Corporate social responsibility versus business ethics: analysis of employee-related policies

Purpose – This study aims to analyze the relationship between companies’ business ethics (BE) and corporate social responsibility (CSR), with particular reference to policies toward employees, with the aim of understanding if and how the two concepts are linked and to foster a better management of the company-employeerelationship throughBEand CSR policies. Design/methodology/approach – Through a content analysis, the authors study three issues related to employees disclosed inCode ofEthics(CE)andCSR reportofa sample ofItaliancompanies. Next, using a multivariate regression model, the authors examine the relation between the BE and CSR initiatives, related toemployees. Findings – The findings show that CE and CSR initiatives are negatively related. They are distinct concepts, but since the authors find that they are connected, they must also be considered in terms of their mutual dependence. To standardize practices toward employees in a code may induce the need to establish additionalcorporate social responsibility initiativesthat elicitlegitimatestakeholdersatisfaction. Research limitations/implications – The analysis focuses on employees, whereas several other CSR aspects that can be explored. Furthermore, additional investigation (through questionnaires or interviews) could deepen this analysis. Furthermore, it might be interesting to consider different countries ormorevariables,suchasculturaldifferencesordifferent regulations. Practical implications – The results of this research reveal that BE and CSR initiatives require precise and personalized observations to be properly understood; however, as they are linked, they must also be studied in their mutual interdependencies; this can be very useful to define governance bodies and organizationalprocedures devotedtoBEandCSR issues. Social implications – This research providesa toolforevaluatingandmonitoringCSR and BEprinciples andcanbeadaptedtomany business contexts and refertodifferent stakeholders. Originality/value – The existing literature on BE and CSR presents opportunities for further study, as theseconceptsareoftenstudiedwithoutinsightsinto their mutualimpacts.


Introduction
Researchers have been studying the function and importance of business in society for decades (Carroll, 1991;Schwartz and Carroll, 2008), often considering both a company's responsibilities toward all its stakeholders (CSRcorporate social responsibility) (Freeman et al., 2010) and the ethical aspects of companies' activities (BEbusiness ethics) (Crane et al., 2010). Many authors frequently use the terms BE and CSR to represent the same aspect of business activity, even though the two concepts are different (Weller, 2020). Weller (2020), starting from Fischer (2004) and integrating Schwartz and Carroll (2008), suggests three possible interpretations of the relationship between CSR and BE: CSR and BE are equivalent, CSR and BE are each part of the other and CSR and BE are distinct but related.
We consider a specific BE and CSR perspective relate to companies' employees. The BE and CSR literature, in fact, frequently overlooks employees even though they are key stakeholders (Hansen et al., 2011;Pedersen, 2011). Few studies analyze BE and CSR from the employee perspective (O'Dwyer and Madden, 2006;Mazza and Furlotti, 2020). We contribute to this stream of literature by conducting an analysis of employee-related policies.
Considering previous literature, we want to verify the existence of a relationship between BE and CSR with reference to the companies' policies toward the stakeholder employee.
To analyze this relationship, we study the disclosure about the issues voluntarily produced by companies, in particular, the Code of Ethics (CE), expression of BE and the social or sustainability report, as expression of CSR.
So, we analyze the relation between BE and CSR employee-related policies with an empirical approach.
Using an Italian sample of listed firms that voluntarily disclose employee-related policies in documents containing social, sustainability or integrated reports and include such policies in their codes of conduct or codes of ethics, we perform a manual content analysis. We use the methodology used in the literature to define keywords and scales and search the examined content for employee-related policies related to the three most important and frequently disclosed issues involving employees: employee health and safety, learning and development opportunities and equal employment opportunities. Next, we perform a multivariate regression analysis to test the relation between BE and CSR.
The primary contribution of our research is the presentation of univariate results of content analysis involving various documents of voluntary disclosure. We aim to provide the first explorative indications of the type of disclosure differentiated by such documents and by the concept of BE and CSR.
