Introduction

In the past decade, there has been growing scholarly and academic interest in corporate sustainability (CS) and its intersection with human resource management (HRM) (Cohen 2010; Ehnert 2009; Mariappanadar 2014; Taylor et al. 2012; Arnaud and Wasieleski 2014). Taylor et al. (2012) argue that HRM systems play a dual role with regard to CS because they can operate both as a means to develop sustainability mind-sets, and as an end to promote the physical, social and economic well-being of employees, thus improving a company’s social performance (Rothenberg et al. 2015). Similarly, HRM can develop the environmental performance of the firm, through so-called green HRM practices (Renwick et al. 2013; Paillé et al. 2014; Dumont et al. 2016). As a result, HRM research—e.g., the edited volume by Ehnert et al. (2014) together with special issues in journals such as Human Resource Management, 51(6), Personnel Psychology, 67(4), Journal of Organizational Behaviour 34(2), Organization, 20(3) and Management Revue, 23(3)—has devoted growing attention to the intersection between HRM and CS, theorizing a new HRM concept, labeled “Sustainable HRM” (Ehnert 2009; Kramar 2013). This concept aims at simultaneously preserving, regenerating, and developing the economic, environmental, social, and human resources of the organization, thus supporting the very idea of CS.

Notwithstanding the growing attention to the development of knowledge on the intersection between HRM and CS, and to the key features of sustainable HRM, available evidence shows that the HR professional community is not (yet) playing a key role in the development of CS via sustainable HRM. For example, in the US context, only 6% of members of the HR community consider their HR departments to play a leading role in defining and implementing sustainability strategies (SHRM 2011). Similarly, in the British context, a survey of 523 general managers shows that only 26% of them consider their HR department to contribute to implementing sustainability strategies and that only 13% consider their HR department to contribute to the definition of those strategies (CIPD 2013).

One possible reason for the reluctance of the HR professional community to engage in CS development and in sustainable HRM is that this professional community follows predominantly the implementation of a different HRM concept, called strategic HRM (Westermann-Behaylo et al. 2013; Wilcox 2012; Apostol and Näsi 2014). This concept supports the managerial view of HRM practices and primarily aims at improving the economic performance of the firm (Lengnick-Hall et al. 2009), and it is assumed to be the mainstream view of HRM (Greenwood 2013). As a result, strategic HRM systems are in practice predominant, and push the professional behavior of the HR community towards maximizing the economic performance of the firm. However, the strategic HRM thinking leads to several dissonances and tensions perceived by members of the professional HR community, which are related to their professional identity (e.g., Hallier and Summers 2011; Pritchard 2012; Wright 2008) and to their professional values (de Gama et al. 2012; Glover and Butler 2012; Guest and Woodrow 2012; Keegan and Francis 2010; O’Brien and Linehan 2014; Roche and Teague 2012).

This study contributes to the growing literature on sustainable HRM, interpreted here as HRM practices aiming at economic, social, environmental, and human sustainability. In particular, we extend the growing literature which explores the evolution of HRM focusing the lives of HR managers and professionals (e.g., de Gama et al. 2012), arguing that, with the key features of sustainable HRM being in line with the original values of the HR profession, the members of the HR community are more satisfied and more attached to their employers when they perceive higher levels of implementation of sustainable HRM practices. Specifically, we develop and test hypotheses on the relation between sustainable HRM and two individual-level attitudes of HR managers and professionals, which are job satisfaction (JS) and turnover intention (TI). We theorized and empirically tested an HRM-specific path in which the meaning of HR work is the mediator of the relationship between sustainable HRM, JS, and TI. Our findings, even when controlling for the role played by a set of economic performance indicators related to the employing organization, confirm all our predictions.

As a result, our findings extend available knowledge on the associations between employees’ perception of CS and employee attitudes, and specifically represent a data-driven argument for a more active role of the members of the HR professional community in implementing sustainable HRM practices, thus overcoming the dominance of economic HRM performance goals in the strategic HRM concept which does not (yet) fully take into account the (intended and unintended) impacts of HRM practices on social, environmental, and human sustainability.

Moreover, the present study makes use of the unique opportunity for collaborating with the European Association of People Management (EAPM, the association of national HR professional associations from 29 European countries), with which the authors co-produced the research questions, designed the research, identified the appropriate tools for data collection, collected cross-country data, interpreted the findings, and developed their practical implications.

Theoretical Framework and Hypotheses Development

Key Features of Sustainable HRM

The concept of CS is related to the broader concept of sustainable development, which has been defined by the World Commission on Environment and Development (WCED 1987, p. 43) as “development that meets the need for the present without compromising the ability of future generations to meet their own needs.” That societal-level concept has been translated at the organization level as CS which “entails the preservation, regeneration, and development of the ecological, economic, and social resources of a system” (Senna and Shani 2009, p. 84). In this sense, CS can be seen as a process of organizational change (Smith 2003). When an organization makes the decision to develop its CS, this decision enables sustainability-driven change to take place (Benn et al. 2014; George and Jones 1996). This change drives the implementation of an ordered set of actions to move the organization to a state in which an equal attention to economic, social, and environmental concerns is incorporated into its strategy, systems, and learning mechanisms (Maon et al. 2009; Stead and Stead 1994). Ultimately, sustainability-driven change has the purpose of transforming an organization into an active agent for sustainable development (Lindgreen et al. 2008; Kemp et al. 2010; Wilkinson et al. 2001).

