. Therefore, only those 122 articles were used for the purpose of all further analysis.

Figure 1

Following the classification method by Laplume et al. (2008) and Cascio and Aguinis (2008), we categorised the articles in the database according to themes. We coded every CSR academic article (n=122) and, based on the number of articles published around the topics, the top five broad categories were identified and can be ranked as follows: concepts and reviews, corporate context, CSR-related strategy, corporate reputation, and links between CSP and CFP (Table 1). Following Cascio and Aguinis’s (2008) approach, in order to improve the reliability of the taxonomy of the content analysis, we attempted to map sets of categories on to each other, regardless of their previous categorisation, with the aim of challenging the existing categories. Because our study is the first to produce a categorical content analysis in CSR research, we cannot claim that no other taxonomy is possible, or that our classification is based on a comprehensive and systematic set of codes.

We present our findings in the following order: first, the CSR development trend is introduced, then the four stages of CSR development and pivotal articles for each stage are explained, followed by a discussion of the five major research themes.

From 774 CSR articles selected, a number of trends were identified. Figure 1 shows that the annual rate of publication of articles in the CSR field has generally been increasing, particularly in the 2000s, with a significant upward trend in the decade from 2001 to 2010. Figure 1 also indicates that the number of CSR articles published annually in the leading academic journals has increased rapidly since the 1970s, with the number of publications reaching a peak every 10 years.

We organised the 122 articles into four stages and labelled them as: incubation (1949–1979), exploration (1980– 1989), adolescent development (1990–1999) and relative maturity (2000–2010). The use of these time periods may appear arbitrary, but it allows for the demonstration of the underlying trends. Because the first period coincides with the initial stage in the development of research and discussion on CSR, it contains a number of works that offer rudimentary explanations of the concept. During the first time period (the incubation stage), early discussions concerning CSR can be found in some managerial practice journals and book chapters. These early studies were primarily designed to increase awareness of social and ethical principles among the business community (Carroll, 1979). During this process of theory-building, several areas were explored in CSR. For example, one CSR structure was proposed by Carroll (1979), while Alexander and Buchholz (1978) discussed the relationship between CSR activities and stock market performance and Abbot and Monsen (1979) suggested a method for measuring CSR. It is also during this period that Friedman’s (1970) controversial article ‘The Social Responsibility of Business is to Increase its Profits’ spurred much future work on CSR. Although there are insufficient empirical and theoretical studies to support comprehensive CSR research at this stage, the studies published during the incubation period provide a foundation for its future understanding.

We describe the second period (1980–1989) as the exploration stage because it is a period in which researchers attempted to verify or support the concepts discussed during the first stage. These attempts included initial efforts to suggest a yardstick for monitoring the understanding of CSR (Tuzzolino and Armandi, 1981), reviews of previous studies to verify and develop the structure of CSR and CSP (Boal and Peery, 1985; Wartick and Cochran, 1985), and the use of Fortune magazine’s corporate reputation ratings to examine the relationship between CSP and CFP (McGuire et al., 1988).

During the third period (1990–1999), the exploration stage, CSR scholars attempted to improve the structure of the concept of CSR at the theoretical level and to enhance the reliability and validity of CSR measurements at the empirical level. CSR received great attention during this stage, with a total of 27 articles being published in leading academic journals. These articles enlarged and consolidated the theoretical and empirical foundations of earlier CSR studies, by employing combinations of other mature themes such as stakeholder theory (Freeman, 1984), marketing themes (Menon and Menon, 1997) and the resource-based view (Barney, 1991). For instance, Freeman’s (1984) influential stakeholder theory came into prominence in the mid-1990s, while the social responsibility element of the theory began to emerge in discussions about business ethics, social issues in management and sustainable development (Laplume et al., 2008).