We find that in the examined voluntary CEs and CSR reports, employee health and safety and employment opportunities are always included, while learning and development opportunities may not be considered. The disclosures on learning and development opportunities contrast with those on employee health and safety. There is more differentiation in the CEs than in the CSR reports on learning and development opportunities, while there is more differentiation in the CSR reports on employee health and safety. A measurable result is always included in the voluntary CSR disclosures on learning and development opportunities, while there are some firms that do not disclose data on employee health and safety. However, in the CSR reports, detailed disclosures on learning and development opportunities are not frequently given, but those on employee health and safety are often provided.
The secondary contribution of our research is the presentation of multivariate results of a regression analysis on the relation between CSR and BE. We study the statistical significance of this relation and its direction. We aim to contribute to the debate on the interpretations of the relationship between CSR and BE by portraying these factors as equivalent, each part of the other or distinct but related (Weller, 2020).
We find that these factors are distinct and negatively related. If companies have good BE policies regarding health and safety and learning and development opportunities, there is little need to provide additional disclosures in voluntary CSR documents about specific strategic initiatives.
Corporate social responsibility and business ethics: theoretical framework and hypothesis development CSR refers to the responsibility that companies assume for all their stakeholders, who may be investors or any other individuals or groups that have stakes in companies (Balluchi and Furlotti, 2013;Gö ssling and Vocht, 2007). According to the 2001 European Commission Green Paper, promoting a European framework for CSR can be defined as "a concept whereby companies integrate social and environmental concerns in their business operations and in their interaction with their stakeholders on a voluntary basis." In most of the existing CSR definitions, it is possible to find, in addition to a mention of the relevance of stakeholders, an implicit reference to BE (Schmidheiny et al., 1997). In this respect, CSR can be viewed as the right thing to do (Allen and Peloza, 2015).
Regarding BE, Lewis (1985) analyses the most often used definition of the concept, particularly in the context of business research, finally concluding that "business ethics is rules, standards, codes, or principles which provide guidelines for morally right behavior and truthfulness in specific situations." He underlines the main aspects of this concept: rules, standards and codes or principles, which represent ethical rules for preventing unethical behavior; morally right behavior, which refers to individual actions that are in accordance with justice, laws or other standards; truthfulness, meaning indications and actions that conform with facts; and specific situations, which refers to the situations of individual ethical dilemmas necessitating ethical choices (Lewis, 1985, pp. 381-382). BE, therefore, is about much more than values or principles; it includes values, but it is concerned with moral rules, guidelines and codes about right and wrong conduct (Harrison et al., 2019). The establishment and formalization of a written document containing a set of rules that guide companies' behaviors according to precise values and ethical principles are one of the most common ways to integrate ethics into the management practices of a company. In this respect, there is a growing tendency among companies to develop formal BE documents that also aim to guide and shape corporate culture. In the academic and professional worlds, these documents are generally designated as codes of ethics or codes of conduct and are documents that contain principles or rules that guide the ethical conduct of companies.
Focusing, in particular, on CSR/BE debate, Rusconi (2019) proposes an ethical firm system theory, starting from a joint and synergistic consideration of stakeholder management theory and firm system theory. In particular, the author proposes that stakeholder theory should be integrated into an ethical vision of firm theory, indicating that ethical duties are strictly linked to the management of a company as a whole. In his study, Rusconi (2019) focuses on ethical firm system theory to clarify or otherwise analyze or approach, the three central ethical points of stakeholder management theory: the dichotomy between economic performance and ethical behavior, stakeholder engagement and ethical responsibility and competition in a cooperative context. Schwartz and Carroll (2008) study the concepts of CSR and BE, starting from their definitions and theoretical frameworks and calling for the establishment of a new, shared paradigm to explain the role of business in society and that of ethics in business. In this respect, they consider several theories, including moral agency theory (Goodpaster and Matthews, 1982), social contract theory (Donaldson, 1982), social power theory (Davis, 1975), interpenetration theory (Preston and Post, 1975), stakeholder theory (Carroll, 1991;Freeman, 1984), property-based theory (Hoffman and Fisher, 1990;Klonoski, 1986), utilitarian theory (Den Uyl, 1984) and religious theory (Pava, 1996). Another important theory in the field of CSR is institutional theory. According to this theory, external institutional pressures can explain how companies' decisions are formed; individuals, in fact, generally seek approval, allowing themselves to be influenced by external expectations and pressures stemming from society and seeking endorsement through compliance with societal traditions and expectations (Verbeke and Tung, 2013). Since companies are social constructions created and managed by individuals, institutional theory can be applied to businesses that are under three main kinds of pressure: cognitive, regulative and normative (Scott, 1995). In particular, cognitive pressure usually stems from companies' need to conform to the structures of their environments (Scott, 1995).
