Introduction

Events involving ethical scandals in companies such as AIG, Countrywide Financial, Lehman Brothers and Siemens AG continue to attract significant attention. In nearly all these cases, researchers point at dysfunctional ethical environments as one of the main reasons for illegal or unethical behaviour (e.g. Arnaud and Schminke 2012). Several contributions, in particular, have singled out the role of organizational ethical climates in influencing the employees’ cognitive and affective states, and thereby directing or inhibiting dysfunctional behaviour (Vardi 2001; Schminke et al. 2005; Martin and Cullen 2006; Parboteeah and Kapp 2008, Parboteeah et al. 2010).

Despite the wealth of studies that have already recognized the role of organizational ethical climates in ethical behaviour and organizational performance, past research leaves several gaps in our understanding of organizational conditions and managerial interventions which can affect the establishment of ethical climates. In particular, while the few available studies have focused on the role of organizational forms and manager orientation (Martin and Cullen 2006; Mayer et al. 2010), if and how human resource management (HRM) practices can also encourage organizational ethical climates has yet to be fully explored (Parboteeah et al. 2010, for a rare exception on the role of communication and empowerment practices).

Human resource management practices were found to be powerful antecedents of different types of organizational climates. Way and Johnson (2005), in this regard, proposed a theoretical model in which the impact of HRM practices on organizational outcomes is mediated by organizational climates. Empirically, this idea has been supported by different studies that demonstrated, for example, the connection between HRM practices and safety climate (Zohar and Luria 2005) or customer services climate (Rogg et al. 2001). Strikingly, the possible cause-and-effect relationships between HRM practices and organizational ethical climates have yet to be explored and this is another interesting development for HR research and practice. Scholars have in fact recognized the inherently ethical nature of HRM professionals and practices (Wiley 2000). With many theoretical papers that support the merits of proposing HRM practices that act on organizational ethics, only a few have tested the expected impacts empirically—and they have done so by focusing on specific practices (e.g. training in Sekerka 2009; or whistle-blowing in Hassink et al. 2007), rather than on the set of interconnected practices that actually make up the organizations’ HRM system. As a result, both the literature on ethical work climate and that on HR practice share a common interest in bridging the two fields, yet only limited theoretical and empirical work can be found on the issue. Our study seeks to address this gap as its main contribution, testing the impact of the set of HRM practices on organizational ethical climates.

In studying this link, our study also makes a second contribution to literature, by testing the role played by the organizations’ orientation towards sustainability in moderating the link between HRM practices and organizational ethical climate. Indeed, sustainability and ethics seem to be closely connected in times where the global financial crisis and the rising of environmental concerns have led several companies to pay greater attention in coupling a traditional care for shareholder profit with employee well-being and the broader impacts on society as a whole. An orientation towards sustainability—which we view as the dynamic balance of economic, social and environmental performance—can play an important role in impacting on the ethical infrastructures of a company, and maybe leading to superior ethical standards. Yet, few studies have investigated this possible link. The present study addresses these relationships by investigating a specific contribution resulting from an organization’s orientation towards sustainability, i.e. its moderating effect on the link between HRM practice and organizational ethical climates—assuming that the higher the employees’ perception of their company that moves towards sustainability, the stronger the impact of HRM practices on organizational ethical climates.

This manuscript consists of five sections. Following a comprehensive review and synthesis of the relevant literature, a set of hypotheses are advanced, linking organizational ethical climates, HRM practices and corporate sustainability. Next, the key features of the study (i.e. sample, measurements, procedures and analysis) are presented in the methodology section. After describing the measurement model and findings, we discuss the results in the light of previous literature. The concluding section highlights the main implications for HRM theory, teaching and practice, discussing the basic limitations of the study and presenting opportunities for future research.

Theoretical Framework and Hypothesis

Organizational Ethical Climates

At the most basic level, climate is an attribute of an organization and refers to the collection of attitudes, feelings and behaviour that emerge on a daily basis within an organizational context (Ismail 2005). Organizational work climates reflect the shared perceptions that employees hold regarding the policies, practices and procedures that an organization rewards, supports and expects (Schneider and Reichers 1983). Based on the patterns of experiences and behaviour that individuals encounter in their organizations (Schneider et al. 2000), work climates influence employee decision-making about what constitutes appropriate and desired behaviour (Zohar and Luria 2005), by reflecting ‘the way things are done around here’ (Reichers and Schneider 1990, p. 22).

Building on these definitions, an ethical climate refers specifically to ‘the shared perception of what is correct behaviour and how ethical situations should be handled in an organization’ (Victor and Cullen 1988, p. 51). An ethical climate therefore obviously reflects the shared perceptions that employees hold regarding the company’s established policies, practices and procedures, and the behaviour that their organization rewards, supports and expects with regards to ethics. As such, an ethical climate reinforces the normative systems that guide ethical decision-making and behaviour (Victor and Cullen 1988; Vidaver-Cohen 1998).

One of the most dominant perspectives on ethical climates comes from Victor and Cullen (1988). They distinguished ethical climates at three different loci of analysis: individual, local and cosmopolitan. The individual locus refers to the personal beliefs and values that generate individual decision criteria and are the bases of moral reasoning; local locus refers instead to the organization itself, while the cosmopolitan locus refers to the community or society outside the organization (Martin and Cullen 2006; Victor and Cullen 1987). In the present study, we will look at the ‘local’ unit of analysis because intra-organizational aspects (or the local referent) have a more functional and pronounced influence over how people perceive ethical climates (Ford and Richardson 1994; Victor and Cullen 1988) and reflect situations that an organization has the ability to change (Parboteeah and Kapp 2008).

At the local level of analysis (also called ‘organizational’), there are three possible ethical climates, the egoistic, the benevolent and the principled. The egoistic climate is where employees perceive that self-interest can guide their behaviour, ‘even to the possible detriment of others’ (Martin and Cullen 2006, p. 178). As a result, employees are moved by instrumental considerations in terms of maximizing their own utility—i.e. company profit. Employees in the benevolent organizational ethical climate perceive, on the contrary, that interest in the well-being of others should guide their behaviour. As a result, employees are likely to be moved by utilitarian ethics that seek to maximize the utility of the social community to which they are connected or are part of. Finally, employees in the principled organizational ethical climate perceive that their behaviour should be guided by applying and interpreting the rules and norms of conduct that have been formally or informally established within their organization.

