Corporate social responsibility and financial performance: The case of the Saudi companies

Article history: Received 12 April 2019 Received in revised form 11 July 2019 Accepted 11 July 2019 This research examines the effects of Corporate Social Responsibility (CSR) on financial performance in the Saudi companies' context. In this context measuring, CSR is a necessary condition for knowledge of their own social responsibility and thus to control environmental and social impacts. Assessing the social and environmental performance, the establishment of a steering system for the performance and accountability on these external dimensions imply the existence of metrics to assess the quality of management of the business-related non-financial. The CSR and the financial performance are measured using two accounting variables: Return on assets (ROA) and return on equity (ROE). The financial data are collected from the last ten years (2007-2017). The results show the absence of a relationship between the CSR and the financial performance measured by ROA, whereas there is a positive relationship if the financial performance is measured by the ROE.


Introduction
*Since the early 1990s, performance measurement systems have continued to evolve. This is reflected in numerous contributions that have not ceased to adapt to the organization's strategies with those of the development of the global business environment. Chiapello and Delmond (1994) proposed to add a qualitative representation, that is to say, they propose to incorporate the non-financial information into the performance management systems. Kaplan (1984) assumed the integration of non-financial indicators into business performance measurement systems such as internal process, organizational learning, customer and shareholder satisfaction in the Balanced Scorecard. Eco-control or societal management control consists of adapting the traditional components of management accounting.
Social responsibility is a management theme that has been widely on the American context and not widespread in the European context. The overall idea of this concept covers all interested parties that have relationships with companies, which take into account the different expectations of its stakeholders, be they financial, social, ecological and environmental. In this context, the notion of CSR covers two major aspects: The taking into account of the demands of the so-called stakeholders as formalizing the response to a request social then coming to confuse social responsibility and social receptivity and the integration of practices related to this concept management logic. Further, Pesqueux (2004) added that "Corporate social responsibility (CSR) has become a management theme today. In such a debate that votes to broaden the responsibility of all members of the social body. And with a revival of financially enhancing societal performance, we have noticed a series of studies that attempt to examine the impact of the implementation of new management tools on the management controller missions. In particular, Simons (1991), Smith and Langfield-Smith (2004), and Gond and Igalens (2008) defined the performance as a system which makes it possible to influence the behavior of the individuals to end to achieve the objectives of the company. Financial performance is no longer enough to assess the performance of a company. It was during the 20 th century that performance expanded to take into account the "Social Responsibility" or societal responsibility of the company towards its "stakeholders". The recent emergence of CSR themes is a new challenge for management control and makes it necessary to propose indicators specific to these particular dimensions. Subsequently, the work on CSR took on a new dimension from the 1970s, thanks in particular to the work of Carroll (1979).
In addition, the concept of social responsibility is used as a translation of the term CSR. The term societal is used in preference to social when it refers to the responsibilities of the enterprise towards multiple stakeholders. For some, the concept of CSR "is the latest innovation in the managerial panoply born of the bursting of speculative bubbles and financial scandals that marked the passage of the 20th century to the 21st century". Carroll (1999) defined this notion as the ability of a firm to respond to social pressures. In Wood (1991), receptivity means setting up relationship management that links the firm with the different "Stakeholders".
This research examines the effects of CSR on financial performance in the Saudi companies' context. In this context measuring, CSR is a necessary condition for knowledge of their own social responsibility and thus to control environmental and social impacts. Assessing the social and environmental performance, the establishment of a steering system for the performance and accountability on these external dimensions imply the existence of metrics to assess the quality of management of the business-related non-financial.

Literature review
By the 1990s, performance measurement systems continued to evolve. This is reflected in numerous contributions that have not stopped adapting the organization's strategies with the development of the global business environment. First, Chiapello and Delmond (1994) proposed adding a qualitative representation that is, incorporating non-financial information into performance management systems. The latter authors share the same assumptions as those mentioned by Kaplan (1984), which assumes the integration of non-financial indicators in business performance measurement systems. As a result, management control goes beyond the traditional thresholds of financial approaches to introduce a variety of indicators that meet societal expectations and that of sustainable development. This has given management control the appearance of a tool for managing overall performance.