In this context, an important area of study concerns the relations existing between CSR and BE, which are often studied through their respective disclosure tools: the CEs and the sustainability (or social/environmental) reports. In particular, CE is a tool in which the company makes explicit the foundations of its ethics and defines the rules of conduct that guide its operations, while social, environmental and sustainability reports are reporting documents in which companies describe what they have done in the field of CSR.
Many research studies have investigated the existing relationship and mutual connections between CSR and BE. Several authors, frequently use the two terms to represent the same aspect of business activity (Fassin et al., 2011;Ferrell, 2004;Pereira et al., 2020), even though the two concepts are different (Weller, 2020). Harrison et al. (2019) explain that although some authors suggest that there is a relationship between CSR and BE (Epstein, 1987), others, even though they recognize the difference between ethics and CSR (De Bakker et al., 2005), consider these different elements under the single concept of CSR (Choi and La, 2013) or characterize CSR as a dimension of BE (De George, 1987). Schwartz and Carroll (2008) explain the existing confusion about these two concepts, providing several examples in which one of the concepts incorporates the other and highlighting studies in which these two terms are used to denote the same meaning. Additionally, the concept of CSR is often included in the general definition of ethics (Joyner and Payne, 2002). Schwartz and Carroll (2008) try to address this misunderstanding, proposing that the concepts of value, balance and accountability should be used as a foundation for future debate and theoretical improvement about the role of business in society. Mason and Simmons (2013) affirm the importance of distinguishing between CSR and BE. They review CSR literature to differentiate CSR and ethical business practices, theorizing BE as an internal expression of CSR; BE is the management's answer to a company's idea of CSR.
Weller (2020), starting from the research of Fischer (2004) and integrating that of Schwartz and Carroll (2008), suggests three possible interpretations of the relationship between CSR and BE: 1. CSR and BE are equivalent because they can be conceptually considered as the same item applied to different contexts: BE is applied to people and employee conduct, while CSR is applied to organizations and business conduct toward outside entities (Davidson and Griffin, 2000).
2. CSR or BE is part of the other, that state that BE is considered a part of CSR or, vice versa, that CSR is an aspect of BE. In this respect, for example, the CSR pyramid by Carroll (1991) encompasses economic, legal, ethical and philanthropic responsibilities (BE is a part of CSR); in turn, Goodpaster (1991) explains that companies' responsibility toward all their stakeholders is a fundamental principle of BE (CSR is a part of BE).
3. CSR and BE are distinct but relate because not only of the different purposes for and reasons underlying BE and CSR but also for an awareness of their mutual influences and importance to each other.
This paper contributes to the BE and CSR literature by exploring a particular aspect of the relationship between these two concepts. The research analyses, in particular, the relationship between BE, which is assessed according to the definitions of the ethical policies in a code (a CE), and CSR, which is measured by the quality of the information provided about CE policies related to a specific stakeholder: employees. This research considers employees because they are key company stakeholders (Pedersen, 2011) but are frequently overlooked in the CSR and BE literature (Hansen et al., 2011). Few studies analyze BE and CSR from an employee perspective (O'Dwyer and Madden, 2006). Simmons (2008) analyses the role of CSR in the context of employee governance; Romi et al. (2018) focus on employee in the B corporations whose owners voluntarily commit to conduct business in a socially responsible manner; Snell et al. (2010) (Manley, 1991), humane working conditions (Williams and Murphy, 1990) and company values (Weaver, 1993). Some authors focus, in particular, on CE characteristics. Ahmad et al. (2019) study the link between CEs and job and work engagement. Houghton et al. (2009) examine employee behavior and investigate the connection between volunteerism and compliance with the CE of a company (Weller, 2020).