Among the various managerial systems involved in sustainability-driven change, the extant literature has recognized the central role played by the HRM system in facilitating and supporting such change. Indeed, previous authors have discussed the centrality of HRM systems for sustainability-driven change, for three reasons, (i) the very nature of the HRM system, which can have a major impact on the design and implementation of practices that enhance CS (e.g., Ehnert et al. 2016; Guerci and Pedrini 2014); (ii) the ongoing evolution of HRM systems, which must meet the needs of a growing number of stakeholders (e.g., Guerci and Shani 2013, 2014; Jackson et al. 2014); (iii) the inherent tension within HRM management systems between the short and the long term, which is a key tension for CS (e.g., Hahn et al. 2014). Accordingly, the recent literature has widely recognized the central role of HRM systems for sustainability-driven change and has put forward the idea that companies engaging in the development of their CS at some point also invest in the development of a more sustainable HRM. The concept of sustainable HRM has been defined as “the adoption of HRM strategies and practices that enable the achievement of financial, social and ecological goals, with an impact inside and outside of the organization and over a long-term time horizon while controlling for unintended side effects and negative feedback” (Ehnert et al. 2016, p. 90). Thus, sustainable HRM can be included in the set of HRM models alternative to Strategic HRM (Ehnert 2009; Kramar 2013), such as for example ethical HRM (Greenwood 2013), critical HRM (Delbridge and Keenoy 2010) or green HRM (Jackson et al. 2011, 2012). Specifically, while strategic HRM aims at improving the economic performance of the firm via HRM practices, sustainable HRM intends to contribute to the success of the organization in a broader sense, i.e., to simultaneously preserve, regenerate, and develop the economic, environmental, social, and human resources of an organization. In addition, the temporal perspective for evaluating those contributions moves from a short-term perspective to a long-term perspective (Ehnert 2009; Kramar 2013), with continuous control of HRM practices’ externalities (Mariappanadar 2014).

Sustainable HRM as an Antecedent of Job Satisfaction and Turnover Intentions of HR Managers and Professionals

In the present paper, we focus on the association between sustainable HRM and two key employee attitudes of HR managers and professional, i.e., JS and TI. In regard to JS, several definitions are available in extant literature, and the most-used is from Locke who described JS as “a pleasurable or positive emotional state resulting from the appraisal of one’s job or job experiences” (1976, p. 1304). This definition implies that the satisfied employee likes and feels good about his/her job (Cammann et al. 1979). TI is defined as an employees’ “conscious and deliberate willfulness to leave the organization” (Tett and Meyer 1993, p. 262). It implies that an employee intents to find a new job with another employer in the near future, and seriously thinks of quitting his/her job, actively looks for a job outside the organization with the concrete idea to leave the organization (Wayne et al. 1997; Park and Shaw 2013).

The two above-cited attitudes have been explored by a wide set of empirical papers embedded in organizational behavior (OB), and several meta-analyses are today available (e.g., Judge et al. 2001; Park and Shaw 2013). Referring to their consequences, previous research has demonstrated that JS and TI are key antecedents of valuable organizational outcomes. In regard to JS, for example, a comprehensive stream of studies has explored its relation with job performance, and an influential meta-analysis has confirmed that the relation is significant and positive (Judge et al. 2001). Similarly, another influential meta-analysis on TI recently showed that its relation with organizational performance is significant and negative (Park and Shaw 2013). In regard to their antecedents, OB scholars have extensively explored (i) personal characteristics (such as gender, or education), and (ii) “in work” factors (i.e., within-job factors, such as job variety of leader member exchange) which predict JS and TI. More recently, OB scholars started exploring the role that “at work” factors (i.e., factors related to the broader organizational environment) play in predicting JS and TI (see, respectively, Judge et al. 2001; Park and Shaw 2013). Studying those factors, the OB literature started merging with the CS literature, as CS is seen as a managerial area that could satisfy a wide set of employees’ needs, and therefore increase their JS and decrease their TI (Bauman and Skitka 2012). In regard to JS, several papers have found a positive association between employees perception about CS and JS (e.g., De Roeck et al. 2014; Glavas and Kelley 2014; Valentine and Fleishman 2008; Ellemers et al. 2011). In regard to TI, empirical evidence shows a negative association between employees perception about CS and JS (e.g., Jones 2010; Hansen et al. 2011; Sinha 2017; Valentine and Godkin 2017; Chaudhary 2017; Lin and Liu 2017).

In this paper, we merge previous results on the relations between CS and JS/TI with a recent theoretical differentiation developed within the CS field, i.e., the differentiation between embedded and peripheral CS (Aguinis and Glavas 2013). “Embedded” means that sustainability is integrated within a firm’s strategy, routines, and operations. In this case, sustainability is considered to be context-specific because it is incorporated into the core competencies of the organization. “Peripheral” means that sustainability is not integrated into corporate core processes and practices, i.e., sustainability is considered to be context generic because it can be implemented by any organization, for example, via “stand-alone” charity or volunteering initiatives. In particular, we argue that the association between CS and JS/TI is stronger when CS is embedded in the company. Indeed, when sustainability is peripheral, it is more likely that social and environmental responsibilities are assigned only to specific people within the organization (such as a CS department) or external to the organization (such as to the employees of a foundation). In these cases, employees might perceive the social and ecological image projected by the organization not to be authentic, and this might result in unintended negative outcomes, such as low levels of JS and higher levels of TI (Glavas and Godwin 2013). However, when CS is embedded in the company, it is more likely that employees see their own work as having a positive impact on the societal system and on the ecological environment. In line with this argument, the deployment of CS into a set of specific practices can be considered an indicator of the embeddedness of CS in corporate business processes, which results in positive employees’ attitudes. We argue here that HR managers and professionals who work in companies implementing a wide range of sustainable HRM practices (intended as embedded CS) are more likely to perceive their individual work as connected to the development of the sustainability of their organization and, thus, to develop greater JS and lower TI. Therefore, we advance the following hypotheses:

Hypothesis 1a

Sustainable HRM is positively associated with the job satisfaction of HR managers and professionals.

Hypothesis 1b

Sustainable HRM is negatively associated with turnover intentions of HR managers and professionals.

Sustainable HRM as an Antecedent of the Meaning of HR Work

Management research on the concept of meaning of work assumes that the process through which work is considered meaningful by a worker is rooted in their subjective interpretation of work experiences and related social interactions (Baumeister 1991; Brief and Nord 1990; Wrzesniewski 2003). Accordingly, we define “meaning of work” as the outcome of the individual interpreting what work means and the role work plays in the broader context of an employee’s life (Pratt and Ashforth 2003). As a result, and in line with extant organizational research (Rosso et al. 2010), we use the expression “meaning of work” interchangeably with both the type of meaning employees make of their work (“meaning,” assuming in particular positive meanings) and the amount of significance they attach to it (“meaningfulness”).