The cause-and-effect relationship between the ethical climate of an organization and a number of organizational outcomes can be found in literature. Martin and Cullen’s (2006) meta-analysis sheds light on the particular consequences resulting from egoistic, benevolent or principled organizational ethical climates. For example, they highlight the negative relationship between the egoistic organizational climate and different organizational outcomes based on the grounds that instrumental principles are more likely to produce behaviour that is convenient for the individual, but dysfunctional for the rest of the organization, or convenient for the organization, but dysfunctional for society (e.g. Appelbaum et al. 2005). Barnett and Schubert (2002) found that both benevolent and principled organizational climates instil broader concerns in employees and are more likely to produce behaviour that is consistent in terms of individuals, organizations and society. For example, staying at the local locus of analysis, organizational ethical climates have consequences on: (i) employee commitment (e.g. Schwepker 2001; Cullen et al. 2003), with the positive influence of benevolent ethical climate and negative influence of egoistic ethical climate; (ii) job satisfaction (e.g. Stewart et al. 2011; Shin 2012), with the positive influence of both benevolent and principled ethical climates; (iii) psychological well-being (e.g. Mulki et al. 2008), with the positive influence of both benevolent and principled ethical climates and negative influence of the egoistic ethical climate; and (iv) dysfunctional behaviour (e.g. Appelbaum et al. 2005), with the positive influence of the egoistic ethical climate and negative influence of benevolent and principled ethical climates.

Such comprehensive treatment for understanding the consequences of organizational ethical climates is not matched in the literature of its antecedents. Martin and Cullen’s (2006) and Mayer et al.’s (2009) reviews revealed that the antecedents of organizational ethical climates have been limited to three broad categories, namely, the role of the context external to the organization, organizational forms, and strategic and management-related orientations. These orientations are the most significant for our purposes, since they relate to what managers and employees can do intentionally to affect organizational ethical climates. As the available studies are scarce and heterogeneous, there are problems in drawing clear-cut implications for both theory and practice. Most of the (limited) work has focused, at the individual level, on the role of leaders and various organizational features. Regarding the leaders’ role, early results suggest, in particular, that self-rated (Shin 2012), moral development (Schminke et al. 2005) and paternalistic ethical leadership (Ötken and Cenkci 2012) can positively influence the organizational ethical climate of a company, with a particular impact on its benevolent ethical climate. Looking at the organizational features, some published research has pointed to the importance of enforcing ethical codes (Schwepker and Hartline 2005) and of organizational empowerment and communication (Parboteeah et al. 2010), accentuating in particular that communication is positively associated with principled-local climate and empowerment is positively associated with benevolent ethical climate and negatively associated with egoistic ethical climate.

With the exception of these research works, overall past studies offer very little guidance on the appropriate actions that organizations could or should pursue to achieve a particular organizational ethical climate. Simha and Cullen (2012), in particular, called for a more extensive understanding of the antecedents of organizational ethical climates. Building on the opportunity to extend our understanding of the above issue, the next section reviews, discusses and hypothesizes the potential role of HRM practices.

HRM Practices and Organizational Ethical Climates

Along with Victor and Cullen’s (1988) definition of ethical work climates, we will use the ability, motivation and opportunity theory (AMO theory, Appelbaum et al. 2000) to define the nature of HRM practices.

AMO theory claims that HRM systems can affect employees by enhancing their: (i) ability to perform as expected and achieve specific organizational goals, e.g. by attracting and developing high-performing employees through recruiting, selection, training, etc.; (ii) motivation to perform as expected through contingent rewards and effective performance management; (iii) opportunity to engage in specific behaviour, for instance with inputs such as job design, industrial relations and workforce involvement policies and practices. See Appendix 1 for a list of HRM practices associated with AMO theory.

AMO theory is one of the most common concepts associated with HRM systems (Lepak et al. 2006; Jiang et al. 2012a; Boselie et al. 2005), in particular to test empirically the impact of HRM practices on organizational performances, such as profit and voluntary turnover (e.g. Subramony 2009; Gardner et al. 2011; Jiang et al. 2012b). Previous studies have not yet developed any hypotheses on the impact of AMO practices on ethical work climates. Consequently, we need to develop our hypotheses following two steps. First, we will describe the general argument proposed by Weaver et al. (1999), which connects HRM practices (regardless of AMO theory) to ethical work climates. Later, we will translate this argument to the specific case of AMO practices.

The Impact of HRM practices on Organizational Ethical Climates: General Argument

Weaver and Trevino (1999, 2001) argued that ethical programmes are successful when they move in two complementary directions: (i) establishing new company values and (ii) ensuring that employees comply with these values on a regular basis. Ethical-oriented practices can therefore be both values-oriented, when encouraging a shared commitment to responsible self-managed conduct; and compliance-oriented, when imposing rules or standards of conduct (Thomas et al. 1994, Paine 1994; Weaver et al. 1999; Collier and Esteban 2007).

Values-oriented programmes have the explicit purpose of driving individuals away from egoistic interests and towards broader concerns over the well-being of organizations, colleagues and other stakeholders. In addition, to drive individuals’ pro-social behaviour (Eisenberg et al. 1990), values-oriented programmes can also be expected to exert a wider impact on the climate in which individuals are embedded. They affect, in fact, the diffused perception of the values and behaviour that would be most appreciated within (but not necessarily by) the organization. Once values-oriented programmes advance extrinsic (pro-social and pro-organization) goals, we can expect them to establish the benevolent ethical climate, over and above egoistic climate that focus on the maximization of self-interest as the dominant objective.

Compliance-oriented programmes also have the explicit purpose of incorporating organizational and social goals into the individuals’ decision-making process. Their approach is, however, different since they do not seek (on their own) to encourage proactive and autonomous behaviour, but more simply, to adhere to rules and standards. While these programmes can inject the same concern for the well-being of others, employees have a significantly different experience—as they are asked to comply with rules, follow guidelines, etc. Once values-oriented programmes establish pro-social goals and meaning within an organization, we can expect the principled ethical climate to be established—again over and above egoistic climate.

Building on these premises, once HRM practices embody both value-oriented and compliance-oriented programmes (Weaver and Trevino 2001) we expect them to have a positive impact on benevolent and principled ethical climates, and a negative impact on egoistic ethical climate.