Evolution of performance measurement systems
The focus of the review of literature evokes two broad categories of measures to approach financial performance. Measurements from Accounting based measures and/or market-based measures. In this case, a CSR-oriented management control system promotes sustainable development and gives companies a dominant competitive position. More concretely, sustainable development can represent a possible balance between the economic, the social and the environmental. The studies conducted by Griffin and Mahon (1997) have tried to allocate the variables of financial performance measure into four categories of variables, three categories relating to the accounting performance and a class of stock market performance. According to empirical studies, accounting measures provide most of the time positive correlations between CSR and financial performance. Similarly, these accounting measures have the advantage of providing a more relevant measure of the company's economic performance and more reliably predicting the potential link between CSR and financial performance. On the other hand, stock market measures represent investors' assessments of the company's ability to generate economic profits. The theoretical foundations of financial performance vis-à-vis CSR, focus between two opposing theories, the first concerns the proponents of the neoclassical theory, rely on the efficiency of markets, and refuse any idea of a CSR other than making a profit for its shareholders. The second is formalized in theories that mobilize the moral responsibility of decisionmakers for future generations and many societal problems. Thus, among the theories that affirm the most or less great convergence between corporate interest and societal interest, we can distinguish three streams:  The current "Business Ethics" or current "Moralist-Ethics" considers that the company must act in a socially responsible manner, which gives the firm status of the moral agent, able to distinguish between good and bad.  The current "Business and society" or current contract-societal believes that there is no barrier between the company and society and that the two are interacting.
According to this theory, the company has relations with the company that is not exclusively market, which emanates a social contract between the company and the company. The "Social Issue Management" stream proposes to provide managers with tools to improve the performance of companies, taking into account the expectations expressed by various stakeholders in society. This current offers a new approach to the environment that is not only economic but also sociopolitical which broadens the field of actors.
Financial performance is no longer enough to appreciate the performance of a company. It was during the 20th century that performance was broadened to take into account the "social responsibility" or social responsibility of the company vis-à-vis its stakeholders. The concept of performance is dematerialized with the concept of sustainable development. The concept of social responsibility is well understood as an essential component of sustainable development. This is of great importance in the various evolutions of the overall performance to reveal its different approaches. Thus, new methods appear, such as social and environmental rating or stakeholder-oriented "Governance" accompanied by "Stakeholder Theory".