Starting from these premises and in the path of this literature, we consider the relation between CSR and BE and study the statistical significance and direction of this relation. In particular, we consider the CEs, the document that summarizes the ethical approach of companies in business activity, and the CSR report, the disclosure document by which a company describes its CSR strategies and actions. According to previous literature (Mason and Simmons, 2013), we consider BE as an expression of CSR, so CE can be considered a management's answer to a company's idea of CSR.
Deriving from the abovementioned literature and focusing on the stakeholder employee, we, define our hypothesis: H1. CE policies about employee have a significant relationship with companies' CSR strategies about employee.

Multivariate regression models
To test our hypotheses, we use ordered probit regressions with robust standard errors: As previous literature suggests (Mason and Simmons, 2013), we consider the BE dependent on CSR because the ethical principle and rules are an expression of a companies' precise vision of corporate responsibility in the context in which they operate (environmental and social as well as economic).
We used the following control variables: Size: the natural logarithm of total assets at the end of each fiscal year; Loss: 1 if net income < 0 and 0 otherwise; ROE: ratio of net income to equity; Leverage: long-term debts/total assets; Sales growth: (revenue t À revenue tÀ1 ) scaled by revenue tÀ1 for fiscal year t.
CE procedures related to employees encompass the following topics: health and safety, learning and development opportunities and equal employment opportunities.
These topics are identified starting from the 10 categories defined by Spiller (2000) for employees. We select those most significant in the Italian context. For the selection, we exclude the categories which in Italy are regulated by law and, therefore, little dealt with in the CE, such as remuneration, communication and job security. We also reject those topics which suffer from high subjectivity in characterization and for which standardization by keywords is ineffective, such as fulfilling work, competent leadership, community spirit and social mission integration. The selected categories, therefore, concern learning and development opportunities, a healthy and safe work environment and equal employment opportunities. These categories are also found to be the most significant in terms of ethical codes and employees in previous research studies (Mazza and Furlotti, 2020).

Sample selection, data collection and content analysis
The sample selection process (Table 1) starts with all the Italian companies listed on the Milan Stock Exchange that are available on Compustat Global. We exclude the financial sector from our analysis because of its dissimilar nature and because of the specific Italian normative and standard of CSR disclosure, compulsory for financial firms.
We download the CSR reports of the firms in the CE sample in any format (social, sustainability or integrated reports). We find 55 firms with CSR reports. This is coherent with the results corresponding to the following questions in the ASSET4 database: CGVSO05V, corporate governance; indicator value, value vision and strategy/transparency -"Does the company publish a separate CSR/H&S/Sustainability report or publish a section in its annual report on CSR/H&S/Sustainability?" and CGVSDP026, corporate governance; datapoint, CSR sustainability reporting -"Does the company publish a separate CSR/H&S/ Sustainability report or publish a section in its annual report on CSR/H&S/Sustainability?". ASSET4 shows that 26 nonfinancial listed Italian firms have CSR reports. ASSET4 covers a portion of the Italian population of firms.
The sample is unbalanced, with a total of 129 firm-year observations; the majority of these observations come from the most recent year, namely, 2016 ( Table 2). Most of the firms (Table 2) prepare sustainability reports (115 observations). Six firms provide social reports.
Only eight integrated reports are found, which is consistent with previous Italian literature on this subject [nine were found in Paolucci and Cerioni (2017); five were found in Camodeca and Almici (2017)]. Most of the firms (Table 3) are in the transportation, communications, electric, gas and sanitary services industries, and these are followed by firms in the manufacturing industry.
The analyzed documents refer to the period 2014-2016. This time period occurs before the introduction in Italy of Legislative Decree 254/2016, which establishes regulatory  requirements about nonfinancial disclosure for some categories of companies, in particular regarding employees and environment. The choice of this period, therefore, allows to analyze employee information without influences due to the introduction of regulatory requirements.
The examined period considers three years because in November 2012 in Italy, the so-called "legality rating" was introduced (Law Decree 1/2012), which allows companies to be rewarded for demonstrating compliance with increasingly high standards of legality, for example, in the case of tenders with public administrations or when accessing credit. In this contest, even in the absence of specific obligations, ethical and CSR choices represent an element capable of increasing the score assigned to companies. So companies are encouraged to design, carry out (and report) ethical and CSR issues in their strategies. The rating is updated every two years and we, therefore, considered a three-year period that could be characterized by greater variability in the CSR and CE choices of companies, though the decree does not recommend specific ethics or CSR rule or strategies, just underlining their importance.
