Restraints and enablers of green initiative-taking among hospitality employees: a mixed-methods approach

Abstract

Green initiative-taking, an employee’s self-starting opportunity-seeking action to improve environmental performance is a desirable outcome of organizations’ green policies. Given prior inattention to this area of study, it is unclear what fosters green initiative-taking, and why. This study attempts to answer these questions using a mixed-methods approach. First, an exploratory qualitative study was conducted. Green human resource management, eco-silence, supervisor bottom-line mentality, and co-worker voice emerged as the major themes of employees’ experiences when seeking to engage in green initiative-taking. Second, building on social information processing and social learning theories, a quantitative study proposes a conceptual model of the inter-relationships between the themes that emerged from the first study. Results from a multinational multisource time-lagged quantitative study support most of the hypotheses and shed light on avenues for future research. It suggests that supervisor bottom-line mentality inhibiting green initiative-taking might be standard procedure bottom-line mentality rather than profit bottom-line mentality. Post-hoc, to enhance the study’s applicability, a fuzzy-set analysis was conducted to offer managers the configurations that best yield green initiative-taking among hospitality employees.

Keywords:

• Mixed methods
• GHRM
• employee eco-silence
• green initiative-taking
• co-worker voice
• bottom line mentality
• fsQCA

Introduction

Employees’ role in achieving environmental performance with important employee green behaviors like green process innovation, voluntary green behavior, environmental commitment, extra-role, and in-role green behavior have been researched (Amrutha & Geetha, 2020; Darvishmotevali & Altinay, 2022; Irani et al., 2022; Ye et al., 2022). From the environmental management performance perspective, the role of employees in the effective management of the environment is crucial, especially in the hospitality industry (Pham et al., 2019; Sourvinou & Filimonau, 2018). The environmental operations management literature pushes this idea further to highlight the role of employees’ environmental proactivity in environmental performance (Graham & Potter, 2015). In hospitality literature, employees’ environmental proactivity research is still emerging as a recent review (Loureiro et al., 2022) contains related constructs such as green creativity, task-related pro-environmental behavior, proactive pro-environmental behavior and organizational citizenship behavior for the environment (Aboramadan et al., 2021; Karatepe et al., 2021; Luu, 2019, 2021; Pham et al., 2020). However, other proactive behaviors such as green role breadth self-efficacy, taking charge and green initiative-taking remain unexplored. Of these, initiative-taking is considered most grounded in proactive attitudes (Crant, 2000) and thus likely to be most crucial for employees’ environmental proactivity.

Green initiative-taking is a self-starting and proactive performance disposition that enables employees actively attack environment-related issues, show great persistence in overcoming challenges that may arise in the pursuit of green goals, and search out and implement opportunities that could improve sustainable work practices and performance (Fay & Frese, 2001; Sok et al., 2021). To minimize negative environmental impact and achieve sustainable environmental performance, organizations need employees who can display such resource-intensive and situated-action activity which is essential for corporate environmental performance ratings. Employees demonstrate initiative-taking behaviors to varying degrees subject to available work resources, action possibilities, peculiarities of the work environment, factors employees interact with, and the context of performance (Ikhide et al., 2023). This suggests a mosaic of green initiative-taking precursors that need to be delineated. However, no study has sought to investigate this.

It is therefore significant to explore antecedents of green initiative-taking for the following reasons. First, environmental management studies, particularly in green human resource management (GHRM) and corporate environmental performance (Graham & Potter, 2015; Irani et al., 2022; Sourvinou & Filimonau, 2018) highlight the need to comprehensively examine the unique interactions and intermediaries that yield specific workplace green behaviors. In the environmental management performance and environmental operations management literature, elements such as organizational attributes, relational factors, as well as organizational and employee green performance measures have emerged as precursors and contingents of different employee green outcomes (Indrayani & Wahyudi, 2020). The antecedents, complex underlying mechanisms and contingencies that could uniquely yield green initiative-taking are however missing. Second, green initiative-taking is a high-leverage concept (rather than just another management fad) that could improve organizational environmental effectiveness and provide a unique resource base, as green initiative-taking is a core ingredient for green strategic orientation in environmental management performance (Deichmann & Ende, 2014; Graham & Potter, 2015). Third, from the initiative-taking perspective, there have been studies that have examined context-specific initiative-taking that occurs in limited domains such as in socialization, stress-coping, and feedback-seeking (Crant, 2000; Ikhide & Ogunmokun, 2022). These studies showed that initiative-taking across different contexts has distinct antecedents. However, no studies have explored initiative-taking in the environmental domain to reveal its distinct antecedents. Finally, although related constructs such as green creativity, task-related pro-environmental behaviour, proactive pro-environmental behaviour and organizational citizenship behaviour for the environment have been studied in the hospitality literature, green initiative-taking conceptually differs because 1) it takes a long-term focus; 2) it is persistent in the face of obstacles; 3) it is goal-directed and action-oriented (beyond idea generation); 4) it is consistent with organizational missions; and 5) it is most grounded in proactive behavior (Crant, 2000).

For these reasons, this study has the following two objectives. First, to explore the green initiative-taking concept and identify its enablers and restraints, particularly in the hospitality sector. This is because the sector is one with major impacts on the environment due to its consumption of abundant resources and energy, production of excessive waste, diverse amenities offered, round-the-clock operational lifecycle, and extravagant consumer behavior (Sourvinou & Filimonau, 2018). Nonetheless, it is also noteworthy that considering the hospitality industry’s unique characteristics such as rigid task definition and highly centralized process of decision-making (López-Cabarcos et al., 2015), green initiative-taking could be difficult though it is essential to successful corporate environmental performance. Second, this study seeks to identify the inter-relationships between the enablers and restraints of green initiative-taking and how best to abate the restraints and boost the enablers.