CSR and financial performance
The concept of CSR originated in the United States. In fact, the words "Corporate Social Responsibility" come from the European Commission. In the European sense, the term "Social" is translated from English and should be, in French, rather close to the term "Societal" which includes the environmental component. Bowen (1953) defined CSR as "a series of obligations entailing a series of policies, decisions, and policies consistent with the goals and values of society". Carroll (1991) suggested that social responsibility extends beyond the boundaries of the firm to serve the diverse interests of the various stakeholders of the enterprise. This notion makes it possible to highlight the expectations expressed by vis-à-vis the organizations. In Carroll (1991), responsibility has four components: economic, legal, ethical and discretionary. Although Carroll (1991) has expanded the company's responsibility by emphasizing ethical and discretionary considerations, not to mention that the fundamental responsibility of the company is economic. Through this observation, the evolution of the concept of CSR appears as a new concept stemming from the societal receptivity (Corporate Social Responsiveness). In this context, Carroll (1999) defined this notion as the ability of a firm to respond to social pressures. In Wood (1991), receptivity means setting up relationship management that binds the firm with the different "stakeholders". This has given this notion a new, more managerial and more operational vision for social responsibility. In addition, we have noticed in the literature the lack of a clear and specific definition of the company's societal performance. According to Carroll's (1991) research, societal performance was a series of three dimensions: the principles of social responsibility (economic, legal, ethical and discretionary), the philosophies of responses to societal problems. The social responsibility of the company began in the work of American academic Howard Bowen in 1953, notably in his book "Social Responsibilities of the Businessman". In his book Bowen (1953), he asserted that businesses are vital centers of decision and power and that their actions affect the lives of citizens in many fields. Through this, Bowen (1953) concluded that companies are responsible for their actions in a wider sphere than just the economic sphere. Thus, he defined CSR as "an obligation of businessmen to make policies to make decisions and to follow policies that meet the goals and values that are considered desirable in our society". In this context, Davis and Putnam (1960) argued that social responsibility refers to: "Decisions and actions taken by businessmen because, at least partially, considerations that transcend those directly related to technical and economic interests". Indeed, Davis and Putnam (1960) proposed the following definition: "For the purpose of this discussion CSR refers to the firm's considerations and responses to problems beyond the economic, technical and legal conditions of society. It is the obligation of the firm to evaluate these decision-making processes, and the effects of these decisions on the external social system in such a way that it achieves social benefits with the traditional economic gains that the firm seeks. Similarly, Gond and Mullenbach-Servayre (2004) discussed the dimensions of the company's social performance and the distinguished process between the company's behavior that could be called social engagement, social responsibility, and social response. Thus, social responsibility is still an emerging concept with somewhat vague meaning. While Carroll (1999) and Wood (1991) considered societal performance as "An organizational configuration of societal responsibility principles, processes of societal sensitivity, and programs, policies and observable outcomes that relate to the social relationships of society. In Déjean et al. (2004), the social responsibility of the company is a recurring debate in the French context, it is about a complete genealogy of the study of the socially responsible behaviors on the part of the French companies. Reynaud (2008) supported the definition of Carroll (1999), which recognized the dimensions of economic, regulatory, discretionary and ethical responsibility. Bollecker and Mathieu (2008), at this point, made the following point: "Research on management controllers also suffers, we believe, from a thematic shift" and it encourages research that contributes to the development of issues that take into account interactions between the attributes of the management controllers, especially in the context of CSR. In such questioning, we estimate quantifying the societal responsibilities of companies and the societal performance, made today as a profitable phenomenon and which directly governs the overall performance of companies.

Social responsibility and financial performance: Theoretical approaches
The review of the literature has found theorists and practitioners who have not ceased to explain the meanings and boundaries of the concept of CSR. For this purpose, in order to contribute more to the meaning of this unifying concept, we will base ourselves on the main models of CSR. Through a first review of the literature (The basic works of Carroll (1979), the model of Wartick and Cochran (1985), and Wood (1991)), we have distinguished three main models that triggered a great debate on the concept of CSR.
The theoretical approaches to CSR are essentially based on the current contractual and sociological neo-institutionalism.
Those questioned the compatibility of market logic and the objective of maximum profit that underpin the economic rationale of the business and societal concerns such as sustainable development, intergenerational equity, the public interest purposes that are appearing prior foreign or contrary to the entrepreneurial logic. The theoretical basis is between two opposite poles: on one side, neoclassical theories, which rely on market efficiency, reject any idea of a CSR other than making a profit for its shareholders. However, the only approach "moralist-ethical" is not sufficient to illuminate the strategic behavior of firms in the field societal because it does not understand the motivations of corporate behavior. The responsibility of the company depends on leaders who are "agents" in explicit or implicit contractual relationships with several categories of stakeholders: shareholders, community, etc., consequently, they must manage the different expectations of a balanced way, which could affect the sustainability of the company. An implicit social contract provides the framework for the dissemination of information and the consultation of stakeholders. The actors representing stakeholders will then exert a monitoring role in order to avoid misleading communications and ensure that social strategies are not simple clearance officers. In this approach, stakeholders influence the strategic decisions of managers and they must be accountable to them about how they took into account their expectations. In this context, Carroll (1979) distinguished four categories of CSR: The economic responsibility, the legal liability, the Responsibility Ethics, and the Responsibility discretion.