We download all the CEs of the examined firms, and we hand collect data on their CE policies using keywords. Appendix 1 shows the sample scoring criteria used to analyze the CEs, including a list of the Italian keywords used, their English translations and an interpretation of the level name. We use a content analysis based on keywords as prior literature (Campopiano and De Massis, 2015;Liao et al., 2018).
The procedures on employee health and safety are coded according to a scale ranging from 0 to 4 that categorizes the keywords into groups related to safety; related to safety and health; related to health, safety and the environment; and related to accident care. The procedures on learning (training) and development (career) opportunities are coded according to a scale ranging from 0 to 4 that categorizes the keywords into groups related to training; related to training and opportunities; related to training, opportunities and professional growth; and related to training types. The procedures on equal employment opportunities are coded according to a scale ranging from 0 to 4 that categorizes the keywords into groups related to equal opportunities in general, related to equal opportunities within specific categories (the most frequent is handicap care) and related to race/religion safeguards and sex nondiscrimination.
CSR disclosures on procedures and performance concern the following topics: number of employee injuries and the related procedures (employee health and safety), number of training hours and career evaluations (learning and development opportunities) and the corresponding procedures, number of people in minority groups (equal employment opportunities) and the associated procedures. We download the firms' CSR reports for each year. We hand collect data based on an ordered variable schema. Appendix 2 shows the sample scoring criteria from Sethi et al. (2017) with an explanation of the level order.
The disclosures on procedures and performance regarding the number of employee injuries in each firm are coded according to a scale ranging from 0 to 4 based on the criteria j SOCIAL RESPONSIBILITY JOURNAL j of disclosure in Sethi et al. (2017) (zero, minimum, measurable, comparison or exceptional disclosure) related to safety injuries. The same type of coding is used for the other CSR disclosures by changing the topic of disclosure. For learning and development, we investigate assessments of competences and training. These disclosures are related to an assessment of the links between competences and responsibilities, employee competence evaluations and training days. Regarding equal employment opportunities, we investigate the number of people belonging to disclosed minority groups, for example, by career level, country and compensation.
Univariate results of the content analysis Table 4 shows the descriptive statistics of the characteristics of the sample firms. The companies with voluntarily disclosed CEs and CSR reports are, on average, profitable (mean return on assets of 5.7%; only 14.7% of the sample has negative earnings), financially stable (mean leverage of 22.9%), slowly growing (mean sales growth of 2.5%) and of a similar size (low standard deviation and mean similar to median). Table 5 shows the univariate descriptive statistics of the content analysis of the CEs and CSR reports and Table 6 shows their frequency distribution based on a Likert scale. The   levels of disclosure regarding employee health and safety in the CSR reports are heterogeneous; in the sample, all the levels from 1 to 4 (minimum and maximum) are represented. Measurable results are not always included in these voluntary CSR disclosures. On the other hand, in the CEs, the disclosure levels are greater than Level 1. The keywords health and safety are never reported individually (0 observations in Level 1 and 73 observations in Level 2), and the environment is often (50 observations) considered (Level 3). However, the average disclosure level score is higher in the CSR reports (3.682) than it is in the CEs (2.473) because Level 4 disclosures (97 observations) are more common in the CSR reports. In the CSR reports, the disclosures in this category are often very detailed, including measurable results, comparisons and other details.
The category of learning and development opportunities is the only category including Level 0 disclosures (minimum). In the voluntary CE and CSR reports, employee health and safety and employment opportunities are always included, while learning and development opportunities may not be considered. The maximum and the mean values of the disclosure level of this category are higher in the CSR reports (Level 4) than in the CEs (Level 3). The CSR_ Learning and development opportunities variable exhibits the highest standard deviation. The voluntary disclosures corresponding to this category are highly differentiated among the listed nonfinancial firms. Regarding CE_Learning and development opportunities, Level 1 disclosures are the most common with a single keyword related to continuous education/ training without policies related to potential development or professional growth.