The fourth stage (2000–2010) is termed the relative maturity stage. The CSR articles published during this stage are well developed in comparison with those from earlier periods. Some researchers emphasised that the analysis of firms’ CSR issues should be conducted in a societal environment (Campbell, 2007; Scherer and Palazzo, 2007), while other researchers continued to show interest in the outcomes of CSR (McWilliams and Siegel, 2000,Hillman and Keim, 2001). In particular, Orlitzky et al. (2003) utilised a meta-analysis to examine 52 previous studies of the link between CSP and CFP, and concluded that CSP was more highly correlated with accounting-based than marketing-based instruments in the prediction of CFP indicators. Although there is still no universally accepted theoretical conception of CSR, this last study at least represents an approach to the question of how to develop CSR research.

Several key or pivotal articles for each stage (Table 2) were identified in the SSCI according to their citation record. The articles from the 1990s and the 2000s are cited 10 times per year on average, while those from the 1980s are cited five times per year on average. Several articles published in the 1970s and earlier also have relatively high citation rates. For instance, at the time of writing Alexander and Buchholz’s work (1978) had been cited 119 times according to the Web of Knowledge citation report (Web of Knowledge, 2013). It should be noted that some publications are also pivotal but were not included in the analysis as they were not published in top academic journals (e.g., Friedman, 1970; Freeman, 1984).

Figure 2 illustrates the number of articles in leading academic journals according to each theme: concepts and reviews, corporate context, CSR-related strategy, corporate reputation and the CSP–CFP relationship. These five themes are discussed further in the following sections.

Figure 2

There have been many different definitions of CSR, for instance, Dahlsrud (2006) identified 37 definitions of CSR. Carroll and Shabana (2010) suggested that even this number may be underestimated, because many academically derived definitional constructs were not included in their study due to the scope of their methodology. A number of articles have discussed the CSR definitions in detail (cf. Carroll, 1999; Dahlsrud, 2006; Carroll and Shabana, 2010). Here we have summarised the CSR concept using the following common perspectives (see Table 1): (1) an integrative perspective, namely, those definitions generated from a spectrum of views ranging from stakeholder concerns to social issues. For instance, some scholars believe that CSR involves corporate activities because they relate to its perceived societal or stakeholder obligations (Sen and Bhattacharya, 2001; Luo and Bhattacharya, 2006); (2) an economic perspective, espoused by those researchers who believe that the aim of CSR activities is the maximisation of profit, which emphasises the benefits to shareholders and internal stakeholders. For example, Friedman (1970) believed that the purpose of CSR is to make as much money as possible for a firm’s shareholders; (3) a voluntary perspective, espoused by those researchers who believe that firms should take some philanthropic action in accordance with local laws to improve social welfare (Carroll, 1991); and (4) a public perspective, where scholars argue that a CSR strategy should be designed to fit the social system (Steiner and Richman, 1971; Buchholz, 1977; Sethi, 1979; Smith, 2003).

Noda and Bower (1996) noted that corporate context is a reinforcement or modification of corporate strategic initiatives. Using this as a starting point for the proper classification of corporate context orientations in our sample of CSR articles, we examined each article in the database and categorised them using the key word ‘corporate context’. Related articles were then identified, that focused on one of three levels, termed ‘individual’, ‘organisational’ and ‘societal’ (Table 3).

Individual perspective. Previous studies have briefly discussed the effects of the personal characteristics of top managers on corporate CSR activities or CSP. For instance, Deckop et al. (2006) found that the more a corporation made use of a long-term focus in the remuneration of its chief executive officer (CEO), the higher the ranking of the firm’s CSP. Swanson (1999) suggested that the personal values of senior managers strongly influence their formulation of corporate social policy. Moreover, Agle et al. (1999) found that there were significantly positive relationships among stakeholder attributions, the personal values of a firm’s CEO and a firm’s CSP. In summary, a leader’s personal values shape managerial perceptions through his or her interpretation of external information, and these values are reflected in a firm’s strategy and, in turn, in its overall corporate performance (Hambrick and Mason, 1984).