The stakeholder theory
From the 1980s, the theory of stakeholders has gradually established itself as a framework to further specify the group's vis-à-vis what the company should have social responsibilities. In fact, the company is in the middle of a set of relationships with partners who are not only the shareholders but the actors interested in or affected by the activities and decisions of the company. The stakeholder theory is not free from a normative and ethical vision but seeks to integrate economic objectives: it asserts that the cooperation agreements establish trust between the firm and its stakeholders and provides a competitive advantage to the company. A first theoretical approach suggests that the company is more powerful socially; it is more efficient economically and financially. Instead, the company will be more successful economically and socially least it will be.
Finally, beyond these two extreme views, it is possible to consider the assumptions of positive synergy and negative cross the different conceptual foundations. With these assumptions also adds a generic assumption of neutrality of interactions and an assumption of more complex relationships.

CSR and financial performance (FP): Empirical approaches
Clarification of the economic impact of CSR has always been a major concern in the field of study on the relationship between business and society. It is therefore not surprising that empirical work on this issue has been very numerous; in 2007 there were over 160 empirical studies on the subject? This work focused on the nature of interactions between the ability of firms to achieve a high level of CSR and financial performance by studying the interaction between, on the one hand, social performance (or societal) of the company, and, secondly, financial performance.
Numerous publications over the last twenty years have highlighted the link between social responsibility and financial performance of the company. But these studies show conflicting results do not allow to clearly establishing the existence of a positive or negative relationship between social responsibility and financial companies. Margolis and Walsh (2003) (2007) lists 160 on empirical studies (3) on this subject, but also that this research was sometimes biased in the direction of the illumination of a positive relationship. Example of the 122 studies, fifty-one concluded a positive association between social responsibility and financial performance, twenty gets mixed results, and twentyseven concluded the absence of links and seven observe a negative relationship. Through the review of the literature, our hypothesis is: H1: CSR has a positive impact on financial performance. H2: CSR has a negative impact on financial performance. H3: There is no link between CSR and financial performance.

Research methodology
The objective of empirical research is to empirically verify our research hypotheses and the theoretical model proposed. In order to test the validity of our hypotheses, we had collected a sample from the Saudi companies.
As part of our research, we chose the method of regression results, using the method of panel data for each variable of financial performance (ROA, ROE) on the five dimensions of CSR.
The results covered a sample of 300 Saudi companies chosen from different sectors. In what follows we will try to present the regression results, using the method of panel data for each variable of financial performance (ROA, ROE) on the five dimensions of CSR.