The disclosures on learning and development opportunities contrast with the disclosures on employee health and safety. There is more variation in these disclosures in the CEs (minimum 1 and maximum 4) than in the CSR reports (minimum 2 and maximum 4). A measurable result is always included in the voluntary CSR disclosures. However, detailed Level 4 disclosures are not common in the CSR reports (only 1 observation). Regarding CE_Equal employment opportunities, the highest frequency corresponds to Level 2 and the disclosed keywords are related to diversity in general and diversity in reference to disabled people. Policies related to other groups based on ethnicities, races, religions, origins, ideologies, beliefs and gender are less common. Table 7 shows a correlation matrix for this analysis. The low correlation among the control variables and the CSR variables shows the absence of multicollinearity problems. An interesting univariate correlation between the size and type of the CSR reports is only present among the CE disclosures related to employee health and safety. The other disclosures are not significantly correlated to firm size or other firm characteristics.
Multivariate results of the regression model Table 8 shows the results of the regressions. Regarding the control variables, the univariate correlation between size and type of CSR report is confirmed in the multivariate regression; this is the case for the employee health and safety category in relation to size and for all categories in relation to type of report. Moreover, leverage is related to CE_Employee health and safety and CE_Learning and development opportunities, and profitability is related to CE_Equal employment opportunities.
The results related to the hypothesis show that the examined CE policies on health and safety are negatively and significantly related to the CSR disclosures on the number of employee injuries in a firm and the related procedures (marginal effect À0.596; p-value 0.052). The CE policies on training and career opportunities are negatively and significantly related to the CSR disclosures on the number of training hours and career evaluations in a firm and the related procedures (marginal effect À0.385; p-value 0.092). The CE policies on equal opportunities are not significantly related to the CSR disclosures on equal opportunities and the related procedures and metrics (marginal effect 0.025; p-value 0.892). The negative sign of this relation can be interpreted as an indicator that if companies have better policies in their CEs about health and safety and learning and development opportunities, there is little need to provide additional disclosures in their voluntary CSR documents about specific strategic initiatives. Of the several theories that can be used to approach this issue, ethical firm system theory can be used to analyze our results (Rusconi, 2019). Stakeholder theory is encompassed in an ethical vision of firm theory that considers the ethical duties defined in CE documents strictly linked to the management of companies as a whole.
We contribute to the few studies that analyze BE and CSR from an employee perspective (O'Dwyer and Madden, 2006). For example, Snell et al. (2010) find that clear ethical rules can simplify discussions and facilitate interactive justice among employees. We add to this that clear ethical rules on health and safety and on learning and development opportunities simplify the work related to this issue and do not require additional initiatives on the subject to achieve the ethical vision of the firm.
Of the three possible interpretations of the relationship between CSR and BE given by Weller (2020), we provide evidence supporting the proposal that CSR and BE are distinct but related. They have different purposes, have mutual influences and are important to each other (Schwartz and Carroll, 2008).
Measuring CSR with specific procedures and metrics (regarding employee injuries, training hours and career evaluations and people in minority groups), we view CSR as a concept whereby companies can integrate social and environmental concerns into their business operations and establish a continuing commitment to behave ethically. Our focus is on actual operations performed. Measuring BE based on policies in the examined companies' codes, we view BE as rules, standards, codes, principles and guidelines (Harrison et al., 2019;Lewis, 1985). Finding a negative relation between these two concepts, we support the interpretation that CSR and BE are distinct concepts.

Theoretical implications
The existing literature on CSR and BE presents opportunities for further study, as these concepts are often analyzed imprecisely or without effective considerations of their mutual influences. Several authors (Harrison et al., 2019) have pointed out that there is ambiguity in the current definitions of CSR and BE, and these concepts are often measured and analyzed as a single element. To eliminate or help to reduce this ambiguity, we consider the existing relationships between BE (CE) policies and CSR initiatives, focusing, in particular, on a key company stakeholder: employees. In this respect, this study extends the theoretical contributions of the previous research on CSR and BE.