Organisational perspective. This level involves institutional investors and owners, including pension funds, mutual funds, investment bankers, insurance and investment firms (Chaganti and Damanpour, 1991). The starting point in this area is Graves and Waddock’s (1994) work, which found that, the larger the number of institutions that hold a company’s shares, the better the firm’s CSP. Johnson and Greening (1999) analysed the effects of different types of institutional investors on CSP and advanced the idea that pension fund equity and the presence of external directors were positively correlated with the people and product quality dimensions of CSP. Furthermore, they found that top management equity also had a positive impact on the product quality dimension, but had no correlation with the ‘people’ dimension of CSP. David et al. (2007) found that shareholder proposal activism had a negative impact on CSP. Institutional investors and owners not only have interests in corporate financial returns, but also pay attention to strategies and activities, relationships with other stakeholders, and corporate sustainable development. As a result, they have a great deal of influence on corporate CSR strategy and CSP (Holderness and Sheehan, 1988; Pound, 1992).

Societal perspective. Social issues can lead to external pressures unless firms are aware of them and incorporate them into their CSR strategy. A number of CSR researchers have written articles on the societal perspective. Their studies can be categorised in terms of different regions, industries and institutions. Maignan and Ralston (2002) found that firms in different countries displayed different levels of engagement in CSR. Matten and Moon (2008) provided an exploration of why CSR activities differ among countries, and how firms should adapt to the local environment. Furthermore, in their review of previous research, Scherer and Palazzo (2011) argued that there was a trend of ‘global governance’ and for the concept of CSR to become political during the process of globalisation. Muller and Kolk (2010) examined CSR activities in the internal business area and found that, compared with extrinsic factors (e.g., trade), intrinsic components (e.g., management commitment to ethics) were dominant drivers of CSP among both foreign and local firms. From an industrial perspective, Bhambri and Sonnenfeld (1988) compared the insurance and forestry industries and suggested that the differences in their institutional environments contributed to various components of structuring public issues in a company’s CSR strategy and, thus, influenced their CSP.

From the institutional perspective, based on 10 years of fieldwork, Arya and Zhang (2009) stated that institutional reforms have increasingly promoted an awareness of CSR and have increased support for CSR activities, while such reforms have also promoted changes in corporate social actions. In other empirical studies based on an institutional approach, attempts have been made to elaborate on the determinants (e.g., firm size, managerial discretion and institutional pressure) (Greening and Gray, 1994; Husted and Allen, 2006), mediators (Campbell, 2007) and strategic investment (Gardberg and Fombrun, 2006) in CSR.

The resource-based view (RBV) (Barney, 1991) is a useful perspective for researchers of CSR strategy, because it helps to transfer actionable prescriptions to practitioners (Priem and Butler, 2001). A large number of CSR studies are based primarily on the RBV (Russo and Fouts, 1997; McWilliams and Siegel, 2001; McWilliams et al., 2006). The RBV is used to account for the intangible assets created by CSR strategy, such as a positive corporate image (Gardberg and Fombrun, 2006), reputation (Hall, 1992), and customer satisfaction (Luo and Bhattacharya, 2006). Our analysis of CSR articles identified the following four perspectives of CSR-related strategy: risk management, environmental management, marketing management and the management of stakeholder relationships (Table 3).

Risk management. Using secondary data, Luo and Bhattacharya (2009) found that CSP could improve shareholder value by lessening undesirable firm-idiosyncratic risk and, consequently, suggested that CSP was a type of investment that is similar to advertising and R & D. Godfrey et al. (2009) found that investments in CSR were aimed at the secondary stakeholder and that society protects firms from potential risk (e.g., negative judgements and sanctions). In addition, the analysis of CSR activities in relation to corporate trading partners shows that these CSR initiatives have no benefit in terms of firm performance (Godfrey et al., 2009).

