The Pyramid of Corporate Social Responsibility ( CSR ) : An Empirical Examination in the Tunisian Context

The concept of CSR was subject to multiple approaches and research perspectives (Gond & Igalens, 2008). Its application in the Anglo-Saxon context revealed a lack of consensus on identifying its dimensions (Auperles et al., 1985). In this study, we will try to empirically explore this concept in the Tunisian context. In this regard, a quantitative questionnaire-based survey of Tunisian companies highlights the four-dimensional nature of social responsibility, with philanthropic responsibility standing out as the most influential dimension in CSR perception in Tunisia.


Introduction
With the advent of globalization and the development of international trade, the concept of corporate social responsibility (CSR) has been receiving an increasing acclaim.The most popular definition of CSR is the one proposed by the European Commission in its 2001 Green Paper; "companies' voluntary integration of social and environmental concerns in their business operations and in their interaction with their stakeholders." However, most research has focused on the nature of CSR in developed countries, yet little research has examined the same concept in developing countries.
In this paper, we will examine CSR in Tunisia.We will try to determine the concept of CSR in Tunisian companies using the model of Carroll (1979).
Our approach consists of a quantitative questionnaire-based survey of managers of a sample of Tunisian companies.First, we will present the CSR concept while showing the absence of a universal conception of it.Then, we will develop an overview of the most common representations of CSR.Second, we report on an empirical analysis as part of a quantitative study of a sample of Tunisian companies.

Literature Review
Despite the important contributions of recent decades, there is still no precise definition of CSR.Several studies have been developed on this topic; the most influential are those of Bowen (1953) and Carroll (1979).
There are several theoretical approaches to social responsibility (Gond & Igalens, 2008).Howard Bowen is considered the founding father of the concept of social responsibility.In his book "Social Responsibilities of the Businessman", the author defined managers' CSR as "a series of obligations resulting in a series of policies, decisions and course of actions consistent with the objectives and values of society" (Germain Trébucq, 2004).His book aims at educating executives the values "considered desirable in our society." Milton Friedman (1970) chooses a different conception of CSR as advanced by his predecessors.Indeed, the author considers the social responsibility of any business as the achievement of gains for shareholders "The social responsibility of business is to increase its profits".
It was not until 1979 when Carroll advocated a conception of CSR that takes into account the recommendations of those affirming the need for companies to be aware that they are responsible for their business towards all their stakeholders, not just shareholders and monetarists like Friedman, advocating thus a CSR based on the creation of value for the company.
According to Carroll (1979), Corporate Social Responsibility should be approached in three ways: Organizational responsiveness, the different categories of CSR and social concerns of the organization.
Carroll defines the concept of CSR as a set of obligations vis-à-vis society.The author distinguishes four types of obligations: economic (be profitable, manufacture goods complying with quality standards, ..), legal (compliance with laws and regulations), ethical (act according to moral principles shared by society) and philanthropic (benevolent actions and charity).According to the author, CSR is "the set of obligations that the company has vis-à-vis society (including) economic, legal, ethical and discretionary categories".This definition has been taken by the European Commission in 2001 which considers CSR as "the voluntary integration by companies of social and environmental concerns in their business operations and in their interaction with their stakeholders." The American literature remains the most exhaustive on the concept of CSR.Indeed, several models have been developed to provide a conceptual framework for social responsibility, like Wood (1991), Watrick and Cochran (1985), Clarkson (1995) Quazi and O'Brien (2000).

Conceptual Model of CSR
The work of Carroll (1979) represents the first real conceptual model of CSR.Indeed, the author proposes a typology of responsibilities in the shape of a pyramid in which economic obligations are the basis of responsibility.Second, we find legal responsibilities followed by ethical responsibilities and at the peak of the pyramid we find philanthropic responsibilities.
Economic responsibility refers to the traditional economic role of a company, which is essentially the production of goods or services under profitable conditions.In fact, the company is expected to first ensure its continuity on the market, create jobs and pay taxes.Definitely, by this economic dimension, the role of the company would meet the requirements of its stakeholders in this case, shareholders, employees, consumers and finally the State.
The second tier of the pyramid is legal responsibility, which refers to the obligation to comply with the laws in society.According to Carroll (1979), the company should achieve its economic responsibilities while respecting the legal and institutional framework.
Ethical responsibility refers to obligations that members of society would expect companies to assume yet they are not codified by laws.In this sense, the company should act honestly towards society even in the absence of legal constraints.
Finally, at the peak of the pyramid, we find discretionary responsibility i.e. discretionary actions of individuals like philanthropic activities.Capron and Quairel (2007) highlights the fundamental differences between the American and European conceptions of CSR, based on different cultural values, unique intellectual influences and the specific structure of the company.In particular, according to the authors, the American conception of the company is contract-based while the European design is institution-oriented. Indeed, the American concept of CSR is based on ethical and religious considerations (Aggeri et al., 2005).In this sense, CSR is summarized by many American companies as foreign philanthropic actions meant to tarnish their business activities (Capron & Quairel, 2007).