Having established these basics, we will now take a further step and show how AMO practices really embody both value-oriented and compliance-oriented programmes—and we can derive the final hypotheses accordingly.

The Impact of Ability-Enhancing HRM Practices on Organizational Ethical Climates

Several contributions have already recognized the importance of ability-related practices in improving organizations’ ethics (Wells and Schminke 2001; Sekerka 2009; Foote and Ruona 2008). An organization can operate at its best when it employs people who already have high standards of ethical sensitivity, i.e. the ability to recognize that any decision-making situation has ethical implications (Sparks and Hunt 1998). Ethical sensitivity is an important precondition for individual ethical behaviour (Treviño et al. 2006), and should be enhanced by the introduction of specific selection and training procedures on the part of companies (Buckley et al. 2001; Weaver et al. 2005). More precisely, employers can implement (i) branding strategies aimed at attracting employees looking for a work environment with high ethical standards, (ii) selection and recruitment strategies directed towards people who share the same ethical values of the organization, and (iii) extensive training that aligns individuals to organizational values and ethical leadership (Ardichvili and Jondle 2009). These interventions are values-oriented ethics programmes because they seek to identify, or develop, higher ethical sensitivity in employees and a greater capacity for making independent decisions in ethically ambiguous situations.

At the same time, ability-related practices are also compliance-oriented programmes. We can advance two reasons for this. On one hand, providing employees with values that are consistent with the organizations’ ethics is also a way of encouraging them to comply with existing rules and standards. On the other, ability-enhancing interventions provide employees with the competencies they need to understand and follow organizational ethical rules and standards.

Overall, since they act as both value-oriented and compliance-oriented programmes, we can expect ability-related HRM practices to: (i) improve the benevolent and principled ethical climates within an organization, and (ii) reduce egoistic organizational ethical climate. Consequently, we propose the following hypothesis:

Hypothesis 1a

HRM practices aimed at enhancing ethically oriented ability are negatively related to the Egoistic Organizational (local) Ethical Climate;

Hypothesis 1b

HRM practices aimed at enhancing ethically oriented ability are positively related to the Benevolent Organizational (local) Ethical Climate;

Hypothesis 1c

HRM practices aimed at enhancing ethically oriented ability are positively related to the Principled Organizational (local) Ethical Climate.

The Impact of Motivation-Enhancing HRM Practices on Organizational Ethical Climates

Motivation-enhancing HRM practices have also been shown to be relevant antecedents of the employees’ ethical behaviour as well as of the organizations’ ethics (Buckley et al. 2001; Weaver and Trevino 2001). In fact, organizations can introduce several practices having the goal of promoting their employees’ motivation, i.e. ‘the degree of commitment to taking the moral course of action, valuing moral values over other values and taking personal responsibility for moral outcomes’ (Rest et al. 1999, p. 101). Employees might engage in unethical behaviour because they would attain (or perceive to attain) benefits that significantly outweigh their costs—especially in those cases where the organization and not the individual bear the brunt of the costs (Piliavin et al. 1986). Motivation-enhancing HRM practices are explicitly intended to address the ethical values in the organization, since they increase the (perceived) costs associated with unethical behaviour—by means of sanctions and punishments—as well as the benefits associated with ethical behaviour—by means of variable pay, awards, performance, etc. (Buckley et al. 1998; Michaels and Miethe 1989; Piliavin et al. 1986).

Punishments and sanctions represent examples of compliance-oriented programmes, since organizations define a precise set of rules and standards that employees should abide to, while actively discouraging any attempt by their employees to break them. The expected result is that individuals will not engage in unethical behaviour, or indeed deviate from ethical behaviour because the costs would outweigh the benefits. Conversely, rewards (in their broadest meaning) are the example of a values-oriented ethics programme. Rewards send in fact an explicit message throughout the organization regarding the behaviour that is prized the most, and seek out employees to be aligned to such behaviour.

Overall, since they act as both value-oriented and compliance-oriented programmes, we can expect ability-related HRM practices to: (i) improve benevolent and principled ethical climates in an organization, and (ii) reduce egoistic organizational ethical climate. Based on the above, we advance the following hypothesis:

Hypothesis 2a

HRM practices aimed at enhancing ethically oriented motivation are negatively related to the Egoistic Organizational (local) Ethical Climate;

Hypothesis 2b

HRM practices aimed at enhancing ethically oriented motivation are positively related to the Benevolent Organizational (local) Ethical Climate;

Hypothesis 2c

HRM practices aimed at enhancing ethically oriented motivation are positively related to the Principled Organizational (local) Ethical Climate.

The Impact of Opportunity-Enhancing HRM Practices on Organizational Ethical Climates

Last, opportunity-enhancing practices are crucial to the advancement of organizational ethics, because they provide employees with the technical and organizational possibility of engaging in ethical behaviour (Winstanley and Woodall 2006). Employees who have the intrinsic motivation and ability to perform ethically may, however, still be inhibited by a negative social or organizational context. Most notably, people may refrain from engaging in ethical behaviour when colleagues, friends or acquaintances display opportunistic behaviour. It follows that organizations must couple initiatives aimed at improving individual motivation and ability with initiatives that improve the social context in which they are embedded—thus, providing the opportunity for ethical behaviour with no fear of negative repercussions (Claydon 2000). Opportunity-related initiatives move mostly in three directions: (1) developing explicit mechanisms to identify unethical behaviour—such as supporting whistle-blowing, monitoring ethical behaviour through surveys, etc.; (2) developing procedures or standards to communicate the organization’s attempts at being ‘more ethical’ and fair in making decisions—such as visible career mechanisms linked to the respect of ethical standards, the involvement of unions in performance measurement, etc.; (3) developing structured situations or contexts in which individuals are involved in the organization’s ethics—such as volunteer programmes, ‘calling for ideas’ on for ethical problems, etc.

It can be argued that opportunity-related practices are not very different to motivation-related practices in terms of expected results.

Similarly to sanctions and punishments, the encouragement of whistle-blowing and monitoring seeks to orient the employees’ behaviour towards respecting established rules and standards. These are compliance-oriented practices, where the organization finds the way to identify and sanction deviant behaviour. The only difference lies in the locus of intervention, since sanctions and punishments address an individual’s infringement directly, while whistle-blowing and monitoring address the social context in which employees are embedded—in order to prevent or identify deviant behaviour.