Environmental management. There has been an increasing awareness of environmental regulations and green management in recent years (McGee et al., 1998; Starkey and Crane, 2003). Menon and Menon (1997) put forward a strategic framework to evaluate the importance of environmental concerns, social performance goals and marketing strategy. Furthermore, Russo and Fouts (1997) provided statistical evidence from 243 firms to demonstrate that a firm’s environmental performance is positively correlated with CFP.

Marketing management. The 1970s marked the starting point for CSR research in the marketing field because at this point the researchers began to argue that CSR provides a means to ensure the survival of a firm (Gelb and Brien, 1971). Several years later, the concept of business ethics began to be used to promote the combination of social responsibility and marketing strategy (Murray and Montanari, 1986; Robin and Reidenbach, 1987), leading to the emergence of cause-related marketing (Varadarajan and Menon, 1988). Later studies in this area emphasise two aspects: corporate context and consumer orientation. As far as corporate context is concerned, CSR scholars suggest that acceptance by society of a company’s CSR involvement has a minimum threshold, below which any profit-maximising actions will be hindered (Handelman and Arnold, 1999). Consumers may be a significant component in the corporate context, since there is evidence that consumer satisfaction mediates the association of CSR actions with market value (Luo and Bhattacharya, 2006). Later findings showed that the response of consumers to a firm’s CSR activities was positive when they found those actions to be strategic or value driven (Ellen et al., 2006). Consequently, CSR researchers have interests in a range of questions including how firms communicate with consumers, how firms disseminate product information, and the manner in which firms build corporate brand dominance (e.g., Simmons and Becker-Olsen, 2006; Biehal and Sheinin, 2007; ) and maintain sustainability (Crittenden et al., 2010; Chabowski et al., 2011; Hult, 2011) through their CSR efforts.

Stakeholder relationships. Stakeholder management is invariably associated with a CSR programme and corporate performance (Harrison and Freeman, 1999). In addition, the stakeholder environment determines the extent of a corporate CSR initiative (Doh and Guay, 2006). Researchers also place emphasis on communication with stakeholders (Basu and Palazzo, 2008), and some scholars indicate that CSR awareness among stakeholders has a positive effect on a firm’s sales income as well as an influence on investment and employment (Sen et al., 2006). In particular, Choi and Wang (2009) offered evidence that high levels of stakeholder awareness facilitated a firm’s corporate ability both to sustain good financial performance and to recover from a substandard financial performance more quickly.

Roberts and Dowling (2002) argued that an important aspect of CSR-related strategy involves sustaining or improving corporate reputation, because a growing body of research now suggests that a positive corporate reputation has the potential to create value. From the resource-based view, an excellent corporate reputation is a type of intangible asset, and cannot easily be replicated by other corporations (Roberts and Dowling, 2002). Previous empirical studies demonstrate that a good corporate reputation has a positive impact on CFP, which we consider further in the following section. In this section, we focus on a corporation’s reputation and its attractiveness to employees and consumers (Table 3).

With regard to a company’s attractiveness to employees, corporate reputation and perceived external prestige are positively correlated with employees’ commitment to their firms (Hunt et al., 1989; Turban and Greening, 1997; Carmeli, 2005). Furthermore, Hunt et al. (1989) found that organisational commitment had a positive impact on other valuable outcomes such as employee satisfaction, employee performance, corporate loyalty and adaptability. There is also evidence that the impact of negative CSR actions can be very damaging to a company’s image (Vlachos et al., 2009). There is also a trend in the communication of a corporation’s CSR values to its consumers wherein positive consumer attitudes contribute to the sales of a firm’s products (Lichtenstein et al., 2004; Wagner et al., 2009).