To achieve these objectives, a mixed-methods research approach is adopted. Of the different kinds of mixed methods design, an exploratory equivalent sequential mixed method design is adopted, and this entails the collection of data in two phases. The purpose of the mixed methods is for “development” (for details, see Molina-Azorín & Font, 2016). In the first phase, using focus groups, the researchers obtained and explored qualitative data to determine the constraints and enablers of green initiative-taking among hotel employees. In the second phase, quantitative data was obtained by the researchers in a multinational multisource field survey to empirically test theoretically grounded hypothesized inter-relationships among the constraint and enabler variables that emerged from the qualitative phase of the study. The qualitative study seeks to answer the what question, as in what the restraints and constraints of green initiative-taking are. Subsequently, the quantitative study attempts to answer the how question, as in how the identified variables from the qualitative study interact to promote or inhibit green initiative-taking. Furthermore, to enhance theoretical development around green initiative-taking, the study primarily draws on the social information processing theory (SIP) and social learning theory (SLT) (Bandura, 1977; Salancik & Pfeffer, 1978).

The study makes the following theoretical and practical contributions. First, it enriches the initiative-taking literature by exploring the concept within the specific context of environmental behavior. Second, for practice, the study offers hands-on insights into how best to elicit and sustain environment-oriented initiative-taking among hospitality employees.

Research design and methods

This study’s design has two phases, each has different methods, goals, and outcomes (Figure 1). We begin with a qualitative phase with categories and themes relating to green initiative-taking as its outcomes. This is followed by a quantitative phase with mechanisms and contingencies that predict green initiative-taking as its outcomes. The detail of each phase is presented next.

Figure 1. Mixed methods research design.

First phase: qualitative study

The procedure employed in obtaining qualitative information was open-ended questions administered during semi-structured focus group interviews. The deficiency of qualitative studies into the subject matter and the simplistic illustrations of previous research necessitate the use of a qualitative exploratory phase to uncover constraints and motivators. Focus groups are a suitable means of exploring underlying mechanisms (Bohnsack, 2004) and it enables participants to engage in open exchange in a casual environment, which allows for more natural conversations to happen instead of rigid question-answer sequences (Garavan et al., 2022).

All participants were employees of 3–5star hotels in Nigeria. HR managers of hotels were approached to suggest 6–10 employees with at least one year tenure who are willing to participate in the study. A total of 8 focus group interviews were conducted electronically between July-August 2021, via Telegram. The interview sessions lasted about 55–90 minutes. Although members of each group were nonacquaintances (from different hotels), with respect to job roles, they were homogeneous. Each group had a minimum of 6, and a maximum of 10 participants with 62% female and 38% male. 50% of participants were between 18–27 years, 35% between 28–37 years, and 15% between 38–47 years. 65% of participants had between 1–2 years tenure, 28% had between 2–5 years and 7% had more than 5 years. 27% work in the restaurant, 11.1% in the front-desk/reception, 9.5% were concierges, 39.7% in cleaning and room service, and 12.7% in procurement (details in Appendix A). Based on the respondents’ consent, all but two of the interviews were digitally recorded for the purpose of transcriptions only, after which the recordings were deleted. For the two sessions not digitally recorded, minutes were taken manually during the interviews. The interviews were conducted in Nigerian pidgin English to foster better self-expression of respondents (The first and second authors are speakers of Nigerian pidgin English).

Questions were developed to guide the interviews, such as

1. What are your positive experiences when you proactively search out opportunities that could improve green work practices and performance? 2. What are your negative experiences when you proactively search out opportunities that could improve green work practices and performance? 3. How do you respond to these experiences? 4. Are there organizational policies and practices that support or deter you in your proactive searching out of opportunities that could improve your green work practices and performance?

Data obtained were analyzed using NVivo. Adopting a hermeneutic phenomenological approach (Colaizzi, 1978; Lauterbach, 2018), the data were coded, and key themes emerged as presented in Table 1. First, the authors independently read the transcripts to understand the key motivators and constraints faced by employees while proactively searching out opportunities that could improve green work practices and performance. This step presented the authors with a complete picture of the relevant issues. Second, sentences, phrases, and words that were of import to the topic at hand were identified. That is, the authors conducted open coding. Third, the authors independently articulated the meaning of the identified sentences, phrases, and words in the previous step. Fourth, the process was repeated by the authors for all responses from each focus group to delineate common themes. Fifth, the themes were integrated by the authors to advance a more cohesive description of the emerging concepts. Finally, sixth, the authors compared their outcomes and jointly developed a lucid narrative of the aggregate themes that emerged, to enhance inter-rater reliability. Differences in coding were settled after discussions with the third author, who reviewed the codes developed from all eight focus groups before arriving at an agreement (Table 1).

Table 1. Qualitative data coding: Aggregated themes, second-order and first-order codes.