Our findings provide evidence showing that CSR and BE should be viewed as distinct but mutually related concepts. In particular, the data show that CE policies on health and safety are negatively and significantly related to CSR disclosures on the number of employee injuries in a firm and related procedures; CE policies on training and career opportunities are negatively and significantly related to CSR disclosures on the number of training hours and career evaluations conducted in a firm and related procedures; and CE policies on equal opportunities are not significantly related to CSR disclosures on equal opportunities or related assessments and procedures. In this respect, it is possible to suppose that if there are precise policies that are properly formalized and described in a company's CE on health and safety, training and career opportunities, there will be fewer initiatives reported in that company's CSR report; however, there is a greater need to undertake such initiatives and report them in a CSR report if they are not included in a CEs.
The conclusions of our work, although referring to a specific stakeholder, suggest the need to consider the two instruments in a synergic perspective and, in particular, to carefully consider what to describe in each one, considering the implications that there will be in the reporting of the other. Surely, this requires reflection on the consistency of BE statements and CSR actions that can be more effectively verified in the presence of both instruments. This is even more significant considering the synergy that should exist between the bodies dealing with BE and CSR, that certainly need to find spaces for comparison and alignment. Such investigations may represent an interesting space for further research developments.

Managerial implications
From a managerial perspective, companies are spending increasing amounts of resources to develop CSR initiatives and to properly consider and incorporate BE in their corporate strategies and corporate governance.
The results of this research reveal that the two examined concepts, being independent, require precise and individualized observations; however, since they are connected, they must also be considered in terms of their mutual interdependence. Since the methods used for developing CE documents and CSR initiatives are independent but connected, companies must consider BE and CSR strategies separately, but they must take into account also their mutual influences. For this reason, the choice to limit or standardize CE practices toward employees may induce the need to establish additional related CSR initiatives to provide legitimate stakeholder satisfaction.
These results can be used by managers to better understand which areas of CSR and BE to invest in and the possible consequences of these investment choices. In this sense, behavior that is perceived to be unethical (e.g. a missing or ineffective CE) can undermine the value of a company, and engagement in CSR initiatives can help companies develop a positive reputation that leads to subsequent business returns.
From an organizational perspective, the research findings can also help companies in setting up and establishing appropriate CSR and BE bodies or procedures to take care of their communication aspects.
This research provides an assessment tool for monitoring CSR and BE principles and can be adapted to many contexts of company activity. Understanding the examined concepts independently of each other will help companies assess their current positions and improve over time.

Limitations and future research
This research has several limitations and insights for future studies. First, this analysis considers specific CSR and BE variables related to employees, selected according to the specific features of the Italian regulatory and business context. The existing literature (Spiller, 2000) identifies other aspects that can be explored, paving the way for further investigations of CSR and BE variables. This direction of further investigation could, moreover, benefit from a broadening of the sample of companies examined, for example, by taking into consideration more countries or more variables, such as cultural differences or different regulations.
A second aspect concerns regulatory interventions on CSR of the past years that could influence companies' choices in this area. Further studies in the coming years, especially regarding the main actions or directions of change adopted by companies, could also be interesting for assessing the effectiveness of regulatory interventions.
Furthermore, we are aware that we are analyzing CEs and CSR reports and not real ethical practices regarding employees; although we think that the content of these documents can provide an interesting analysis and give information about the employee policies that companies voluntarily define and disclose, further investigation, e.g. through questionnaires or interviews, could deepen this analysis.
Finally, an interesting area for further research developments may be the analysis of the bodies and procedures existing in companies and dedicated to BE and CSR. Because of the close associations highlighted, in fact, the composition and functioning of the bodies dedicated to these issues should also ensure spaces for coherence and collaboration, opening research directions in the field of corporate governance and organizational studies as well.
publications and reviewer of several scientific journals. Katia Furlotti is the corresponding author and can be contacted at: katia.furlotti@unipr.it Tatiana Mazza, Associate Professor in Business Economics at the University of Parma. She has been visiting PhD at Louisiana State University (USA). Her research interests include several topics of auditing and accounting (audit rotation, enforcement, internal audit, internal control, earnings management, corporate governance and voluntary disclosure). She is a member of American and European Associations (EAA, AAA) and Italian Academy (AIDEA, SIDREA). She is the author of publications and reviewer of several scientific journals.
For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: permissions@emeraldinsight.com