Conversely, practices such as calling for ideas, job designing, etc., are values-oriented programmes because they seek to communicate to the organization how importantly ethical behaviour is regarded, and how much the employees’ involvement is appreciated. While a rewards system sends this message directly to the individual responsible for positive behaviour, interventions of this kind provide the infrastructure and climate, where ethical behaviour is possible.

Overall, since they act as both value-oriented and compliance-oriented programmes, we can expect opportunity-related HRM practices to: (i) improve benevolent and principled ethical climates in an organization, and (ii) reduce egoistic organizational ethical climate.

Hypothesis 3a

HRM practices aimed at enhancing ethically oriented opportunity are negatively related to the Egoistic Organizational (local) Ethical Climate;

Hypothesis 3b

HRM practices aimed at enhancing ethically oriented opportunity are positively related to the Benevolent Organizational (local) Ethical Climate;

Hypothesis 3c

HRM practices aimed at enhancing ethically oriented opportunity are positively related to the Principled Organizational (local) Ethical Climate.

The Moderating Role of Corporate Sustainability

The theme of sustainability has recently been of burgeoning interest to corporations, industries, regions, communities and policy-makers (Mohrman and Shani 2011). This increased focus has broadened the lens of social responsibility and environmental sustainability to searching, at a corporate level, for a more dynamic balance between economic, social and environmental performance—what is labelled in this manuscript as corporate sustainability. A corporate sustainability orientation has led many companies to integrate corporate sustainability into their strategy, policies, structures and management systems (Docherty et al. 2008; Lindgreen et al. 2008; Mohrman and Shani 2011).

A corporate sustainability orientation has an inherently ethical dimension (Steurer et al. 2005; Becker 2012; Sadler-Smith 2013; Florea et al. 2013). For example, the concept of corporate sustaincentrism developed by Gladwin et al. (1995) highlights the ethical dimensions of sustainability by clarifying its guiding principles—namely, inclusiveness, interconnectedness and equity—which are all distinguished by a moral compass. More recently, Florea et al. (2013) have suggested that four moral values—namely, altruism, empathy, positive norms of reciprocity and private self-effacement—spread among sustainability-oriented organizations. Consequently, an orientation towards sustainability would push managers and employees into paying more attention to ethically related issues. Indeed, the typical sustainability orientation model is not limited to paying proactive attention to stakeholders or voluntary environmental practices, but is an organizational way of life (Sharma and Henriques 2005). Because of that, we expect an orientation towards sustainability to establish an ‘ethical framework’ in which HRM practices have greater impact on ethical climates.

Within the present study, we thus hypothesize that corporate sustainability—being defined by an inherently ethical dimension—moderates the relationship between HRM practices and organizational ethical climates. In particular, the higher the employees’ perception of an orientation towards sustainability on the part of their company, the stronger the impact of HRM practices in increasing its principled and benevolent ethical climates and reducing its egoistic ethical climate. As such, we advance the following hypotheses:

Hypothesis 4a

Corporate Sustainability moderates the relationship between HRM practices (Ability, Motivation, and Opportunity) aimed at enhancing ethics within an organization and the Egoistic Organizational (local) Ethical Climate, namely, the higher the employee perception of corporate orientation towards sustainability, the higher the negative impact of HRM practices on the Egoistic Ethical Climate.

Hypothesis 4b

Corporate Sustainability moderates the relationship between HRM practices (Ability, Motivation, and Opportunity) aimed at enhancing ethics within an organization and the Benevolent Organizational (local) Ethical Climate, namely, the higher the employee perception of corporate orientation towards sustainability, the higher the positive impact of HRM practices on the Benevolent Ethical Climate.

Hypothesis 4c

Corporate Sustainability moderates the relationship between HRM practices (Ability, Motivation, and Opportunity) aimed at enhancing ethics within an organization and the Principled Organizational (local) Ethical Climate, namely, the higher the employee perception of corporate orientation towards sustainability, the higher the positive impact of HRM practices on the Principled Ethical Climate.

Method

Sample

For this study, we sent an online questionnaire to 6,000 employees in six different EU countries: Italy, Germany, Poland, UK, Spain and France. We organized the data collection in partnership with a global company which, in the six above-mentioned countries, provides workforce solutions, such as training and development, employer and employee matching, outplacement services, HR consulting and payroll outsourcing. We built a probabilistic sample, based upon gender, age, area, industrial sector and occupation, and, for each country, the company delivered questionnaires until they obtained 1,000 observations that were consistent with the probabilistic sample created. A cover webpage clearly stating the purposes of the study and the respondents’ rights for anonymity was sent with each questionnaire. Table 1 provides specific information regarding the key features of the sample.

Table 1 Information on the sample
Full size table

Measure

Organizational Ethical Climates (Egoistic, Benevolent, Principled)

Ethical climate questionnaires (ECQs) have been used in many studies where ethical climate is the principle variable (Parboteeah et al. 2010). Although Victor and Cullen proposed nine possible ethical climate types, the local type is one of the best to understand the employees within an organization (Parboteeah and Kapp 2008). Employees were asked to rate the organizational ethical climate as they perceive it in the workplace on a six-point Likert-type scale, where 1 is ‘Strongly disagree’ and 6 is ‘Strongly agree’. Example of items included are ‘What is best for everyone in the company is the major consideration here’ and ‘It is very important to follow the company’s rules and procedures’.

Cronbach’s alpha was equal to 0.79 for benevolent, to 0.80 for principled and to 0.67 for egoistic ethical climates. Consistently with Hair et al. (2006), we considered Cronbach’s alpha higher than 0.6 to be a good indicator of internal reliability. With regard to the measure of Egoistic climate, we also observed that other studies had accepted the construct with Cronbach’s alpha slightly above 0.60 (e.g. Cullen et al. 2003).