First-order codes (Representative quotes)	Second-order codes	Aggregate themes
We are trained on how to reduce the consumption of water, electricity, and gas	Green Training	Green human resource management
We were trained on food recycling through third-party organizations
Trained on how to gently nudge guests to reuse
Management appreciates us when utilities such as gas don’t run out quickly	Green reward
Supervisor struggles to balance between financial and environmental performance	Supervisors’ intense focus on profit maximization	Supervisor bottom-line mentality
Supervisors expect environmental performance to lower costs
Supervisor seems mostly motivated toward performance directly related to financial performance	Supervisors tone down environmental issues
When green issues are raised, managers seem to dissenter and don’t respond as quickly as when other issues are raised
Supervisor tone down environmental management ideas suggested by employees
Colleagues talked more about non-green issues in response to management’s willingness to reward and train toward environmental performance	Colleagues challenging behavior/voice	Co-worker voice
Colleagues challenge management on perceived neglect of other issues
Colleagues suggest ideas to improve overall organizational performance	Voice Engagement
Employees demand feedback on promised improvements in non-green areas
The workforce is bolder to engage the management on all issues
Refuse to ignore environmental issues at work	Employee green feedback	Employee eco-silence
Realized how much green issues we had
Management told us not to ignore green issues	Employee participation
Opinions about green issues were solicited by management
Not out-of-place to talk about environmental issues	Decreasing silence on environmental issues
Not anxious to suggest environmental-related ideas

Second phase: quantitative study

From the first phase of exploratory qualitative study, GHRM, eco-silence, supervisor bottom-line mentality, and co-worker voice emerged as the major themes of constraints and enablers of green initiative-taking among the study’s sample of hotel employees. In this follow-up quantitative study, based on theory and extensive literature review, a conceptual model of potential relationships among these variables is developed (Figure 2) to empirically test how these variables interconnect and relate to green initiative-taking.

Figure 2. Conceptual model.

Theoretical background

Social Information Processing Theory (Salancik & Pfeffer, 1978) posits that individuals, as adaptive organisms use information from their social work environment to interpret events, and their way of functioning, attitude, and behavior becomes a byproduct of the received cues. In simple terms, employees socially construct their attitudes and perception based on environmental cues they receive within the workplace. Supervisors and co-workers are particularly dominant sources of information in the immediate work environment and scholars affirm this (Lu et al., 2022). It is based on the received information and perceived cues from colleagues and supervisors that an employee develops their cognition by processing information conveyed, which in turn results in favorable or unfavorable behavioral disposition at work (Babalola et al., 2020). The focal employee considers what their colleagues or supervisor focus on and use that understanding as a compass about “how things are done around here” or as a representation of what is supported, valued, and expected. Accordingly, employees decide the suitability and applicability of their code of conduct and tend to act based on their behaviors’ situational desirability (Salancik & Pfeffer, 1978). Similar to SIP is the SLT (Bandura, 1977) which for this study is described as the modification in understanding that happens via interactions within one’s social network. SLT entails relational understanding, normative expectations, and cognitive knowledge acquisition. The premise of SIP and SLT thus leads inexorably to the conclusion that what can be learned about an employee’s green outcome can be obtained from looking at the social environment including relationships cultivated at work, and other organizational work contexts within which the behavior occurs, and to which it is adapted to. Both SIP and SLT offer a fitting theoretical lens for this study, as their tenets better relate to less-structured tasks requiring proactivity than well-structured routine tasks (Woodman et al., 1993). In addition, they have both been efficient in recent hospitality pro-environmental behavior research (e.g. Thabet et al., 2023; Vila-vazquez et al., 2023).

Literature review and hypotheses development

Employee eco-silence and green initiative-taking as two direct outcomes of GHRM

The vast conceptualization and practice of GHRM majorly include green staffing, green involvement, green training, green goal, green compensation, and green performance appraisal (Chaudhary, 2020). In the hospitality industry, GHRM is purported to be an antecedent of organizational-level and individual-level pro-environmental behaviors (Farooq et al., 2022; Pham et al., 2019, 2020). On the other hand, building on the hospitality literature’s perspective of silence (Aboramadan et al., 2021), eco-silence can be regarded as a non-voice, failure to voice, reluctance to voice, or suppressive voice behavior choice towards environmental issues. It can also be described as when employees withhold information or suggestions about environmental issues and fail to recommend improvements to procedures designed for green activities and performance. It could entail the reluctance to voice suggestions or speak up about practices within the workplace that contributes to environmental challenges such as pollution, emission of light, sound, or other various environmental pollutants that impact water and air could impede organizational environmental performance.

However, effective human resource management practices have been found to promote a more direct voice mechanism and choice that improves work processes and enhances organizational strategy (Nechanska et al., 2020). Social exchange theory helps explain that employees are likely to develop and reciprocate with a beneficial implicit/felt obligation to their organization (through avoiding Silence) when they perceive their organization aims to address needs and expectations with corresponding human resource management practices (Hu & Jiang, 2018). Employees’ reduced silence about critical issues (like environmental sustainability) can be attributed to human resource management practices (Mowbray et al., 2021). This could be facilitated by the feeling of psychological safety, as GHRM could contribute to employees feeling of psychological safety. GHRM demonstrates the organization’s commitment towards environmental sustainability and such perception could make employees more comfortable being themselves and openly voice opinions when it comes to environmental issues without fear of negative consequences to their career, status, or self-image (Moin et al., 2021). GHRM could also be negatively associated with employee eco-silence because it could promote a green organizational culture with environmental sustainability values, behavior, and participation such as the contribution of ideas to environmental management performance. Likewise, GHRM could also dampen employees’ eco-silence by facilitating green organizational culture wherein voicing opinions to improve environmental responsibility is a norm, appraised, and even incentivized. Thus, it is hypothesized that:

H1: GHRM is negatively associated with employee eco-silence.