Measure of CSR
We present in what follows the procedure that we used to verify the reliability and validity of the measurement scale of CSR. The analysis was performed using the software for data analysis SPSS13.0. Different steps must be followed to ensure that the analysis is properly conducted.
We first present the scale factor analysis. In terms of "characteristics", we selected the "initial structure" in the "Statistical Area" and "coefficients", "significance levels" and KMO index and Bartlett (1980) test in the "correlation matrix». The KMO test used to quantify the degree of correlation between variables and the appropriateness of factor analysis. This indicator is between 0 and 1.
This test must be significant that the data are factorizable (p<0.05). As a result, we used principal component analysis as the extraction method with orthogonal rotation (Varimax). Through the results emerged, each variable must be correlated to a single axis. This happens when the difference between the saturation on the principal axis and the saturation of any other axis is greater than 0.3. Otherwise, it means that the variable is correlated with both axes, we must eliminate it and re-factor analysis. After determining the number of items presented in each scale, we calculate the Cronbach alpha. The reliability coefficient serves to estimate the degree of internal consistency of the isolated structure. The value of alpha which is generally considered acceptable in basic research is 0.7, but this value can be lowered to 0.6 in exploratory research (Hair et al., 1995). Measuring CSR consists of four dimensions (economic, legal, ethical and discretionary) developed by Carroll (1979) and psychometrically validated by Maignan et al. (1999). Our extension is the addition of the environmental dimension whose items are inspired by the Global Compact in 1999.
In what follows, we analyze the results of the factor analysis of each dimension of our variable CSR. Subsequently, we present the Cronbach's alpha which is used to verify the reliability of items.
On the first economic dimension, it includes 4 items. The KMO index (0.693), the significance of Bartlett (p= 0.002), shows that the original data matrix is factorized. Factor analysis shows that four items are related to the first factor (maximize profit), this factor is 52.285% of the total variance and a value greater than 1 is to 2.091. The correlations of items with factor chosen are all above 0.5 (with the exception of item 4 that a correlation equal to 0.489). The Cronbach's alpha releases worth 0.639, so the scale used to measure the economic dimension is reliable. The second dimension which is legal shows a KMO index of 0.610, the significance of Bartlett is equal to 0.002. Two successive factor analysis was performed where we removed the first item because it has the lowest correlation (0.437) compared to other items. The Cronbach's alpha is equal to 0.509 we can admit it because our sample is small. The third ethical dimension consists of five items. The KMO index (0.681) and significance of Bartlett is equal to 0,002 are satisfactory. Factor analysis shows that five items are related to the first factor, which represents 46.571% of the total variance, and it is above 1 (2.329). The Cronbach's alpha is 0.7 which shows the reliability of the scale measuring this dimension.
The fourth dimension is discretionary, it consists of 4 items. Two iterations were performed where it was eliminated the third item. KMO index is 0.578 and the significance of Bartlett shows a value (p= 0.000), releasing an acceptable result. Cronbach's alpha showed a value of 0.793, which means that this scale is reliable. Finally, the environmental dimension is measured by 4 items. The KMO index (0.762) and significance of Bartlett (p= 0.000), showing that the original data matrix is factorized. The two-factor analysis was conducted in which we removed the first item. Regarding the reliability of this measurement scale item selected show good internal consistency. Indeed, Cronbach's alpha has a value of 0.854. The first factor obtained is the environmental dimension. The value of Cronbach's alpha was 0.854. This factor includes items relating to undertaking a great responsibility to the environment; promote the development and dissemination of technologies that respect the environment and finally the provision of a program to reduce energy consumption and water.
This result seems logical since, after the ecological disasters suffered by our planet; there is great pressure from various stakeholders to ensure that businesses take into account the natural environment in their business processes. Tunisian companies are aware of the need to protect the environment where they operate. The second factor is the dimension discretion for Cronbach's alpha of 0.793, indicating the potential role of enterprises in Tunisian society. So there is no doubt that the company as an organization located in the heart of social, expands its environment to take account some aspects of social, human and other. The third factor, it is the ethical dimension to a Cronbach's alpha of 0.7 this position may be explained by the fact that the Tunisian companies want to appear first as trustworthy to its stakeholders, therefore created a climate of trust shared, especially after the financial crises that affected everyone.
Secondly, provide accurate and complete information, this may pose a strategic asset to attract more investors. In fourth place comes the economic dimension, which leads us to relativize the neoclassical theory which states that the only responsibility of a company is to maximize profit. The last dimension concerns the legal dimension. This can be explained by the fact that the law is a requirement that all companies must comply, so the law does not present liability that the company must ensure voluntarily such as ethics or the protection of the environment. In the case of a dismissal or recruitment company is required to respect the labor code. All these results support the idea of social responsibility to the company that expands the environment of the firms understood in its dimensions, not only economic and legal but also social, human and ecological. Thus we can conclude that CSR consists of five dimensions mentioned above.