CSR under Its Different Perspectives
On the other hand, and in terms of research on CSR, empirical tests of Caroll's model ( 1979) point to differences between countries.Indeed, the study of Maignan and Ferrell (2003), whose objective is to identify consumers' representations of CSR, confirms the superiority of the legal dimension followed by the ethical dimension in the French and German contexts.
The primacy of economic responsibilities was highlighted by Pinkston and Caroll (1996) in several European countries like England, France, Germany, Switzerland and Sweden, but also in Japan and the USA.
In the Tunisian context and despite the predispositions of Tunisian managers to engage in a socially responsible perspective, the practice of CSR remains very shy.Indeed, according to a study under the Development Partnership with the Private Sector framework (GIZ-DPP) in 2012, the perception of CSR by Tunisian companies is "partial and differentiated and responds more to immediate concerns resulting from an external constraint or an expected benefit than to a long-term vision".
Furthermore, the study suggests the existence of major difficulties in the implementation of CSR.These barriers mainly relate to Tunisian managers' lack of conviction, but also to a cultural resistance to the "very principle of accountability".The study also puts the emphasis on the difficulties associated with means.Indeed, Tunisian firms suffer from a skills shortage, both internally and externally, and does not have the financial resources for the implementation of a CSR policy.
Academically, research on CSR in Tunisia is still scarce.Although the Tunisian context seems to be a fertile ground to study CSR because of the existence of several multinational companies and the cultural specificity of the Tunisian society, which is primarily centered on Islamic morality, conceptual contributions remain very limited (Bousourra & Zribi, 2008).

Methodology
Carroll's typology is a reference model in the literature on Corporate Social Responsibility (Jamali, 2008;Capron & Quairel-Lanoizelée, 2007;Igalens, 2008;Maignan et al., 1999).Carroll's model (1979) shows that CSR has four dimensions: economic, legal, ethical and philanthropic (or discretionary) responsibility.These four dimensions were operationalized by Maignan et al. (1999).More other studies (Lin et al., 2010;Maignan & Ferrell, 2003) empirically adopted the four-dimensional nature of CSR.From there, it is necessary to study CSR as a set of four variables.
Our study's objective is to determine how Tunisian leaders integrate CSR into their businesses using Carroll's pioneering model of CSR as a framework.
We will try, through a "questionnaire-based" quantitative study as a data collection method, to answer the following research question: What is the most influential CSR dimension adopted in Tunisia?

The Measurement Scale of CSR
To measure CSR practices in Tunisia, we used the scale of Maignon et al. (1999), which has already been adopted by several scientific studies.This scale, which is in line with that proposed by Aupperle et al. (1985), and which takes into account the four-dimensional nature of CSR (the economic, legal, ethical and philanthropic dimensions) replicates Caroll's model (1979).Our questionnaire reproduces the 18 items of the scale.However, we will proceed with adjustments to the Tunisian context according to the methodology of Besson and Haddjadj (2003).
In our study, we will use a factor analysis to determine the main factor structure likely to reproduce the dimensions of Caroll's model (1979).We will interpret the percentages of explained variances to detect the most influential responsibility dimension of CSR in Tunisia.
In another hand we will use the multiple linear regression analysis to determine with more precision the order of responibilities.
-Independent variable: CSR ( Y ) -Dependant variables: Economic component (ECO)-Legal component (LEG)-Ethical component (ETH) -Philanthropic component (PHIL) Model: We conducted an exploratory survey of a total sample of 140 unlisted Tunisian companies belonging to different sectors and having different sizes (industry, business + services).In this regard, we administered a questionnaire to line managers (61% administrators, 17% technical managers, 22% managers) between April and June 2014.A total of 110 companies is selected with a response rate of 78% and 330 retained questionnaires (some questionnaires were eliminated because of negative correlations between respondents from the same company).
The responses from the same companies were aggregated around the average to form a score per company (El benna & Child, 2007;Gherib, 2011).Cronbach's alpha coefficients pointed to the most representative items of the dimensions.The percentage of variance explained indicates which of the four components that contributes most in explaining the construct CSR.