According to Chaudhary (2020), GHRM practices reflects an organization’s pro-environmental stance, drawing on social identity theory (Mael & Ashforth, 1992), such an image could cause employees to identify strongly with the organization and display necessary green performance behaviors such as green initiatives where necessary. Darvishmotevali and Altinay (2022), add that employees will be more likely to engage beyond task-related duties in an environmentally friendly way by the influence of their organization’s GHRM practices. Employees’ training, reward, and participation in GHRM practices are essential factors that could stimulate extra-role activities toward the environment based on social exchange theory (Aboramadan et al., 2021). Precisely, green training delivers green skills and knowledge that subsequently help to identify environmental problems and enhances green abilities to proffer novel solutions. Further, due to green training, employees are more cognizant of green standards and act more proactively to promote the spread of environmental values and cheer extra-role green behaviors among colleagues (Pham et al., 2020). Also, appraising environmental performance helps to better understand environmental tasks and information, and boosts disposition towards extra-role green behavior (Pinzone et al., 2016). Equally, involvement in green activities is considered a factor that enhances ecological behaviors and encourages employees to initiate novel ideas for green activities (Masri & Jaaron, 2017).

Finally, as earlier mentioned, GHRM could facilitate psychological safety which motivates employees to take initiative. When employees feel psychologically safe, they more than overcome the reluctance to not speak up, they are also more likely to take innovative actions for the goals they desire to see realized in the organization.

H2: GHRM is positively related to green initiative-taking.

The moderating role of witnessing a co-worker voice

Witnessing co-worker voice engagements is an exemplary behavior that stimulates positive changes at work and can impact subsequent behavior (Ng et al., 2021). Though employees observe and witness their colleagues’ behavior, only very few studies have considered the enabling (or weakening) effect of witnessing co-workers’ behavior. This is especially significant considering the “role model” and “spillover” effect of relational contacts and sources of social information such as supervisors and co-workers. Chaudhary (2020) argues that GHRM practices do not only directly influence employees’ behavior but do so also through some socio‐psychological or motivational mechanisms and boundary conditions. Applying SIP theory, the relationship between GHRM and eco-silence can be influenced by contextual relational factors such as witnessing co-worker voice. Likewise, drawing on contagion theory, witnessing a co-worker’s expressive voice behavior, in terms of speaking up and voicing concerns or ideas could create a cognitive schema that could guide an observing employee’s voice choice and their attempt to reproduce voice.

We argue that witnessing co-worker voice would strengthen the negative association between GHRM and employee eco-silence. As an essential source of information, the kind of expressive voice choice exhibited by coworkers could deliver the message to employees that they can, rather than remain silent about an environmental concern, express their green-related ideas and opinion that could improve green-compliant processes, outcomes, and objectives at work (Ng et al., 2021). In other words, a voice instrumentality belief can be increased when the employee sees a colleague engage in voice about general concerns. This will more likely make that employee feel more disposed and confident in their competence to replicate that behavior by sharing or being vocal about green-related information that can bring about an impactful and meaningful general change in the workplace, as well as in their behavior such as their ability to take green initiative.

While GHRM has been hypothesized to minimize undesirable green outcomes or enhance certain green work behavior (Chaudhary, 2020; Paulet et al., 2021), its effect has also been posited to be strengthened (or weakened) by relational or contextual work resources (Lu et al., 2022). Consequently, we argue that GHRM may interact with witnessing co-worker voice to influence employee’s eco-silence and that, in turn, may influence green initiative-taking. It is suggested that the strategic and resourceful role of GHRM in enriching employees’ distinctive competencies and in countering unfavorable green behavior (such as eco-silence) may be strengthened by witnessing co-worker voice. Drawing on the SIP, here co-worker voice is an additional source of information and model for action which could weaken hesitations towards voicing green-related concerns and this, in turn, could subsequently enhance green initiative-taking. This is because while GHRM enhances psychological safety and as a result employees speak up more, witnessing voice behaviors could provide a litmus test for the acceptability of voice actions. In addition, speaking up is a strong indicator of initiative-taking, as well as a mechanism through which an employee takes initiative (Deichmann & Ende, 2014). Thus, GHRM will interact with witnessing voice to affect initiative-taking by weakening eco-silence. It is thus proposed that:

H3: Witnessing co-workers voice moderates the relationship between GHRM and Ecosilence such that when witnessing co-worker voice is high, the relationship between GHRM and Ecosilence is stronger.

H4: Witnessing co-worker voice moderates the indirect relationship between GHRM and green initiative-taking through eco-silence such that when witnessing co-worker voice is high, the indirect relationship between GHRM and green initiative-taking through eco-silence is stronger.

The moderating role of supervisor bottom-line mentality (SBLM)