Impact of CSR on financial performance measured by ROA
Before presenting the results we first define the regression function as follows: The result of the study of the impact of social responsibility on financial performance measured by ROA reveals no relationship between these two variables (see table below). Indeed, the low value of R² (3%) and adjusted R² Absolute (5%) show a linear fit small, so a small explanation of financial performance based on social responsibility. The overall significance of the model is examined by Fischer's test that has a value greater than 0.005, which proves that the model is not globally significant. A review of estimates inherent in each of the five dimensions of CSR shows no significance, except that the economic dimension has a coefficient (α) negative and statistically significant. The analysis, therefore, shows a neutral effect of CSR on PF. When we moderate the relationship by size we observed that R ² increases and becomes 10%, which indicates that the variable size slightly improves the estimate. We noted also that F Fischer displays a value of 3.9% <5%, which shows that the model is globally significant. Despite this improvement, the relationship remains neutral. With the introduction of variable risk (F= 0.039) and industry (F= 0.46) we noticed that the model is generally not significant, which means that the risk variables and sector have no effect on the relationship between CSR and financial performance. Therefore, we conclude that with the introduction of control variables (size, risk, and industry) the relationship between social responsibility and financial performance remains neutral. In this respect and according to the study by Ross (1986), this indicated that there are many variables that can intervene and moderate the relationship between social responsibility and financial performance. The author argues further that the possible exception that social responsibility moderates the financial performance, there may be random, and the problem of measuring social responsibility can mask the potential link between the two built. In this context, other authors have argued that the relationship between social responsibility and financial performance was so complex and indirect that it is not possible to postulate the existence of a stable relationship between the two variables (Gond and Igalens, 2008).

Impact of CSR on financial performance measured by ROE
The regression function is defined as follows: Contrary to ROA, the social responsibility of the company explains better financial performance which is measured in this case by the ROE. Indeed, R² is around 38%, that is to say, the change in ROE is explained at a rate of 38% by social actions. The variable components of social responsibility have no effect on financial performance measured by ROE at 95% of confidence. But at the risk of 10%, certain variables become relevant in explaining the financial performance like the discretionary dimension.
The probability of Fisher displays a value (p= 0.000) indicating that the model is globally significant. With moderation by the control variables, we found no improvement. In fact, decreases R² and has a value of 27% with variable size, 33% with the risk variable, and 27% with the variable area. So we can conclude the lack of moderating effect of control variables on this relationship. Similarly, with the introduction of variable size and risk, we have noted that the discretionary dimension has a positive effect on financial performance measured by ROE. This can be explained by the fact that large firms and less risky Debt undertake more in philanthropy to gain in terms of image. It is also appropriate to conclude that the environmental dimension in both cases of the measurement of financial performance produced a negative effect. This can be explained by the fact that investment in the environment is very expensive, which negatively affects financial performance. In conclusion, we can say that the social responsibility of the company has a positive impact on financial performance measured by ROE. Indeed, the inclusion of stakeholders will result in improved economic performance, that is to say, the most successful companies on non-financial criteria, benefit from a more favorable behavior of the stakeholders than their competitors and will, therefore, have a higher financial performance.

Conclusion
This research examines the effects of CSR on financial performance in the Saudi companies' context. In this context measuring, CSR is a necessary condition for knowledge of their own social responsibility and thus to control environmental and social impacts. Assessing the social and environmental performance, the establishment of a steering system for the performance and accountability on these external dimensions imply the existence of metrics to assess the quality of management of the business-related non-financial. In fact, the existence of these metrics is also of particular importance to other stakeholders that ethical investors who require such information to select the best-performing companies on the main criteria the quality of resource management Human and respect for human rights.
This leads companies to establish a legal and socio-technical infrastructure to make measurable CSR stakeholders. In theoretical terms, the extent of CSR faces similar problems to those identified to define the concept of CSR: The multiplicity of approaches and dimensions of this complex concept, difficult to report objectively its components more subjective often linked to an assessment based on criteria related to ethics or a social context. In this work, we tried to study the impact of CSR on financial performance on a sample of Saudi companies. We had specified the factors of social responsibility we then studied the impact of that on the financial performance indicators measured by ROA and ROE. The results show the absence of a link between social responsibility and financial performance measured by ROA, while the relationship is positive when the latter is measured by ROE. We can also conclude that the control variables (size, risk, and sector) do not moderate the relationship between social responsibility and financial performance. Hence, we can conclude that social responsibility has a positive impact on financial performance if the latter is measured by the ROE, but we note that there is no relationship between the two built if financial performance is measured by ROA. We can also conclude that the control variables (size, risk, and sector) do not moderate the relationship between social responsibility and financial performance.