HRM Practices

The measurements for HRM Practices could not be generated exclusively from what was established in literature. We had, therefore, to generate a measurement for the set of HRM practices aimed at developing organizational ethics. In order to do so, we followed three steps. First, we carried out an extensive literature review that allowed us to obtain a set of practices considered by literature to have the aim of developing organizational ethics (mostly based on Beatty et al. 2003; Buckley et al. 2001; Budd and Scoville 2005; Caldwell et al. 2011; Deckop 2006; Greenwood 2002, 2013; Pinnington et al. 2007; Weaver and Trevino 2001; Wiley 2000; Winstanley and Woodall 2000, 2006). Later, given that the reliability of a measurement hinges on the precision and clarity of the questions being asked (Steinberg 2004), and that the content validity of a measurement is verified if it reflects important contents of the domain being measured (Carmines and Zeller 1991; Donald 2003), we created six ‘national’ steering committees for the six countries covered in the research. Each committee included five HRM professionals operating in each country. The committee had the purpose of assessing whether the proposed HRM practices were relevant for the purpose of the study, and if the proposed survey items were clear and adequate. Following the input of the six committees, we considered as being out of our scope practices that were not clear or were not applicable to organizations with specific features (in terms of size, industry, country, corporate governance, etc.) or employees with specific features (in terms of organizational level or job types). Given these inputs, the authors went back to the literature and modified the items accordingly. The revised questionnaire was again sent to the steering committees to fine tune the wording.

The questionnaire was then submitted to the employees, who were asked to rate to what extent their companies implement the particular set of HRM practices—divided according to the AMO theory—on a five-point Likert-type scale where 1 is ‘Never’ and 5 is ‘Always’. Respondents used a Likert frequency scale to evaluate how frequently these practices are implemented in their organization. Examples of the practices included in the questionnaire are: ‘presence of ethical leadership programmes’ (seen as an ability-enhancing practice), ‘promoting awards for good citizenship’ (seen as a motivation-enhancing practice) and ‘encouraging the reporting of unethical behaviour’ (seen as an opportunity-enhancing practice). Cronbach’s alpha coefficient was equal to 0.91 for ability-enhancing HRM practices, 0.92 for motivation-enhancing HRM practices and 0.91 for opportunity-enhancing HRM practices. The questions included in the final questionnaire are given in Appendix 1.

Employee perception of Sustainability (Sustainability)

Questions relating to sustainability were adapted from the 2012 MIT Sloan Management Review Research Report (Haanaes et al. 2012). The questions were selected for this study because they implicitly assume sustainability to be a continuous process flowing from the past, through the present, to the future. Before listing the specific questions, the respondents were given the following definition of sustainability: ‘The dynamic balancing of economic performances, social performances and environmental performances’. Employees were asked to rate the status of sustainability in their companies’ agenda—in terms of management attention and investment—today, how has the organization’s commitment towards sustainability—in terms of management attention and investment—changed in the past 3 years, and how they expect their organization’s commitment towards sustainability—in terms of management attention and investment—to change over the next 3 years. Cronbach’s alpha coefficient was equal to 0.82. The list of questions used is given in Appendix 2.

Procedure

In this study, the items in the questionnaire were translated from English into Italian, German, Polish, Spanish and French, and then back-translated to ensure the accuracy of the translations. Prior to the full survey, we ran a pilot test of 20 employees working in the companies of the HRM professionals belonging to the steering committees. During the pilot test, participants were encouraged to ask questions so that the ‘national’ version of the questionnaire could be amended where there was the risk of any ambiguity. The results of the pilot test indicated that respondents had no problems in understanding the questions, and the items included in the questionnaire were then back-translated into English by a professional translator. Three English-speaking experts checked that the back-translation corresponded to the original. The survey was then extended to the entire sample of employees.

It must be noted that selecting employees as survey informants came from an explicit indication found in literature—where there is the distinction between ‘intended’ or ‘implemented’ HRM practices (which provide a managerial perspective) and ‘perceived’ HRM practices, which reflect the employees’ perceptions (Nishii and Wright 2008).

Several studies have recognized the low correlation between employee and manager perceptions in terms of corporate HRM practices (Den Hartog et al. 2013; Liao et al. 2009)—which means that managers and employees have intrinsically different perspectives on the state of the implementation of HRM practices (Nishii and Wright 2008). We then decided to focus on the employees’, rather than the managers’, perception, because this is the most significant unit of information to use for understanding the consequences of HRM practices (Purcell and Hutchinson 2007).

Analysis

We followed the same data analysis approach used by Parboteeah et al. (2010) in a closely related study. We carried out a factor analysis to examine the factor structure of the organizational ethical climate and the items relating to AMO practices. The scales with their appropriate items were then developed into constructs based upon their factor loadings. The hypotheses were then tested using partial least squares (PLS) path models.

Results

A PLS path model was used to study the relationships between different variables and the weight played by each on the others. PLS is a family of alternating least squares algorithms, which extend principal components analysis and canonical correlation analysis. The method, developed by Wold (1982), was first introduced in the context of linear multiple regression models; more recently (Lohmöller 1989; Wold 1985), the PLS estimation algorithm has been adopted in the SEM-LV (Structural Equation Models with Latent Variables) framework as an alternative to the covariance structure estimation procedure (Bollen 1989), which makes use of the well-known LISREL procedure. LISREL is the acronym for linear structural relationships, used for the model and also for the computer software developed by Jöreskog and Sörbom (2006). SEM-LV models, known also as path analysis models, make reference to the analysis of the relationships existing between latent, not directly observable, variables, which are also defined as latent constructs.

Partial least square is used as an alternative for situations where the theory is scarce or when, as in the present case, the measurements or variables are not adjusted for a real distribution. Estimations in PLS do not imply any specific statistical model and, therefore, avoid the need for carrying out suppositions concerning the scales of the measurement (Fornell 1982). In this model, latent variables (LVs) are expressed by ability, motivation, opportunity, sustainability, egoism, benevolence and principles.

The sets of exclusive proxy variables linked to the latent variables included in the model should possess a high level of reliability, also called the ‘internal consistency’ of the scale. Cronbach’s alpha coefficient is used for this kind of measurement, obtained as an average of the correlations between every proxy variable pair.

The relationships between latent variables and manifest variables are defined using reflective measure specifications. Generally, PLS path modelling includes two different kinds of outer models: reflective (mode A) and formative (mode B) measurement models. Theoretical reasoning could induce us to choose the outer mode (Diamantopoulus and Winklhofer 2001). In a reflective model, the latent construct exists independently of the measurement, causality flows from the construct to the indicators and changes in the latent variables must precede any variation in the indicators, so that the indicators all share a common theme and are interchangeable.