To develop the hypothesis regarding the positive association between sustainable HRM and the meaning of HR work (as perceived by HR managers and professionals), it is useful to herald some basic contributions from meaning of work studies, in regard to how meaning of work emerges. Indeed, it is important to recall here the differentiation between the processes through which individuals make sense of their work (Bellah et al. 1985; Wrzesniewski 2003), which can be based (i) on monetary and material benefits combined with job security, (ii) on career advancements opportunities, and (iii) on the so-called “calling orientation,” where individuals attach meaning to their work in relation to the extent to which it has a positive impact on the social system. In line with this calling orientation, Rosso et al. (2010) identified seven mechanisms through which meaning of work emerges. In particular, a specific mechanism is transcendence, which refers to connecting or superseding the ego to an entity greater than the self or beyond the material world (Maslow 1971). Through this mechanism, the employee attaches meaning to work in which he/she is subordinated to groups, experiences, or entities that transcend the self (Frankl 1959; Weiss et al. 2004) and to work that provides the perception of contributing to something outside of or greater than the tangible self (e.g., Lips-Wiersma 2002). Following these authors, work transcends the self when it provides the employee with the opportunity to “perceive that he/she is positively impacting broader society or the world” (Rosso et al. 2010, p. 112). On that specific topic, the already-cited recent article that addressed the positive impacts of CS on employees’ attitudes (Bauman and Skitka 2012) has theorized CS to potentially contribute to the satisfaction of four possible employee needs, namely, the need for security and safety, for distinctiveness, for belongingness, and for meaning. Notably, the authors put forward the idea that employees’ needs for security and safety, for distinctiveness and for belonging can be satisfied by using different levers through different organizational interventions, whereas CS is considered to be a key organizational area through which a company can satisfy its employees’ need for meaning, as it broadens the scope of the jobs within the organization (Bauman and Skitka 2012, p. 77). Putting together the above-presented considerations, it is arguable that the calling orientation is an underlying mechanism that creates room for theorizing an association between CS and meaning of work at the individual level. However, as argued above for JS and TI in relation to the differentiation between embedded and peripheral CS, this association makes sense when CS is embedded in the company, because when CS is embedded in the company it is more likely that HR managers and professionals (which we focus in this paper) see their own HR work as having a positive impact on the societal system and on the ecological environment. Assuming therefore that sustainable HRM represents the embeddedness of CS into the HR area, we argue here that HR managers and professionals who work in companies implementing a wide range of sustainable HRM practices are more likely to perceive their individual work as aiming at objectives which transcend economic performance but are related to a greater scope and, thus, to develop greater meaning of their HR work. Therefore, we advance the following hypothesis:

Hypothesis 2

Sustainable HRM is positively associated with the meaning of HR work.

Meaning of HR Work as Antecedent of HR Professionals and Managers’ Job Satisfaction and Turnover Intention

In the last several years, management research has tested the impact of meaning of work on a wide set of desirable outcomes at the individual level (Michaelson et al. 2014). The basic idea behind this stream of studies is that employees who perceive their work as meaningful develop positive emotions that “broaden people’s momentary thought-action repertoires and build their enduring personal resources, ranging from physical and intellectual resources to social and psychological resources” (Fredrickson 2001, p. 19). Thus, employees who perceive their job as meaningful are more likely to find a rationale for the required job effort (Cohen 2008). Drawing on those theoretical premises, management literature provided evidence for the impact of meaning of work on a range of individual outcomes, such as, for example, work engagement (May et al. 2004), organizational commitment and identification (e.g., Cardador et al. 2011; Liden et al. 2000), well-being (e.g., Campbell et al. 1976), and of organizational outcomes, such as job performance (e.g., Fried and Ferris 1987; Grant 2008; Hackman and Oldham 1976), organizational citizenship behavior (e.g., Piccolo and Colquitt 2006), occupational identification (e.g., Bunderson and Thompson 2009), and customer satisfaction (e.g., Leiter et al. 1998).

Following this line of reasoning, researchers have explored the effects of meaning of work on the two employee attitudes considered in the present study, i.e., JS and TI. Specifically, it has been theorized that a work perceived by an employee as meaningful is more likely to activate an appraisal of the job resulting in a positive emotional state (i.e., to enhance JS) and to retain the employee in the employing organization minimizing his/her intent to find a new job with another employer (Rosso et al. 2010). Following the above proposed line of theoretical reasoning, several recent studies confirmed a positive association between meaning of work and JS (e.g., Hackman and Oldham 1976; Wrzesniewski et al. 1997; Glavas and Kelley 2014).

In regard to the second dependent variable addressed in this study, i.e., TI, we found several recent contributions which tested and demonstrated, in different settings and on different kinds of employees, that when an employee attaches more meaning to his/her work s/he is less likely to quit (Arnoux-Nicolas et al. 2016; Wang and Xu 2017; Fouchè et al. 2017; Leunissen et al. 2016). Theoretically, this result has been interpreted in light of the Hackman and Oldham job characteristics model (1976), which predicts that meaning of work, intended as a psychological state, mediates the relation between the characteristics of a work and its attitudinal outcomes on the employee (such as TI).

Therefore, putting together the above-reported theoretical arguments and empirical studies, we advance the following hypotheses:

Hypothesis 3a

Meaning of HR work is positively associated with the job satisfaction of HR managers and professionals.

Hypothesis 3b

Meaning of HR work is negatively associated with turnover intentions of HR managers and professionals.