Supervisors fit into the category of factors and information sources that SIP proposes to play a role in enhancing (or attenuating) employees’ workplace perception and eventual display of work behavior and expected performance (Lu et al., 2022). Supervisors provide information resources by communicating expected goals, clarifying work expectations, providing directives about work processes and strategies as well as promoting task attainment. All of these shape employee experience and behavior so that the organization’s goal can be achieved (Bush, 2020). However, there are more conflicts and competition between the different goals of an organization than previously considered, especially as it relates to environmental performance (Lu et al., 2022). For example, earlier literature indicates that economic performance (e.g. earnings maximization, profitability, and operational efficiency) is the principal obligation of an organization (Carroll, 1991). Thus in many cases, environmental responsibility is merely seen as means of boosting the economic benefits and performance of organizations (Ikhide et al., 2021). It is noteworthy that this does not argue for a trade-off between environmental sustainability and financial performance, as more recent research (e.g. Guenther et al. (2018) meta-analytic review) reported a significant positive relationship between environmental sustainability and financial performance (though the effect is small). Nonetheless, the possibility that supervisors may overly focus on and drive certain organizational goals at the expense of other performance goals exists. This notion depicts a bottom-line mentality where supervisors are characterized as having uni-dimensional thinking that revolves around securing bottom-line outcomes (profit often) to the negligence of other competing work priorities and obligations to stakeholders and the environment (Babalola et al., 2021). According to Greenbaum et al. (2012), the bottom line is frequently referred to in terms of financial consequences. In many cases, a bottom-line mentality entails a tunnel vision concentration, which disregards the significance of upholding several other values that could equally ensure the organization’s long-term success (Babalola et al., 2020). Organizational profitability’s preeminence in top-level management interest makes the bottom-line about attaining profitability, although nowadays issues of environmental performance are gaining attention at the top-level management (Amrutha & Geetha, 2020). Similarly, while it has been argued that organizations pay lip service to environmental performance and that unlike financial performance it cannot be effectively measured, calculated, or audited (Norman & MacDonald, 2004), more recent corporate environmental performance studies maintain that environmental performance can indeed be measured and effectively managed (Trumpp et al., 2015). This presents an interesting field to explore as environmental metrics and performances are also increasingly being set at par and on the same pedestal of importance as the economic performance of organizations. Hence, the need to investigate the moderating role of SBLM on environmental-related work frameworks such as our hypothesized association between GHRM and green initiative-taking behavior.

SBLM could be a “mixed blessing” as limited studies have also demonstrated it can facilitate certain desirable outcomes. For example, Chen et al. (2022) found that SBLM promoted work effort and helping behavior through employees’ bottom-line goal commitment. Likewise, SBLM has been found to improve thriving at work (Babalola et al., 2022) and work performance (Zhang et al., 2021). However, none of the positive outcomes are sustainability-related. On the other hand, SBLM has been found to have several undesirable work outcomes. SBLM could affect the full implementation of GHRM practices by undermining green work expectations, goals and strategies communicated to employees as well as green evaluation, rewards, and enforcement. The inconsistencies between organizations’ GHRM and SBLM could result in conflicting and ambiguous situations for employees’ in-role performance as well as employees’ proactivity.

According to Bush (2020), supervisors are boundary factors who shape and decide the extent to which multi-organizational goals are sustained over time. However, what happens when SBLM signals the preeminence of financial performance and interferes with organizational green efforts? According to SIP theory, employees would largely form their cognition and mirror the outlook conveyed by their supervisor. It is therefore hypothesized that SBLM rather than facilitating green values and instigating green behavior among employees through an attentive implementation of GHRM, would weaken and possibly counter the effect of GHRM, and consequently affects employees’ green initiative-taking behavior.

Hypothesis 5: SBLM moderates the relationship between GHRM and green initiative-taking such that when SBLM is high, the relationship between GHRM and green initiative-taking becomes weakened.

Further, supervisors have been linked to subordinates’ voice choice and also silence climate has been said to ensue from the human resources policies and programs organizations adopt (Azevedo et al., 2020; Kaufman, 2015). Nechanska et al. (2020) argue that supervisors may seek to encourage a particular voice choice and behavior by favorably rewarding or appraising it. Drawing on SIP, employees will evaluate their supervisors’ disposition towards certain voice behavior, available opportunities to voice green ideas and supervisors’ encouragement for other green behaviors. Thus, SBLM’s contrast with GHRM will suggest to employees that they have limited opportunities to voice environmental-related opinions or influence organizational affairs relating to green goals. Further, as a result of the supervisors’ minimal effort to implement GHRM and environmental-friendly processes, in favor of financial success, employees will likely tune down their ideas about environmental issues, resulting in eco-silence, which in turn influences their initiative-taking (Nechanska et al., 2020).

Finally, though employees’ voice-related behaviors antedate performance-based behaviors (like green initiative-taking) in response to HRM and supervisor characteristics, we argue that this voice-related behavior could be eco-silence. Eco-silence could consequently weaken employees’ green initiative-taking disposition because it is not appraised or rewarded and because employees are not offered constructive supervisor directives to support green-related work voice and behavior. In a nutshell, GHRM practices although established could be undermined by SBLM as they might not support green voice and initiative-taking behavior.

Hypothesis 6: SBLM moderates the indirect relationship between GHRM and green initiative-taking through eco-silence such that when SBLM is high, the indirect relationship between GHRM and green initiative-taking through eco-silence is weakened

Participants and data collection

Data is obtained to empirically test the relationships among the variables that emerged from the qualitative study. Considering that the qualitative data was obtained from Nigeria, to improve the generalizability of the study’s findings, multinational data was collected from hotel employees in Nigeria and the UAE. The UAE not only provides a diverse national context to test the generalizability of the findings of study 1, but it also offers an ideal hospitality context for the study as a prominent hospitality service provider. Similarly, Nigeria, a top hospitality hub in the sub-Sahara (Adeola & Ezenwafor, 2016), presents an ideal context as its hospitality industry enjoys increased employee innovative behaviors that drive the continuous growth of the sector (Ogunmokun et al., 2020). Finally, in addition to their distinct cultural context, hospitality employees are increasingly aware of the important role of pro-environmental behaviors in both Nigerian and Emirati national contexts (Alameeri et al., 2018; Joshua et al., 2022), which makes it possible to provide support for the practicability of the study’s model. In total, 267 employees voluntarily participated in the survey, yielding 267 valid sample data. This sample seems about adequate to test the hypothesized relationships based on Kline’s (2005) recommendation of a benchmark sample size of 200, and the 10-times rule of Hair et al. (2011). In addition, a post hoc G*Power analysis using the study’s model with the least R squared to generate effect size (at 0.05 significance level), revealed that a sample of 264 has a power of 0.9, which exceeds the ideal 0.8 (Hintze, 2008). Thus, the study’s sample is sufficient.