Results of the Tests and Discussion
We find that the philanthropic dimension is highly represented by the item "partnerships with local associations, institutions and schools" with a coefficient of about 0.727 and an explained variance of 42.19% for the same component."Respect of legislation on employees' salaries and recruitment" is the most representative of the legal dimension (Cronbach = 0.763 and an explained variance of 27,10%).
The economic dimension is highly represented by the item "profit maximization" with a Cronbach's alpha equal to 0.640 and an explained variance of 17.37%.Finally, the ethical dimension is represented by the item "a company which we can trust" (Cronbach = 0.696 and explained variance = 13.34%).The regression results confirm the predominance of the philanthropic dimension in the CSR construct followed by the legal and the economic dimension.However the ethical component is classified at the last position.
Indeed in the Table (2) the social responsibility of Tunisian companies is measured primarily by philanthropic responsibility.This clearly indicates that philanthropic contributions are important in the construct of CSR in Tunisia (coeff = 2,713).The dominant role of legal responsibility, which comes second (ceoff = 2,591), would suggest that there is a strong corporate commitment to respect the law and comply with the laws regulating workers' rights.
Rule of law is crucial when it comes to corporate social responsibility since laws represent a process of codifying the moral values that are endorsed by society (Crane & Matten, 2004).In third place, there is economic responsibility (coeff = 2,120), which does not seem to be of a particular interest for Tunisian companies.This could be explained by the economic downturn experienced by Tunisia in recent years.
The ethical dimension of CSR comes fourth indicating the weak interest of ethics.Indeed, our sample is not subject to the constraints imposed on listed companies, which should show a certain degree of accountability to stakeholders, especially in terms of transparency of information, to establish trust (Boussourra & Zribi, 2008).
On the other hand, refining our analysis by size (Waddock & Graves, 1997;Stanwick & Stanwick, 1998) and by sector (Waddock & Graves, 1997;Griffin & Mahon, 1997) the importance given to each of the responsibilities tends to slightly vary.
Indeed, an analysis of average trends (Boussourra & Zribi, 2008) by size and sector of activity (factor analyse method) shows that large companies (a net fixed assets of more than 4 million dinars) belonging to the industry sector lack interest in the economic dimension (third place), however they have a strong interest in legal responsibility (first place).
This may be explained by the fact that large and industrial companies (with a strong trade union presence) are controlled and subject to internal and external social pressures.
Philanthropic efforts remain very present in second position.The economic role remains low at the before last position.Business + services companies with a small size follow the same order of responsibilities found above but a slight reverse on the top where the ethical dimension goes into third position.
The order of responsibilities for Tunisian managers are represented as follows: The social component Religion teaches charity, mutual aid and collective support values to community members.Indeed, philanthropy (based on the concept of "Zaket" which is one of the five pillars of Islam) is widespread.It manifests itself in sponsorship programs, donations to various charity initiatives and financing of schools and associations (Bousourra & Zribi, 2008).In this sense, philanthropy is the right thing to do by the companies.

The legal component
The strong commitment of Tunisian companies to comply with legislation and labor code would reflect the role of trade unions and the pressure they put on companies especially since the popular uprising of late 2010.In fact, the powerful trade union center, named UGTT, gained more strength and legitimacy because of the unstable political situation in the country which allowed it to fully play its role as a defender and guarantor of the rights of workers.
On the other hand, compliance with laws relating to workers' rights by Tunisian companies could possibly be explained by the fact that they have an incentive to comply with the legal framework to attract partnerships with foreign multinationals often demanding in terms of CSR.

The ethical component
The ethical concerns do not yet seem present in the Tunisian companies.They come last.Our sample is not subject to the constraints of transparency and corporate governance practices, as is the case for listed companies (Boussourra & Zribi, 2008).