Sustainable HRM as an Antecedent of Job Satisfaction and Turnover Intentions of HR Managers and Professionals

In sum, in our conceptual model, we argue that meaning of HR work mediates the positive association between sustainable HRM and JS, and mediates the negative association between sustainable HRM and TI. The proposed mediation mechanism which links sustainable HRM and employee attitudes is built on Glavas and Kelley’s work (2014), which theoretically supported (and empirically demonstrated) that CS makes employees feel that their job is more meaningful, as it seems contributing to a cause greater than “simply” making money. These feelings lead to positive employee attitudes towards the organization. Specifically, in the present paper, we extend the mediation model proposed by Glavas and Kelley (2014) arguing that sustainable HRM (intended as a specific form of embedded CS in the HRM area, which makes HRM practices aimed at economic social environmental and human sustainability simultaneously) makes, through the above-presented mechanism of transcendence, the HR work more meaningful. Indeed, sustainable HRM makes HR managers and professionals to perceive their jobs as aimed at to the generation of outcomes which go beyond the solely economic performance, as in the case of the more traditional concept of strategic HRM. Since, in line with meaning of work studies, we then predict that this greater meaning of work attached by HR managers and professionals to their HR work is positively associated with JS and negatively associated with TI, our final model proposes that meaning of work mediates the association between sustainable HRM and JS/TI. Therefore, we hypothesize that:

Hypothesis 4a

The positive association between sustainable HRM and the job satisfaction of HR managers and professionals is mediated by meaning of HR work.

Hypothesis 4b

The negative association between sustainable HRM and turnover intentions of HR managers and professionals is mediated by meaning of HR work.

Methods

Research Process and Actors Involved

The trigger for this study was a set of conversations that one of the authors had with the management of the European Association for People Management, which is the association of national HR associations from 29 European Countries (http://www.eapm.org). In those conversations, the management of EAPM shared the idea to explore the association between sustainable HRM and HR managers’ and professionals’ working lives. Indeed, the management of EAPM was concerned about the possibility that the hard choices and actions that European HR managers and professionals had to take in the times of crises could have undermined the social legitimacy of the HR profession, therefore generating a shared discontent in the members of the HR professional community. In this context, the management of EAPM shared with one of the authors its intellectual interest in exploring whether sustainable HRM, supposed to increase the meaning of work of the HR managers and professionals, could be a way for addressing this discontent. The authors, first, performed an extensive literature review about the state-of-the-art of the HR profession, which was shared with the management of EAPM. On the basis of that review, the decision to activate a research project, aimed at addressing the broad research question about how to improve the professional lives of HR professionals and managers, was made. In order to perform the research project, a researcher-practitioner research team was established, and composed of the authors and eight professionals involved in EAPM. Those professional members were one from each of the six target countries, the manager in charge of the research at the European level and a representative of the President of the association. Concretely, the researcher-practitioner research team met several times, before, during and after data collection for (i) co-producing the research question to be addressed and setting the specific objectives of the study; (ii) developing the research instruments on the basis of some proposals from the authors; (iii) selecting the sample and coordinating the data collection; (iv) collectively interpreting the results and developing practical implications, which actually lead to several interventions implemented by EAPM, and implications for research.

The researcher-practitioner research team decided to develop a survey-based research project, with a questionnaires administered in English (the operating language of EAPM), and directed to the HR department of the organizations. This decision was made for several reasons. First, being HR managers and professionals organizational positions that require a high level of education and a considerable amount of skills and experience, the researcher-practitioner research team recognized that English language is a common requirement for those jobs. Second, since the decision was to focus on HR managers and professionals which are members of HR professional associations connected at the European level (which for example provide their members with training activities and professional resources in English), the researcher-practitioner research team considered their membership to be an additional indicator of their English skills. Last, the choice to have the survey in English seemed in line with current practices adopted by international research projects both in the HRM field (e.g., Piekkari and Tietze 2012) and in the business ethics field (e.g., Rettab et al. 2009).

Procedure and Respondents

The research team collaboratively designed a questionnaire and collected cross-country data across six European countries. The questionnaire focused on both the organization (i.e., sustainable HRM) and the respondent (i.e., HR managers’ and professionals’ JS, TI and meaning of HR work).

In all target countries, two emails from the HR professional association were sent to their members (excluding those members operating as HR consultants). Overall, the response rate was approximately 18%, and the resulting sample consists of 176 HR managers and professionals (97 female; 79 male) from organizations in Italy (n = 88), France (n = 24), the Ukraine (n = 8), Slovenia (n = 15), Macedonia (n = 16), and Croatia (n = 25). On average, the respondents held HR-related positions for approximately 14 years (M = 14.09, SD = 8.85) and worked in either a service—(56.3%) or manufacturing—industry (43.7%). The average number of employees working in the respondents’ organization was 4222.29 (SD = 31082.36).

Measures

To obtain high levels of content validity, we derived our measures from the literature to the greatest extent possible. The development of the survey instrument was performed on the basis of a wide set of literature-based measures. In particular, we followed a three-phase-process: (i) we carried out an extensive literature review in order to collect the measures for the considered constructs used in other studies involving employees in general because we could not rely on measures specifically developed for HR managers and professionals; (ii) the research team selected the measures, considering their “closeness” with HRM-related roles and omitting those considered not clear enough or not applicable to organizations with specific features (in terms of size, industry, country, corporate governance, etc.) or not applicable to HR managers and professionals with specific features (in terms of organizational-level or HRM job types); (iii) we ran a pilot test with 5–10 HR professionals from each country, which confirmed the clarity and appropriateness of the tool. Unless noted otherwise, all items were scored on seven-point Likert scales ranging from 1 = totally disagree to 7 = totally agree.

Sustainable HRM

As a result of our review of available measures, we identified two options: (i) the six-item scale by Orlitzky and Swanson (2006), which includes items related to the extent to which the employment organization devotes resources to sustainability-related issues in recruitment and selection, training to promote sustainability as a core organizational value, training to develop employees’ skills in receptive stakeholder engagement and communication, considering employee social performance in promotions, performance appraisals, rewards, and compensation; (ii) the development of an index based on “the role of HR in corporate sustainability” (CIPD 2013), which includes 16 possible HRM practices that can be used to embed CS in an organization via HRM (e.g., leadership and management training on CS issues, policies to improve employee well-being). The final decision was to go with the latter because this index was considered to be more specific and comprehensive (16 items instead of 6) and more understandable for our sample. Thus, HR managers and professionals were asked to indicate the existence of each of the HRM practices in their organization (0 = non-existent; 1 = existent). We used all 16 items as an indicator for the use of sustainable HRM. To be consistent with the other measures, we rescaled the 0–16 scores to 0–7.