We earlier noted that the ranking of corporate reputation is an effective means of measuring CSR outcomes. Similarly, we discussed the notion that some systematic measurements of CSP reflect the results of CSR actions. There is an ongoing debate about the linkage between CSP and CFP, because theoretical and empirical studies support a number of positions that are often contradictory (e.g., Aupperle et al., 1985; McWilliams and Siegel, 2000; Godfrey, 2005; ). In order to summarise the conclusions reached in previous studies and to reveal the underlying link between CSP and CFP, Orlitzky et al. (2003) conducted a meta-analysis of 52 studies and concluded that CSP had a positive relationship with CFP across all industries and within all corporate contexts (Margolis and Walsh, 2003).

Regarding measurements, Orlitzky et al. (2003) found that the correlation between CSP and CFP was more obvious if accounting-based firm performance data was used, rather than marketing-based data. However, Barnett and Salomon (2006) caution that the statement of Orlitzky et al. (2003) was based only on a compilation of existing evidence, and, therefore, cannot yield any persuasive conclusions, since the 52 previous works were all flawed in various ways. Furthermore, Peloza (2009) examined 159 studies in the CSP-CFP linkage and proposed a figure of ‘stages of financial impact from corporate social performance’ to illustrate the manner in which CSP influences CFP. Barnett and Salomon (2006) developed the concept of social screening, which was used by CSR investment managers to measure the level of a company’s CSP. Based on their analysis using social screening, they indicated that financial returns initially declined, but there was a rebound effect when the firm improved its level of social screening (Barnett and Salomon, 2006). Subsequently, Brammer and Millington (2008) conducted an investigation of the CSP–CFP link and concluded that it was curvilinear that firms with unusually high or low CSP may have a higher CFP, but that firms with unusually low CSP only promised short-term financial returns, while firms with unusually high CSP promoted more long-term financial returns. Recent studies of US corporations have shown that charitable contributions stimulate revenue growth, especially after controlling for sales growth (Lev et al., 2010). Overall, despite all the attention paid to the link between CSP and CFP, the nature of this linkage remains uncertain.

Another debate in the literature concerns the causal link between CSP and CFP. Orlitzky et al. (2003) state that CSP and CFP are more likely to be mutually influenced by each other because firms with a good CFP can afford more CSR activities that lead to a better CSP. Meanwhile, a good CSP helps companies to achieve better financial returns. Although the data of Orlitzky et al. (2003: 417) show that ‘a virtuous cycle with quick cycle times or concurrent bidirectionality’ occurs, Barnett and Salomon (2006) argue that this compilation of data does not represent a definitive conclusion. Furthermore, they elaborate that ‘a compilation of findings cannot produce a definitive conclusion given the limitations of the underlying studies’ (Barnett and Salomon, 2006: 1105). The empirical data of Lev et al. (2010) demonstrate that the growth of CFP has a significant association with future charitable contributions. Surroca et al. (2010) found that CSR activities stimulated the development of corporate intangible resources (e.g., innovation, human capital, reputation and culture), which subsequently improved financial performance. Similarly, Doh et al. (2011) found that institutional intermediaries influenced market assessments of a firm’s social responsibility and highlighted the importance of the legitimacy-conferring function of expert bodies in the analysis of the link between CSP and CFP.

There is a growing body of research on the linkage between CSP and CFP that emphasises different factors. For instance, some research on building a CSP–CFP model examines the purchasing behaviour of consumers (Schuler and Cording, 2006), investor preferences (Mackey et al., 2007) and corporate responsiveness to employees (De la Cruz Deniz-Deniz and Saa-Perez, 2003). There are a large number of studies concerning the CSP–CFP link (20 per cent of the total), which reflects the fact that the pursuit of a good CFP is a vital motivation for firms to engage in CSR activities.

In sum, the objective of this section was to provide a general overview of the five major CSR research themes that this study has identified, namely: concepts and reviews, corporate context, CSR-related strategy, corporate reputation and the CSP–CFP relationship. It is worth noting, that not every article selected is discussed in-depth in the paper. Further discussions and synthesis of the literature will no doubt provide a richer and more comprehensive understanding of CSR studies.