Only 3-5star hotels were considered using the convenience sampling method for the following five reasons. First, these hotels have satisfactory levels of services and facilities that provide the opportunity to explore green initiative-taking across different hotel departments as this study has done. Second, they are more likely to have many employees which arguably influences environmental behaviors (Fraj-Andre et al., 2009). Third, they are an interesting context as they often serve as “role models” pacesetting for other sectors within the hospitality industry (Sourvinou & Filimonau, 2018). Fourth, they were selected to get enough samples as these hotels have large employee-base. Finally, 3-5star hotels are said to be appropriate for many environmental-management-related studies (Pham et al., 2020).

From September 2021–February 2022, survey questionnaires were distributed and collected via the drop-collect method by research assistants in multi-waves and multisource, to overcome common method variance (CMV). In the first leg GHRM, SBLM, witnessing co-worker voices and respondents’ demographics data were collected. Following earlier time-lagged studies (Joshua et al., 2022), a two-month lag was allowed before the second wave of data collection; when eco-silence and green initiative taking were assessed. Further, for the green initiative-taking, each respondent was required to ask their supervisor to complete that section of the survey on their behalf. In addition, to promote a candid assessment, the questionnaires were submitted directly to the researcher by managers after completion (see Appendix B for respondents’ demography).

Measures

The study’s variables were measured using a Likert scale ranging from 1 (Strongly Disagree) to 5 (Strongly Agree). As presented in Appendix C, Eco-silence was modified from Aboramadan, Turkmenoglu, Dahleez, and Cicek (2021) and Tangirala and Ramanujam (2008), GHRM was adopted from Dumont, Shen, and Deng (2017), witnessing co-worker voice adopted from Maynes and Podsakoff (2014), SBLM is adapted from Babalola et al. (2021) and Greenbaum, Mawritz, and Eissa (2012) and green initiative taking was modified from Frese et al. (1997). Further, green self-efficacy, tenure, gender, and education are included as covariates to control for their effect on the study’s hypothesized relationships (Crant, 2000).

Preliminary analysis and measurement model

The data’s normality distribution was examined using kurtosis and skewness and they are within a satisfactory range (see Appendix C). Before a structural model analysis is conducted, it was essential to evaluate the validity and reliability of the study’s measurement models (see Appendix D). A confirmatory factor analysis (CFA) was conducted using SPSS AMOS23. Results revealed the hypothesized five-factor model has a good fit with the data: χ²[260] = 567.180; χ²/df = 2.181, TLI = .945, IFI = .953, CFI = .952, RMSEA = 0.067 and SRMR = 0.078. Further, each indicator significantly loaded on respective latent factors with standardized loadings greater than 0.5 except for one item each in Eco-silence and SBLM, which were dropped. To further confirm discriminant validity, HTMT ratios were obtained, and the values were satisfactory at < 0.85 (see Appendix E). The average variance extracted (AVE) values are >0.5, the composite reliability (CR) values exceed the 0.7 threshold and all items’ standardized factor loading is > 0.5, thus establishing convergent validity (Ademilua et al., 2020; Cheung & Wang, 2017).

Certain measures have been undertaken to prevent CMV. First, evaluation apprehension was reduced by ensuring that participation in the survey was voluntary and anonymous (Podsakoff et al., 2003). Also, jargon and ambiguous words were avoided in the questionnaire to enhance respondents’ comprehension and accurate responses. Then, the data were collected in two waves and from more than one source. Finally, to discourage respondent fatigue, the questionnaire was designed to be as simple and brief as possible. Then, after data collection, Podsakoff et al (2003) post hoc assessment of CMV using latent factor test was conducted. Findings reveal that CMB is not a threat in the study’s data as the highest variance extracted by one factor is 30.11%, which is below the maximum threshold of 50%. A T-test was conducted to determine if there is a significant difference between the UAE and Nigerian respondents, and the results revealed no significant difference. Thus, subsequent analysis was carried out on the samples without potential group difference considerations.

Structural model and hypotheses testing

The variables’ composite scores were obtained from the data imputation function in AMOS23, they were mean-centered to calculate the interaction terms. PROCESS macro model 1 was employed to test hypotheses 1 and 3. With education, gender, and tenure controlled for, the relationship between GHRM and eco-silence is insignificant; β = .0670, p > 0.10. Thus hypothesis 1 is not supported. Further, witnessing co-worker voice is negatively associated with eco-silence; β = −.5000, p < .000. Most importantly, the GHRM*Witnessing co-worker voice interaction is significant; β = −.2550, p < .001. In detail, GHRM raises eco-silence among employees who scarcely witness co-worker voice (β = .2818, p < .001) while it reduces eco-silence among employees who repeatedly witness co-worker voice (β = −.1479, p < .05). Thus hypothesis 3 is supported. Figure 3 presents a graphical depiction of the substitutionary moderating role of witnessing co-worker voice in the relationship between GHRM and employee eco-silence.

Figure 3. Interaction between witnessing co-worker voice and GHRM on eco-silence.