Meaning of HR Work

We used the 10-item Work and Meaning Inventory (Steger et al. 2012) to measure the meaning of HR work. The used scale covers three specific sub-dimensions, which are as follows: (i) experiencing positive meaning at work, and a sample item is “I understand how my work contributes to my life’s meaning”; (ii) sensing that work is a key avenue for making meaning, and a sample item is “My work helps me make sense of the world around me”; and (iii) perceiving one’s work to benefit some greater good, and a sample item is “I know my work makes a positive difference in the world.”

Turnover Intention

Turnover intention was measured by three items (Wayne et al. 1997). A sample item is, “I am seriously thinking of quitting my job.”

Job Satisfaction

The three item Michigan Organization Assessment Questionnaire—JS Subscale (Cammann et al. 1979) was used to measure JS. A sample item is, “All things considered, I feel pretty good about this job.”

Control Variables

We included several variables that may influence the results of this study. First, because this study was carried out across different countries, we controlled for the context of the country based on possible differences that might exist between types of HRM models used in post-communist and Latin countries (Mayrhofer et al. 2012). For instance, organizations in former Soviet countries (i.e., post-communist countries) had to address the transition from public to private organizations (Morley et al. 2012). Therefore, we dummy coded the Ukraine, Slovenia, Macedonia, and Croatia as 0 to represent post-communist countries and Italy and France as 1 to represent Latin countries. That is consistent with previous findings in comparative HRM research across Europe (e.g., Claus 2003, p. 745). Second, in line with other studies exploring the individual reactions to CS (see Gond et al. 2017 for a review), we controlled for the organization’s performance. Indeed, working for an employer with superior performance affects employee attitudes, being organizational success positively related to job satisfaction (e.g., Ryan et al. 1996) and negatively related to turnover intention (e.g., Park and Shaw 2013). The respondents were asked to rate their organizations’ return on investment, market share and earnings compared to their competitors by scoring three items on a scale from 1 = much worse to 5 = much better. A sample item is, “Considering the last 3 years, could you please rate the actual earnings of your company in comparison with competitor earnings?”. Third, we relied on the expertise of HR professionals as respondents of the questionnaire. Therefore, we controlled for their tenure in HRM-related professions. Finally, we also controlled for the industry (0 = services; 1 = manufacturing) and firm size (number of employees).

Analytical Procedure

To test our hypotheses, we followed the two-step procedure recommended by Anderson and Gerbing (1998) using the lavaan package (Rosseel 2012) in R. In a first step, confirmatory factor analysis (CFA) was used to assess the measurement properties of all multiple-item scales used in this study (meaning of HR work, TI, JS and the control variable organizational performance). In a second step, a structural path model was tested. Both the measurement model and structural path model were compared to other plausible models. We used the Satorra–Bentler Chi-square (χ2) difference test, which adjusts for non-normality with ordinal data (Satorra and Bentler 2001), to compare different models. We calculated and reported various indices to assess the fit of both the measurement model and the structural path model. In line with Hu and Bentler (1999), we concluded a good fit between the hypothesized models and the data when the χ2/degrees of freedom (df) ratio was lower than 3, the root mean square error of approximation (RMSEA) and the standardized root mean square residual (SRMR) were lower than .06 and .08, respectively, and when the Comparative Fit Index (CFI) and Tucker Lewis Index (TLI) were close to .95 (Hu and Bentler 1999). Finally, we reported the Akaike Information Criterion (AIK). The lower this value, the more parsimonious a certain model is (Akaike 1974). We regressed country, tenure in an HRM role, industry, firm size and organizational performance on each construct to control for the potential impact of these control variables.

Common Method Variance

Common method variance (CMV) is often considered as a potential problem in self-report research. We observe limited consensus on the extent of the CMV in these surveys (Williams et al. 1989). For example, some authors have empirically demonstrated that these surveys do not have the negative consequences that CMV-related critics have alleged (e.g., Crampton and Wagner 1994; Malhotra et al. 2006). On this premise, we included both ex ante and post ante procedures as recommended by Podsakoff, MacKenzie, Lee, and Podsakoff (2003). First, we tried to avoid CMV by assuring respondents about the confidentiality and anonymity of their responses to reduce social desirability. Second, the survey was carefully developed with attention to minimize bias. To undermine bias caused by preceding questions on subsequent judgments, we used different response scales (i.e., five-point and seven-point Likert scales, binary scales) to minimize the influence of initial ratings on subsequent ratings (Sudman et al. 1996). Third, we added an unmeasured latent method factor to capture the common method variance among all observed variables in the model and to control for its potential bias.

Results

Descriptive Statistics

The means, standard deviations, correlations, and Cronbach’s alphas for all measures used in this study are reported in Table 1. The Cronbach’s alphas for all multiple-item scales show good reliability (meaning of HR work: .91; TI: .88; JS: .83; organizational performance: .81). All correlations are below .80, indicating clearly that multicollinearity is not a problem (Gujarati 2008; Ramanathan 2002). The correlations are in line with the hypotheses. sustainable HRM is positively associated with JS (r = .32, p < .001) and meaning of HR work (r = .30, p < .001), and negatively with TI (r = − .25, p < .01). Meaning of HR work relates positively to JS (r = .72, p < .001) and negatively to TI (r = − .46, p < .001). Furthermore, Table 1 shows that post-communist countries apply more sustainable HRM (r = − .20, p < .01) and experience greater meaning of HR work (r = − .25, p < .001). Organizational performance is related to higher levels of sustainable HRM (r = .34, p < .001), meaning of HR work (r = .39, p < .001) and JS (r = .45, p < .001) and to lower levels of TI (r = − .39, p < .001).