The economic component
If the economic component is a priority for companies in the US, Japan and England (Aupperle et al., 1985;Aupperle, 1982) for Tunisian companies, this dimension is far from being considered a priority.We believe that this latter attitude may be explained by the difficult economic environment resulting from the political events that have shaken the country since the uprising of 2010.The economic slowdown and political instability reduced the attractiveness of Tunisia and discouraged domestic and foreign investors (GIZ-DP 2012).
In sum, the analysis of the Tunisian case confirms the presence of the four dimensions of CSR as theorized by Carroll (1979) and empirically proven by several studies.The results of our study are different from those in the US and European contexts.
The work of Aupperle et al. (1985) and Aupperle (1982) on the American context confirms Caroll's model ( 1979) putting first economic responsibility followed by legal, ethical and philanthropic responsibilities.These results have been supported by Pinkston and Caroll (1996) in a study of seven countries (USA, England, France, Germany, Switzerland, Sweden, and Japan).
In the same vein, the work of Raynaud and Roques ( 2007) on a sample of French companies confirms the four-dimensional nature of CSR, although the order of responsibilities does not reproduce with the order assumed by Carroll (1979).
French companies seem to mind first legal aspects and give the other responsibilities less weight.
In the Tunisian context, the results of relevant studies conducted so far seem to be controversial.Boussourra and Zribi (2008), through a quantitative study of a sample of Tunisian companies, conclude with the primacy of legal responsibility followed by ethical, economic and philanthropic responsibilities.
Furthermore, the qualitative study of Golli and Yahiaoui (2009) to determine managers' perception of the different dimensions of CSR using Carroll's model in Tunisia allowed them to extract two different conceptions of CSR by Tunisian managers.The interviews analysis found that "a first group believes that the rule of law is the very nature of corporate social responsibility and a second group believes that social responsibility primarily focuses on pursuing economic objectives.
The study of Gherib (2011) shows that in Tunisia CSR focuses on compliance and regulation.According to the author, perception of CSR in Tunisia remains minimalist or values economic objectives and the search for respecting law and regulations.The author explains this thesis by the social compromises limiting Tunisian companies due to the popular uprising of late 2010.
In sum, absence of homogeneity in results on CSR in Tunisia lets us assume that this "object of study" remains a vague and ambiguous concept in Tunisia, not yet mature enough in the culture of Tunisian companies (Gherib, 2011).
On the other hand, we believe that implementation of a CSR policy attracted several countries with a variable degree of importance and expansion, depending on the specificity of each country.This diversity in CSR conception could be explained by contextual factors such as culture or institutional variables (Boussourra & Zribi, 2008).
The cultural context is important in determining appropriate CSR priorities and programs.Campbell (2007, p. 950) assumes that corporate social responsibility "may mean different things in different places to different people at different times."

Conclusion
In this study, we sought to understand CSR as perceived by Tunisians managers using a conceptual model developed in the USA and validated in several contexts.After introducing methodological adjustments that consider contextual differences, we conducted a quantitative questionnaire-based study of a sample of Tunisian compagnies.The obtained results were different from those obtained in the US and Europe.
Indeed, our study empirically confirms the presence of the four components of CSR as theorized by Carroll (1979).Our results support therefore the current research emphasizing the four-dimensional nature of CSR (Carroll, 1979;Maignan et al., 1999).However, the Tunisian context seems clearly different from the European-American contexts insofar as in Tunisia reaching economic gains is not the first concern for Tunisian managers.Indeed, the results obtained from factor analyses allowed us to infer that philanthropic responsibility is the most influential dimension of CSR practice in Tunisia, followed by legal responsibility.Next comes successively economic and ethical responsibilities.
Our study aimed at contributing to a better understanding of the concept "CSR" which seems to be complex to define in a given context, like that of a developing country.Using a quantitative approach, rather different from previous studies which used qualitative methods, we obtained a more precise picture of CSR conception by Tunisian companies.
On the one hand, and from a theoretical point of view, our study did not consider in depth the cultural and religious aspects to better explain the concept of CSR (Frederick, 1994).On the other hand, institutional power would be very important to better understand the concept "CSR" (Campbell, 2007).
Methodologically, we found it a limitation the adaptation of a CSR conceptual framework developed for the Anglo-Saxon context, which is completely different from the Tunisian context.On the other hand, we believe it would be useful to retain a larger sample to possibly generalize our results.
It would be useful to conduct further research that takes into account: (a) The parameters of the Tunisian context, as a developing country in this case; (b) The institutional aspects; (c) The presence of foreign multinationals that play a significant role in transferring principles of CSR to domestic companies (GIZ-PDD, 2012), or (d) the Arab-Muslim culture of Tunisia, which would likely influence social responsibility practices (Gherib & alii, 2009).

Figure 1 .
Figure 1.Social responsibility pyramid in Tunisia

Table 1 .
Results of the factor analyses