PROCESS macro model 7 was employed to test hypothesis 4. With green self-efficacy, education, gender, and tenure controlled for, the direct relationship between GHRM and green initiative-taking is positive and significant (β = .1773, p < .01), while eco-silence is positively associated with green initiative-taking (β = .3056, p < .0001). The GHRM*witnessing co-worker voice interaction is negative (β = −.2206, 95CI [−.3271, −.1141]), likewise the index of the moderated mediation is negative (β = −.0674, 95CI [−.1122, −.0293]). In detail, while GHRM significantly raises green initiative-taking through eco-silence among employees who scarcely witness co-worker voice (β = .0701, 95CI [.0236, .1305]), the insignificant opposite is true among employees who repeatedly witness co-worker voice (β = −.0435, 95CI [−.0902, .0041]). Hypothesis 4 is supported.

PROCESS macro model 1 was employed to test hypotheses 2 and 5. With green self-efficacy, education, gender, and tenure as covariates, GHRM is positively associated with green initiative-taking (β = .1201, p < .01). Thus hypothesis 2 is supported. Further, SBLM is negatively associated with green initiative-taking; β = −.3993, p < .0000. Most importantly, the GHRM*SBLM interaction is non-significant; β = −.1236, p > 0.10. Thus hypothesis 5 is not supported. Finally, PROCESS macro model 5 is used to test hypothesis 6. With green self-efficacy, education, gender, and tenure as covariates, the direct relationship between GHRM and green initiative-taking is positive and significant (β = .1159, p < .01), while eco-silence is positively associated with green initiative-taking (β = .2091, p < .001). The GHRM*SBLM interaction is negative (β = −.1077, 95CI [−.2379, .0224]), however, the indirect effect through eco-silence is positive (β = .0273, 95CI [.0017, .0679]). In detail, while GHRM significantly raises green initiative-taking through eco-silence among employees with supervisors that have low bottom-line mentality (β = .1974, 95CI [.0515, .3433]), the relationship is attenuated among employees with supervisors that have moderate bottom-line mentality (β = .1159, 95CI [.0080, .2238]) but non-significant among employees with supervisors that have high bottom-line mentality (β = .0344, 95CI [−.1118, .1807]). Thus hypothesis 6 is supported. Figure 4 presents a graphical depiction of the moderating effect of SBLM in the relationship between GHRM and green initiative-taking in the moderated mediation model (more details on the results in Appendix F).

Figure 4. Interaction between SBLM and GHRM on green initiative-takin.

Post hoc analysis

To enhance the applicability of the study’s findings, a fuzzy set analysis was conducted. This proposes all possible configurations (recipes) of antecedents that yield high levels of green initiative-taking as shown in Table 2.

Table 2. Sufficient configurations toward green initiative-taking.

Configurations	Raw coverage	Unique coverage	Consistency	Description: causal recipes conditions for high green initiative-taking
Configurations for a high level of green initiative-taking = f (GHRM, Co-worker voice, Eco-silence, SBLM)
Eco-silence*GHRM	0.6656	0.0579	0.8724	In hotels with good GHRM, employees who are silent about environmental issues are most likely to engage in green initiative-taking
∼Coworker voice*∼SBLM*GHRM	0.6130	0.0338	0.9044	In hotels with good GHRM, employees who seldom witness co-worker voice and less supervisor bottom-line mentality are most likely to engage in green initiative-taking
Co-worker voice*SBLM*Eco-silence	0.4494	0.0619	0.9119	Employees who witness high co-worker voice, high supervisor bottom-line mentality, and are highly silent about environmental issues are most likely to engage in green initiative-taking

Freq cut-off = 5, Consistency cut-off = 0.90402, Solution coverage = 0.761293; Solution consistency- 0.866578.

Likewise, a necessary condition analysis was carried out, and it does not include a consistency and coverage value >0.9 for any of the configurations, indicating that no single condition or variable is necessary to generate high levels of green initiative-taking.

Discussion and implications

The study’s findings represent a foundational understanding of what leads hospitality employees to engage in or refrain from green initiative-taking. An initial qualitative study revealed four constraints and enabling variables of green initiative-taking among the study’s sample. They are GHRM, eco-silence, supervisor bottom-line mentality, and co-worker voice. A follow-up quantitative study empirically tests six hypothesized relationships among these variables that demonstrate their interconnection and how they relate to green initiative-taking.

Eco-silence and green initiative-taking are examined as integrated green outcomes moderated by witnessing co-worker voice and SBLM, which are contextual relational factors with colleagues and supervisors. Contrary to expectations, the relationship between GHRM and eco-silence is nonsignificant. Though the link between HRM and employee silence is established in the literature (Aboramadan et al., 2021), and the association between GHRM and green outcomes is evident (Paulet et al., 2021), our finding (being the first to examine this green domain of employee silence) opens a new line of research that asks whether the previously established relationships are true in an integrated domain-specific context. Nonetheless, a plausible explanation for the nonsignificant result could be the coexistence of negative and positive influences of GHRM, which could lead to an insignificant relation between GHRM and eco-silence. A positive influence could be that GHRM policies provide employees with the impetus and knowledge to contribute to environmental objectives, as one of the respondents in the focus group mentioned regarding green training:

Dem dey show us how we fit sabi wetin we dey do wey no good for the environment… and dem dey tell us say make we dey follow put mouth for wetin go help the company protect the environment (Verbatim in Nigerian pidgin English)

We are being trained to identify and minimize activities that harm the environment… and they encourage us to not be silent when we should contribute to the company’s environmental performance (English translation)