Table 1 Descriptive statistics, correlations, and cronbach alphas
Full size table

Measurement Model

Table 2 presents the results of a series of CFA’s to assess whether all multiple-item scales measured their proposed latent factor (i.e., meaning of HR work, JS, TI and the control variable organizational performance). Sustainable HRM was measured as an index of 0’s and 1’s with no assumed item-intercorrelations and was therefore not subjected to CFA to avoid bias. A four-factor model including the 19 items showed a good fit with the data. After inspection of the factor loadings, one item for meaning of HR work was omitted due to a factor loading lower than .5 (Tabachnick and Fidell 2001). A Satorra–Bentler χ2 test shows that the resulting four-factor model including 18 item fitted the data better (TRd = 37.31, df = 17, p = .003). This model was retained as the best representation of the data. We compared this model with several alternative models in which different factors were assumed to be one factor. Satorra–Bentler χ2 tests show that the retained measurement model showed a better fit than a three-factor model in which TI and JS were combined (TRd = 72.66, df = 3, p < .001), a three-factor model in which TI and meaning of HR work were combined (TRd = 110.47, df = 3, p < .001) and a three-factor model in which meaning of HR work and JS were combined (TRd = 63.19, df = 3, p < .001). We compared the average variance extracted (AVE) and the shared variance (i.e., the squared correlations) between meaning of HR work, TI, JS and organizational performance to assess their discriminant validity (e.g., Fornell and Larcker 1981). All AVE estimates (i.e., 0.54, 0.71, 0.65, 0.60 for meaning of HR work, TI, JS and organizational performance, respectively) were lower than the shared variances (the highest shared variance was 0.53 for TI and JS). Overall, the above results support the convergent and discriminant validity. In addition, Table 2 shows that a one-factor solution did not fit with the data. This model showed a significantly worse fit than our retained measurement model (TRd = 180.64, df = 6, p < .001). This indicates that a single method-factor is unlikely to account for the variance in our data (e.g., Mossholder et al. 1998). To statistically control for CMV, we added an unmeasured latent method factor to the measurement model and allowed each observed variable to load on it. Adding this factor revealed an improved fit (TRd = 12.05, df = 4, p = .02). From this model, the average CMV could be estimated as 9.24% Therefore, we added this unmeasured latent method factor to our hypothesized structural equation model to control for potential CMV.

Table 2 Fit statistics for all tested measurement models
Full size table

Hypothesized Model

Figure 1 presents the results of the hypothesized structural equation model. This model shows a good fit with the data (M1: = 278.69, df = 203, χ2/df = 1.37, RMSEA = .05, SRMR = .05, CFI = .96, TLI = .95, AIC = 15642.29). Collinearity statistics indicate no problems with multicollinearity as the tolerance levels were above .20 and the variance inflation factor (VIF) below 3 (Gujarati 2008). The squared multiple correlations (R2) were .10 for meaning of HR work, .56 for JS and .47 for TI. Consequently, the tolerance levels of meaning of HR work, JS, and TI were .89, .44, and .53, respectively. The VIF was 1.12, 2.27, and 1.89 for meaning of HR work, JS, and TI, respectively.

Fig. 1
figure 1

Hypothesized structural equation model. Note: n = 176. The impact of the control variables is not depicted. Tenure in an HRM-related profession, industry of the organization and organizational performance had no significant impact on any of the variables. Country showed a negative relationship with sustainable HRM (beta = − 19, p < .05) and meaning of HR work (beta = − .19, p < .05). The number of employees was positively related to sustainable HRM (beta = .09, p < .001) and TI (beta = .03, p < .05)

Full size image

We compared the hypothesized structural model (M1) with plausible alternative models. The fit indices of these models are reported in Table 3. At first glance, the fit indices support that M1 has the best fit with the data and is the most parsimonious model. Adding the direct effect of sustainable HRM on TI (M2) did not result in a significantly improved fit (TRd = 0.56, df = 1, p > .05) nor did adding the direct effect of sustainable HRM on JS (M3: TRd = 0.93, df = 1, p > .05), or on both TI and JS (M4: TRd = 7.59, df = 2, p > .05). Moreover, none of the coefficients of these additional paths were significant (p > .05). The findings support that the hypothesized model is the most suitable to validate our hypotheses.

Table 3 Goodness of fit statistics for all structural path models
Full size table

The standardized path coefficients presented in Fig. 1 are in line with the hypotheses. Sustainable HRM is positively related to meaning of HR work (beta = .16, p < .05). Meaning of HR work is negatively related to TI (beta = − .32, p < .001) and positively related to JS (beta = .70, p < .001). For clarity purposes, Fig. 1 does not depict the paths between the control variables and the constructs of interest. The respondents’ tenure in an HRM-related profession, industry of the organization and organizational performance did not relate to any of the constructs. Post-communist countries reported higher levels of sustainable HRM (beta = − .19, p < .05) and meaning of HR work (beta = − .19, p < .05). A higher number of employees was positively associated with sustainable HRM (beta = .09, p < .001) and TI (beta = .03, p < .05).

We followed the bootstrap procedure by Preacher and Hayes (2004) to assess the mediating role of meaning of HR work in the relationships between sustainable HRM and both JS and TI. We computed the unstandardized indirect effects for each of 10,000 bootstrapped samples at a 95% confidence interval. The total effect of sustainable HRM on JS was .31 (SE = .07, p < .001) and the direct was .11 (SE = 0.05, p < .05). The indirect effect through meaning of work was .20 (SE = 0.05) and the 95% confidence interval ranged from .11 to .32. The total effect of sustainable HRM on TI was − .34 (SE = .10, p < .001) and the direct effect was − .17 (SE = 0.09, p = .08). The indirect effect through meaning of work was − .17 (SE = 0.04) and the 95% confidence interval ranged from − .28 to − .10. Analogously, Sobel (1982) tests support the mediating role of meaning of HR work in the relationship of sustainable HRM with JS (Sobel = 2.30, p < .05) and with TI (Sobel = − 2.08, p < .05). Overall, these results are in line with the hypothesized model and suggest that sustainable HRM relates to JS and TI because sustainable HRM contributes to meaning of HR work first.

Discussion

This study was designed, in partnership with EAPM, for exploring the individual-level reactions of members of the HR professional community to sustainable HRM. The results of this study indicate significant support for the hypothesized path model, which predicted that the meaning of HR work mediates the relation between sustainable HRM and JS/TI. In the following paragraphs, we discuss our main findings in the light of extant literature.