The difference in casual direction has profound implications for both measurement error and model estimation (Diamantopoulus and Siguaw 2006). MacKenzie et al. (2005) found that construct misspecification affects the results of the structural model analysis. Freeze and Raschke (2007) provide guidelines and suggestions for researchers. The analysis was developed using Smart PLS 2.0 software (http://www.smartpls.de/).

Findings and Hypothesis Testing

To assess the ‘indicator reliability’, we examined the loadings of all the PLS analysis reflective indicators and found that all the outer loadings were greater than 0.4, as suggested by Churchill (1979).

Establishing convergent and discriminant validity in PLS requires an appropriate AVE (average variance extracted) analysis. The values of the AVE indices, which are used to measure the percentage of variance explained by each factor and which are applied within each latent construct, calculated for all the latent variables considered were all greater than 0.5. This confirms the convergent validity and the goodness of the model (Henseler et al. 2009). Discriminant validity was also satisfied, with the square root of the AVE of each construct being greater than the correlation of the specific construct with any of the other constructs in the model.

Applying the hypothesized model, we began with the complete model reported in Table 2.

Table 2 Results from the complete model
Full size table

Based on the results in Table 2, we eliminated the least significant links and ran a ‘reduced model’ for a re-estimation of its parameters. The values of the AVE indices for the ‘reduced’ model were again greater than 0.5 (Table 3), thus confirming the convergent validity and the goodness of the model here also (Henseler et al. 2009). Discriminant validity is satisfied, with the square root of the AVE of each construct being greater than the correlation of the specific construct with any of the other constructs in the model.

Table 3 AVE indices and correlation matrix
Full size table

Therefore, we formulated the model as reported in Table 4.

Table 4 Estimations for the resulting model
Full size table

The path coefficients presented in Table 4 are standardized regression coefficients, which connect latent variables to each other and quantify the direct impact of each explanatory variable on the concepts to which it is linked by constrained causality. Variable significance is assessed by bootstrap re-sampling, using a minimum of 200 resamples of size 6,000. According to Efron and Tibshirani (1998), 200 replications are generally sufficient for estimating standard errors.

Results indicate that Hypothesis 1a is not supported, because this link is not significant (t = 1.629); Hypotheses 1b and 1c are supported, because the links are significant and the coefficients are positive (respectively, β = 0.208 and 0.199); on the contrary, Hypothesis 2a is not supported, because, while the link is significant, motivation directly affects egoism (β = 0.226); Hypotheses 2b and 2c are also not supported, as, for Hypothesis 2b, the link is not significant (t = 1.271), while, for Hypothesis 2c, the link is significant, but the coefficient is negative (β = −0.093). For the third set of Hypotheses, also, the results vary: Hypothesis 3a is not supported, because this link is not significant (t = 0.999), while Hypotheses 3b and 3c are supported, because the links are significant and the coefficients are positive (respectively, β = 0.216, 0.180). Lastly, Hypothesis 4a is not supported, because the moderations are not significant (t = 1.247, 2.133, 1.615); Hypothesis 4b is partially supported, because only the moderations between ability and benevolence, and between opportunity and benevolence are significant and the coefficients are positive (β = 0.048, 0.170); Hypothesis 4c is supported, because the links are significant and the coefficients are positive (β = 0.064, 0.109, 0.052).

The coefficients of determination of the endogenous latent variables R2 are equal to 0.156 for principles, 0.210 for benevolence and 0.051 for egoism; in literature, a value of the index R2 greater than 0.1 is acceptable (Duarte and Raposo 2010) and, for this reason, the model is not sufficiently good to explain egoism.

Discussion

Our study contributes to the field of organizational ethics by providing empirical evidence of the relationship between HRM practices and ethical work climates. Three main results emerge from our study: (i) HRM systems comprising ability-, motivation- and opportunity-enhancing practices influence all sorts of ethical work climates and (ii) different HRM practices have distinct effects on each different type of ethical work climate; (iii) an organizations’ sustainability orientation affects the relationship between HRM practices and ethical work climates.

The Ethical Role of HRM Systems

Past research on organizational ethics had investigated the role of HRM practices almost exclusively at the individual level—showing how they could affect ethical behaviour. The exclusive focus on the individual level is inconsistent with the idea that HRM practices should have a broader impact on organizational life and, in particular, alter the context in which employees operate and make ethical decisions (Way and Johnson 2005). Our results demonstrate that this role exists, since HRM practices alter the climate in which employees are embedded in three ways: (i) by increasing the benevolent climate, whereby ‘taking care of others’ is a shared value in the social system; (ii) by increasing the principled climate, whereby complying ethical rules and standards is also a shared belief and, (iii) surprisingly, by increasing the egoistic climate.

As such, a first general implication of our study is that HRM practices do not simply align employees towards specific ethical behaviour (Werbel and Balkin 2010), but affect the overarching ethical values in which employees are embedded. This is an encouraging result, because it possibly shows that there is a more powerful role for HRM practices within the ethical domain. HRM practices, in fact, are more effective when they create a context where the expected behaviour emerges spontaneously rather than when they instigate a particular conduct directly through incentives or sanctions. Direct incentives to ethical behaviour can be in fact ‘too expensive’ for the organization, and also generate ‘crowding-out’ effects, whereby employees act ethically only when they expect an explicit incentive. Rather, by altering the social and organizational context in which employees are embedded, employees can act ethically regardless of direct incentives.

Different Ethical Roles for Different HR Practices

The results also highlight that each type of HRM practices provides a particular contribution to the ethical work climate. Ability- and opportunity-enhancing practices exert the expected—positive—impact on benevolent and principled organizational ethical climates. The combined effect on principled and benevolent climates provides specific evidence that ability- and opportunity-enhancing practices embody both values-oriented and compliance-oriented ethics programmes (Weaver and Trevino 2001). Consistently with our hypotheses, the generalized ‘perception’ that an organization ‘rewards, supports and expects’ attention to rules (principles) and care for stakeholders and colleagues (benevolence) is signalled when the company creates the conditions for higher ‘sensitivity’ and competencies regarding the topic (ability-enhancing practices) and when it provides social norms and infrastructures explicitly meant for ethical purposes (opportunity-enhancing practices).