The first finding is about the empirically supported positive relation between sustainable HRM and JS (HP1a) and the empirically supported negative relation between sustainable HRM and TI (HP1b). This first finding complements available (and established) evidence on the positive association between employees’ perception of CS and JS (e.g., De Roeck et al. 2014), and on the negative association between employees’ perception of CS and TI (Hansen et al. 2011). Indeed, referring to the distinction between embedded versus peripheral CS, we have focused here on embedded CS, intending sustainable HRM as the deployment of CS principles in the HRM field, which resulted associated with increased JS and decreased TI. This conception of CS allowed us to test a mediation mechanism which assumes that embedded CS (in our case, sustainable HRM) develops within the employees (in our case, HR managers and professionals) more meaningful views of their job, as our second finding demonstrates.

Indeed, our second finding indicates that sustainable HRM is positively related to the perceived meaning of HR work attached by HR managers and professionals to their job (HP2). Because one important criterion of CS and of sustainable HRM is to include societal and ecological concerns into corporate and individual performance criteria, this finding can be interpreted as being in line with the HR roles literature, which criticizes the overemphasis on the strategic role of the HR function and the dominant focus on economic performance in practice and research (see Caldwell 2003; Caldwell et al. 2011; Francis and Keegan 2006). We argue that this particular finding shows the importance of sustainable HRM systems for HR managers’ and professionals’ perception of the meaning of HR work. Indeed, this finding can be interpreted in such a way that the focus on economic performance which characterizes strategic HRM does not fully fit with the expectations and needs of the HR professional community, while this development has reduced the meaning of the HR work as perceived by HR professionals and managers. In fact, several critical concerns have been advanced regarding the concept of strategic HRM. Marchington (2015), for example, recently highlighted how the shift to strategic HRM has pushed the HR community to adopt narratives and metrics associated with short-term and easy-to-measure performance criteria; to become obsessed by satisfying top managers and shareholders, downplaying its responsibilities to other stakeholders and employees; to selectively invest in top talent, ignoring the majority of the workforce; and to manifest a lack of concern for HRM practices implemented by (especially, global) suppliers in the supply chain. Those key features of strategic HRM have pressed the professional HR community to (i) develop an exclusively inward-looking function based on a reification of labor (Islam 2012) and (ii) to constrain its capacity to exercise moral agency and engage in ethical behaviors (Wilcox 2012). Consequently, the HR professional community is today facing a crisis of trust and legitimacy with both employees and the society (e.g., Kochan 2007; Thompson 2011). Not surprisingly, this crisis of trust and legitimacy reverberates in HR managers’ and professionals’ individual experiences, as witnessed by the studies addressing their reaction to the strategic HRM concept. For example, previous studies have shown that this concept generates ethical and professional dissonance (de Gama et al. 2012; Glover and Butler 2012; Guest and Woodrow 2012; Keegan and Francis 2010; Pritchard 2012) and role conflicts (O’Brien and Linehan 2014; Roche and Teague 2012; Wright 2008) in practicing managers and have led several HRM students to reject the idea of defining themselves in terms of any kind of HRM practitioner’s identity (Hallier and Summers 2011). In this context, sustainable HRM can be seen as an opportunity for the HR professional community to reintegrate aspects of social legitimacy or stakeholder needs into the HR function and to maintain high meaningfulness of their work by balancing the economic, environmental, social, and human sustainability performance of the firm (Beer et al. 2015; Guerci and Shani 2013).

Our third finding provides support for the hypothesized positive association between meaning of HR work and JS (HP3a) and the negative association between meaning of HR work and TI (HP3b). Although this finding is in line with previous OB research on individual employees, we extend prior knowledge by showing that the effects of meaning of work are also relevant for the specific group of HR managers and professionals. This finding is in line with Rosso et al. (2010), who highlighted how meaning of work becomes increasingly important for explaining a diverse and wide set of individual reactions of workers at different levels of the organization and with different personal and professional backgrounds.

Last, the predicted mediating role played by meaning of HR work in the relations between sustainable HRM and JS (HP4a) and TI (HP4b) was empirically supported. This mediation mechanism complements those already put forward by previous literature. Indeed, previous studies have proposed two possible mediation processes which link CS to employee attitudes (reported in De Roek and Maon 2016), which are related to (i) organizational identification, assuming that CS increases the external prestige of the employing organization and employees react to that increased prestige developing a feeling of membership (or organizational pride) and fostering their propensity to strengthen their relationship with the organization; or to (ii) social exchange, assuming that CS increases in the employees their belief of being treated fairly by the employing organization and this justice-based belief thereby influences employees’ propensity to enter into an exchange relationship in which they might feel obligated to reciprocate the organization’s favors with positive attitudes. The here proposed (and demonstrated) mechanism based on meaning of work, which has been drawn on the work by Glavas and Kelley (2014), provides an alternative explanation, showing that employees’ perceptions about CS do not only affect the relationship between the employee and his/her employer—as theorized by both the above-cited mechanisms—but also affect the relationship between the employee and his/her own job. Indeed, when CS becomes embedded in ongoing processes (in our case, sustainable HRM), the employee (in our case, HR managers and professionals) attaches more meaning to his/her job which in turn is associated with more JS and less TI. The existence of this third mechanism presents several implications, which are presented in the following section.

Implications, Limitations, and Future Research

The study has implications for practice and education (developed by the researcher-practitioner research team), as well as research implications.

In terms of practice, the research team highlights key implications for three specific actors, namely individual HR managers and professionals, HR professional associations, and senior leaders of organizations. The first set of implications regard individual members of the HR professional community, and in particular those HR managers and professionals aiming to increase their satisfaction with their HR jobs, aiming to decrease their intention to leave their current employer, and aiming to develop the meaning of their HR work. Our results recommend those members to play a proactive role in pushing the organization to develop those HR-related sustainable practices, as they are determinants of their JS, TI, and meaning of HR work. Indeed, although the implementation of sustainable HRM might be difficult for those actors because it requires alignment among organizational leaders (Guerci and Pedrini 2014) and a set of more “structural” features of the organization for example related to corporate governance systems (Martin et al. forthcoming), we recognize that individual HR managers and professional still have the possibility to influence the decision-making processes in that specific direction (Sheehan et al. 2014). Related to those first implications, we developed suggestions for HR professional associations. Indeed, being those associations committed to the objective to make the HR work more satisfactory and meaningful for their members, and being them a key learning tool in the hands of associated HR professionals and managers (Pohler and Willness 2014), HR professional associations should focus their attention to support their members in developing skills related to the “new” idea of sustainable HRM, being this an antecedent of JS, TI, and meaning of HR work. Recognizing that current practices of the most visible HR professional associations seem more oriented to the diffusion of the more traditional paradigm of Strategic (instead of Sustainable) HRM (Westermann-Behaylo et al. 2013), the research team composed of researchers and practitioner which steered the research effort considered this second implication to require a major change in the current culture of leading HR professional associations. The last set of implications regards the senior leaders of those organizations employing high performing HR staff. For those actors, we argue that investing time and resources in developing HR practices which fall into the idea of sustainable HRM, and making those investments visible to the employed HR professionals and managers, could support the organizational HR community to perceive their HR work more meaningful, and therefore could develop their satisfaction with their HR jobs and facilitate their retention.

The researcher-practitioner team interpreted the findings as insightful analytical material also for management education, to help institutions and professors in moving forward what has been called Critical HRM education (Bratton and Gold 2015). This pedagogy in teaching HRM, indeed, emphasizes the need to help practitioners and students in adopting the process of “reflective critique” (Bratton and Gold 2015, p. 498). Indeed our results provide evidence around the positive impact of the implementation of sustainable HRM as an alternative to the dominant strategic HRM concept and could encourage a critique of prevailing assumptions. At the same time, our focus on the individual reactions of the members of the HR professional community to sustainable HRM could support HR learners in “avoiding a sole reliance on sanitized representations of corporate HR,” and in “looking outwards in order to connect personal and workplace problems to larger macro and global social structures” (Bratton and Gold 2015, p. 498).

The above-reported practical and educational implications of the study developed by the researcher-practitioner research team have been seriously taken by EAPM, in order to address the issue that triggered the collaboration, i.e., how to improve the working lives of European HR managers and professionals after the long economic downturn which has characterized the European economy in the last years. Specifically, EAPM made use of the results in two different ways: (i) informing its members about the findings, using professional publications and magazines, as well as the websites of the national associations; and (ii) forming its members about the key concepts included in this research project with a specific attention to the idea of sustainable HRM. The association, specifically, designed a specific session in the training dedicated to young members of the HR professional community, and summarized those contents to the more professionally mature members of the HR community in short sessions included in the annual meetings of the national associations involved in the project.

Last, the study has significant research implications, as future studies should address its basic limitations and extend its scope. This study’s first basic limitation is that it is based on cross-sectional data and that the responding population was a convenience sample confined to specific European regions. Future research could address the generalizability of our findings in probabilistic samples, as well as in more demographically and, as such, more institutionally diverse samples. Particularly relevant are the changes in institutional settings (changes in law, regulations, budget constraints, etc.), which need to be taken into account when examining the addressed relationships. Beginning, for example, from the contextually based HRM theory framework (Paauwe 2009), we recommend that future studies examine the possible impact of the cultural and institutional environment on the adoption of sustainable HRM. Moreover, longitudinal research is needed to examine if the relationships between the cultural and institutional environment, sustainable HRM and outcomes remain stable. A second limitation is related to the fact that the survey was administered in English; indeed, relevant methodological literature on cross-national surveys highlights that a questionnaire in one language (also called Lingua Franca) provides some benefits in terms of comparability, but presents challenges related to the possible effects of cultural differences on the perception of questions and response categories (Harkness et al. 2010). A third key limitation of the present study is its focus on the association between sustainable HRM and individual-level attitudes of HR managers and professionals. It would be interesting to examine the effects of sustainable HRM on actions and behaviors of all the employees within the organization. In addition, future research should not only focus on the employee level (i.e., individual attitudes and behavior) to determine whether sustainable HRM has an impact, but also link sustainable HRM and employee-level attitudes and behavior with relevant organizational-level outcomes. Finally, future research could adopt a paradox perspective and explore how individual HRM professionals experience the tensions related to their professional identity and whether implementing sustainable HRM practices supports HRM professionals in coping with these identity paradoxes over time (see Smith and Lewis 2011).

Beyond the research strategies and approaches that could overcome this study’s limitations, we believe that our study opens up further avenues for future research. First, we call for future studies which could “zoom” into some of the associations that we explored here, but which—given the scope of the present study—we could not deeply assess. Specifically, we refer here to the relations between embedded CS and meaning of work, and between meaning of work and individual-level attitudes. For example, in this paper, we found significant, sizable, and negative correlations between JS and TI, and it might be possible for example that the relation between meaning of work and TI could be mediated by JS (Griffeth et al. 2000). Since testing the specific effects of meaning of work was not the aim of this paper, which was designed in a HRM (rather than in a OB) framework, we leave this opportunity to future studies. In addition, we argue that future studies on the individual-level effects on the HR professional community of the emergence of the strategic HRM concept could shed more light on the relation between this HRM concept and the traditional approaches to employment relations (e.g., Godard 2014). In fact, the historical views on the evolution of HRM research and practices (e.g., Kaufman 2014) show how, originally, personnel management was meant to encompass a broad range of employee concerns and employment policies, in accordance with human relations (Mayo 1933) and human resources (Argyris 1957; McGregor 1960; Vroom 1964) schools of thought, avoiding some of the organization-centric employment practices and perspectives on employees engendered by scientific management (Taylor 1903) and administrative theory (Fayol 1949). Exploring the individual-level reactions of the members of the HR community to the strategic HRM can be a valuable strategy to explore “from within” the contradictions and unintended effects of this HRM concept.

Conclusion

The present study explores the impact of sustainable HRM on the members of the HR professional community, focusing in particular on their JS and TI. The empirical analysis found support for the predicted positive association between sustainable HRM on JS, and for the predicted negative association between sustainable HRM and TI. In addition, in line with growing interest on meaning of work (Michaelson et al. 2014), we theoretically supported and empirically demonstrated that those relations are both mediated by the meaning of HR work. Overall, these findings—which extend the knowledge on the employees’ perception of CS—employee attitudes relationship—can be used to encourage the members of the HR professional community (and their professional associations) to play an active role in supporting their employing organization in the implementation of sustainable HRM practices.