Unexpectedly, though, the results do not highlight any significant effect of ability and opportunity practices on egoistic climate. Past research supports two contrasting interpretations of egoistic climate. On one hand, benevolent and principled climates are often negatively correlated with the egoistic climate (Bulutlar and Oz 2009; Parboteeah, et al. 2010), which might suggest that they ‘substitute’ opportunistic, self-interested goals with others oriented towards the well-being of stakeholders and colleagues. This interpretation views egoism as a negative feature of the context, where self-interest equates opportunistic behaviour at the expense of others. On the other hand, the egoistic climate is typically conceived as being distinct from benevolent and principled climates (Victor and Cullen 1988), and the latter do not necessarily affect the former. In this regard, the egoistic climate has a more neutral meaning, simply indicating the pursuit of self-interest. Since practices that increase principled and benevolent climates do not affect egoism, we believe that our study supports the second interpretation. The egoistic climate does not stand in stark contraposition to principled and benevolent climates because employees can pursue both personal and social goals. The organization also ‘rewards, supports and expects’ behaviour and goals that require some degree of self-interest—e.g. career ambitions, production or selling targets, etc. If the pursuit of self-interest does not necessarily harm others or involve unethical behaviour, ability- and opportunity-enhancing practices—which mostly aim at spotting or avoiding unethical behaviour—do not ‘get a grip’.

The results concerning motivation-enhancing practices are compelling, since none of our expectations were confirmed. Rather, motivation-enhancing practices manifest a positive impact on egoistic climate and a negative one on principled climate—with no reported significant impact on benevolent climate. Motivation-enhancing practices therefore work in the opposite direction from what is typically expected by organizations: instead of building a context that is supportive of ethical, pro-social, behaviour, they engender the egoistic climate which nurtures self-interest—which, as we argued before, does not necessarily equate unethical behaviour. While unexpected, the result is not fully surprising. A few previous studies have in fact suggested that motivation-enhancing HRM practices could be cautiously included (e.g. compensation systems and performance appraisal) in programmes concerned with ethics. While it is true that many authors consider motivation-enhancing HRM practices as potentially important for organizational ethics (e.g. Winstanley and Woodall 2006; Claydon 2000), other authors (e.g. Weaver and Trevino 2001) suggest that practice based on punishments and rewards does not necessarily guarantee a working environment where the aim is for people to support one another (and for this reason related to benevolent organizational ethical climate) or to respect the company principles (and for this reason related to principled organizational ethical climate). Our results support the latter interpretation. We can explain it by suggesting that performance goals, bonuses and awards relating to ethical behaviour can contribute towards self-interest, rather than inhibiting it. Employees, in fact, do not react in a disinterested way to bonuses, sanctions, etc., but, rather, they can be interpreted as the means of achieving self-interested goals (such as promotion, money, etc.). Employees may ‘use’ ethical behaviour not as a new intrinsic value, but as an extrinsic motivator. Research from cognitive psychology shows that ethical activities—which individuals perform because of their intrinsic motivation—are affected negatively by extrinsic motivations (Gneezy and Rustichini 2000). Osterloh et al. (2002) described this in terms of ‘crowding out’. The authors argued that linking performance with extrinsic motivators would actually have a negative impact on the employees’ motivation because they would be willing to perform a particular task only when the extrinsic motivators are in place. It follows that individuals who were formerly motivated intrinsically to conduct themselves ethically would progressively consider it to be ‘normal’ to receive incentives in return—and could pursue personal rather than collective interests. So, receiving rewards for normally expected ethical behaviour may be in conflict with some employees’ beliefs that ethical behaviour should be its own intrinsic reward, and that ethical behaviour is diminished if it is rewarded. Therefore, short of customizing them to the specific motivations of individual employees, rewards for ethical conduct can be best presented in the form of long-term rewards, such as promotions, rather than as pecuniary rewards (e.g. bonuses), so as not to diminish the status of ethical behaviour in the minds of people who act ethically.

The Ethical Role of Corporate Sustainability

Last, our results confirmed the role that corporate sustainability orientation has on the different variables examined in this study. The results provide evidence that the perception of such orientation does make a positive difference and should be taken into account. Recent research suggests that the importance for an organization in ‘investing’ in sustainability has, however, remained largely elusive, leading us to wonder whether the increasingly ‘mainstream’ concept of sustainability is nothing more than a managerial fad (e.g. Burritt and Schaltegger 2010). Our results suggest otherwise, highlighting its positive role in establishing benevolent and principled ethical climates. The results further suggest that the overall orientation of a company towards sustainability works as a sort of ‘ethical framework’ within which ability- and opportunity-enhancing practices become ‘more powerful’ in their capacity to build benevolent and principled climates—and so align employees towards ethical behaviour. Accordingly, the perception that the company is moving towards a better sustainability balance can be considered as a key condition for maximizing the impact of HRM practices aimed at developing organizational ethics on the ethical climate of an organization.

Practical Implications, Limitations and Future Research

In this final section, we will present the implications of our findings for HRM practice, research and teaching, the basic limitations of this study and the possible avenues for further research.

In terms of practical implications, we can derive two managerial recommendations for HRM practice from the results of this research. First, the results suggest that there is the opportunity to assign different priorities to the different components of the HRM system (i.e. ability-, motivation-, and opportunity-enhancing practices) when the purpose is to establish principled and benevolent ethical climates. According to our results, we recommend that an HRM department wanting to improve its benevolent and principled ethical climates (shown to be connected to positive outcomes in the workplace, such as employee well-being, job satisfaction and staff commitment) to focus on selection/training processes (ability-enhancing practices) and employee involvement (opportunity-enhancing practices).

The second managerial recommendation concerns the context in which the company’s HRM system affects organizational ethical climates. Results suggest that linking an organization’s overall orientation towards sustainability is critical in order to maximize the impact of HRM ethical practices on organizational ethical climates. Indeed, when included in a more comprehensive organizational strategy that stresses the company’s orientation towards sustainability and makes this clear to its employees, the practices being studied have a greater impact on ethical climates. This requires a vertical integration between the organisation’s strategy regarding people and its overall strategy, as well as a strong collaboration between HRM managers and professionals and other managers in the organization. According to our results, a special working partnership should be established between the managers and professionals responsible for sustainability/corporate social responsibility (CSR) in the company, as highlighted by other studies on the topic (e.g. Gond et al. 2011).

The present study also has interesting implications for HRM teaching. Recent studies on the topic (e.g. Van Buren III and Greenwood 2013) have called for an educational focus to explicitly address the ethical roles and responsibilities of HRM functions and practitioners. Those making the calls suggest that HRM teaching which does not equip students with the wherewithal to manage the ethical implications of any employment relationship cannot help them in assuming a strategic role in their present or future companies. Following the findings that have emerged in this study, we will add our voice to this call: on top of the need to make students aware about the inherent ethical component within an employment relationship, this study has demonstrated that HRM practices have a significant impact on the overall ethics of an organization (i.e. its organizational ethical climates). As a result, HRM education should also provide students with research-based recommendations regarding what HRM practices are, and how and under what conditions they can affect organizational ethics.

As in most research projects, the present study has several limitations that should be acknowledged, all of which provide opportunities for further research. The first limitation regards the fact that it tested the direct correlation between a set of HRM ethical practices and organizational ethical climates, without focusing on the processes and mechanisms under which that relationship takes place. The second limitation relates to our adopted concepts of HRM practices: according to AMO theory, a ‘functional’ view of the HRM systems (i.e. a coordinated set of practices aimed at influencing the employees’ ability to perform, their motivation to perform and their opportunities to perform) is implicitly adopted, while it is still under debate in literature (Guest 2011; Boxall and Macky 2009; Godard 2010), because it does not include any ‘functional’ work practices, such as diversity management or team-working. The third limitation regards the stages of implementation for HRM practices (e.g. intended or actually perceived HRM practices, Nishii and Wright 2008); this study focused on perceived HRM practices, by examining the employees’ point of view. As a result, the study relied on self-reported surveys from one source and at one point in time. In addition, since all data were obtained from a common questionnaire, it is possible that common method variance might have inflated the true relationships between the antecedents and the outcomes. The final basic limitation is about the indicators used for measuring the HRM practices. The questionnaire used measured how frequently specific HRM ethical practices are implemented, without focusing, for example, on their coverage or level of sophistication.

On the basis of the limitations presented above, we will both advance and discuss specific avenues for future research.

The first possible research path relates to studying the correlation between a set of HRM ethical practices and organizational ethical climates. This study can be seen as first step in analyzing that relationship, as certain intriguing points remain open. In particular, future research might focus on the processes and mechanisms whereby that relationship takes place, such as, for example, longitudinal studies which can explain how specific interventions gradually contribute to establishing positive organizational ethical climates. Furthermore, since the model being tested also includes a strict linear view of causality, which goes from inputs to outcomes, future research about the processes that connect HRM practices and organizational ethical climates could test for ‘reverse causality’, following those who argue that outcomes exert an important role in defining HRM practices (Boselie et al. 2005). In addition, in reference to the recent research advancement in the area of moral psychology and behavioural ethics, future research might take into consideration some individual-level variables that can affect the relationship among HRM practices, organizational ethical climates and ethical behaviour. For example, assuming the frequently used framework proposed by Rest et al. (1999), specific individual factors regarding moral awareness, moral judgement, moral motivation and moral behaviour could be tested as moderators of the relationship between HRM practices and organizational ethical climates and those of the relationship between organizational ethical climates and individual ethical behaviour.

A second possible avenue for future research relates to the ‘functional’ view of the HRM systems adopted in this study and presented above. Future research can overcome that particular limitation by testing the impact on organizational ethical climates of both HRM practices—relating to the management of the people who perform the work in the organization—and work practices—relating to the management of the work domain, such as team-working practices (Boxall and Macky 2009). By extension, the model tested in this study can be used as a stepping stone for more comprehensive models that are likely to be multi-level in nature, in order to address the concerns posed by Huselid and Becker (2000) who, in one of the first studies on strategic HRM, recommended that readers should not consider the HRM system as the only possible explanation of organizational performance.

Furthermore, the stages of implementation of HRM practices can open interesting roads for future research. As noted above, this study focused on perceived HRM practices, by considering the employees’ point of view. Concerning the ongoing debate regarding the reliability of raters in HR literature (e.g. Huselid and Becker 2000; Gerhart et al. 2000; Wright et al. 2001), future research can be based on research designs involving multiple respondents, which could acknowledge the points of view of HRM system designers (i.e. HRM managers and professionals) and of the HRM system ‘implementers’ (i.e. line managers). In addition, future research can test the implementation of ethical HRM practices in terms of their quality, using different measurements, related, for example, to the coverage, intensity or sophistication of HRM practices (Boselie et al. 2005). Alternatively, the use of subjective evaluative items could be useful, since HRM literature has argued that inadequate implementation of specific HRM practices may be more damaging than its absence (Wright and Gardner 2003).

A final suggestion for future research relates to the European scope of the present study, which, in HRM terms, is a mixture of homogeneity and heterogeneity (Mayrhofer et al. 2012). Indeed, while this sample is suitable for the current study because it focused on six European countries in order to control heterogeneity, the background and practices of European organizations may limit the transfer of these findings to other cultural or organizational settings. In fact, it is arguable that the establishment of common public policies driven by the European Union—relating, for example, to the development of human capital or the ‘internal’ labour market mobility—has triggered a homogenization process of HRM practices among national states. Recognizing the importance of national cultures and institutional settings for HRM practices, future research in other countries and different institutional settings would be valuable to investigate the possible relationship between HRM ethical practices, sustainability and organizational ethical climates.

In conclusion, the present paper contributes towards the understanding of the impact of HRM practices on organizational ethics, by focusing on their specific relationship with organizational ethical climates. The study found that ability-enhancing and opportunity-enhancing HRM practices are related to benevolent and principled organizational ethical climates, whereas motivation-enhancing HRM practices are related to the egoistic climate. In addition, the findings highlighted the moderating effect of employee perceptions regarding corporate sustainability orientation on those relationships. These results present an insight into the implications for HRM practices, because they can be used by practitioners as a broad guideline for developing specific ethical climates via HRM; for HRM teaching, because these findings emphasize the need for greater integration of ethics in HRM education; and for HRM research, because this study, beyond its actual findings, suggests a direction for critical research that can enhance our understanding of the relationships between HRM practices, organizational ethical climates and a corporate orientation towards sustainability.