On the other hand, a negative influence could be better explained using the diffusion of responsibility concept, which suggests that the enhanced availability of information about issues in the workplace does not necessarily result in employees’ response (Hussain et al., 2019). That is, the reluctance to speak up on environmental issues could persist as environmental knowledge becomes widespread within the organization due to GHRM practices. This is because, with the widespread of green knowledge no one takes personal responsibility for speaking up on green issues, since everyone is aware of it and would assumably not be silent about it. The paradoxical potential of human resource policies has been explored in literature from diverse perspectives (see Guerci & Carollo, 2016; Keegan et al., 2019)

As expected, the relationship between GHRM and eco-silence is moderated by witnessing co-worker voice in a reversing moderating effect (hypothesis 3). This finding confirms Lu et al. (2022) claim that relational resources could weaken or strengthen the link between GHRM and green outcomes. Particularly, witnessing co-worker voice is proven to dissuade others from eco-silence. When witnessing co-worker voice is high, the relationship between GHRM and eco-silence is negative, and when it is low, the negative relationship isn’t just attenuated, it becomes positive. This means, witnessing co-worker voice (which could signal voice culture) fosters a negative relationship between GHRM and eco-silence, and when there is no frequent instance of co-worker voice, the negative relationship between GHRM and eco-silence turns positive. This finding further lends credence to the potential co-existence of a positive and negative influence of GHRM.

Further, witnessing co-worker voice moderates the indirect relationship between GHRM and green initiative-taking through eco-silence (hypothesis 4). This supports our supposition that the consequences of eco-silence could include implications for employee performance. However, the moderation is negative. Taken together, through the mechanism of eco-silence, the higher the co-worker voice experienced, the lower the chances of taking green-related initiatives. First, findings from the qualitative study could provide a plausible reason, as respondents report witnessing voice that are not necessarily pro-environmental. For example:

When our oga dem dey tell us say environment this, environment that, my people for work con dey talk say our oga dem no put mouth for things wey dem talk before say dem go do for us…me sef go come reason with dem soh tey if I wan follow our oga dem advise before, I no go follow am again. (Verbatim in Nigerian pidgin English)

Often in response to management’s focus on the environment, my colleagues voice their discontent with what they perceive as management’s neglect of improving our welfare… this discourages me whenever I try to proactively seek out opportunities to improve individual and organizational environmental performance. (English translation)

On the other hand, even when co-worker voice is pro-environmental, it could still mitigate green initiative-taking due to a phenomenon Darley and Latané (1968) describe as the bystander effect. Akin to the diffusion of responsibility concept, Darley and Latané (1968) argued that one of the reasons an individual sometimes refrains from intervening in a situation is the presence of multiple potential actors. That is, it is not because employees are not concerned enough about environmental issues to engage in green initiative-taking, but because of the prevailing GHRM policies at work, they assume that others might take action and their own initiative might be unrequired and redundant (Diekmann, 1985). Thus, the decline in green initiative-taking when co-worker voice increases does not stem from the failure of GHRM in inspiring green initiative-taking through reducing eco-silence, but rather from a likely bystander’s response to other employees. Carnevale et al. (2020) have reported a similar moderating influence of co-worker voice on promotive voice.

Further, the significant relationship between GHRM and green initiative-taking (hypothesis 2) generally supports Hong et al. (2016) findings that certain HRM systems are initiative-enhancing, and the consensus in the literature that GHRM often yields employee green outcomes (Paulet et al., 2021). Equally, based on the findings from the study’s data, the GHRM and green initiative-taking relationship is not contingent on SBLM (hypothesis 5). That is, SBLM does not significantly interact with GHRM’s ability to directly trigger green initiative-taking. This is probably due to initiative-taking’s persistence in the face of obstacles (Fay & Frese, 2001). Thus when employees are faced with supervisors overtly fixated on financial bottom-line success, they do not give up so easily on their green initiative-taking. This could explain why SBLM does not significantly moderate the direct relationship between GHRM and green initiative-taking, as employees’ persistence neutralizes the effect of SBLM. However, this relationship should be further explored using a sample taken from a varied group before a more decisive statement can be made. Nevertheless, SBLM has a significant negative moderation on the relationship between GHRM and green initiative-taking through eco-silence (hypothesis 6). This indicates that based on the study’s model, SBLM significantly moderates the relationship between GHRM and green initiative-taking “only” through the mechanism of eco-silence.

Practical findings from the fsQCA suggest three combinations of configurations that might aid in identifying scenarios that promote green initiative-taking (Table 2). In the first combination, green initiative-taking is most likely in contexts where although there is no speaking up about environmental issues, there are good green human resource policies. In addition, other configurations reveal that green initiative-taking is most likely under conditions where though there is little co-worker voice and low SBLM, there are good green human resources policies as well as, under conditions where co-workers speak up, supervisors have a high bottom-line mentality and employees are reluctant to speak up about environmental issues.

Theoretical contributions

The study’s findings highlight the antecedent role of formal and informal social cues in prompting green initiative-taking, thus supporting the tenets of SIP and SLT that behaviors are formed not only by formal sensemaking cues but also by informal cues. Further, this study contributes to the hospitality literature by being the first to identify the restraints and enablers of green initiative-taking. The study uniquely demonstrates that voice witnessing, employee silence, leadership bottom-line mentality, and green human resource policies interact to facilitate and constrain green initiative-taking among hotel employees. The study’s mixed method approach strengthens the epistemology of sustainable hospitality as it identifies antecedents overlooked in hospitality pro-environmental behavior literature (Loureiro et al., 2022), although they have been acknowledged in proactive behavior literature (Deichmann & Ende, 2014). Likewise, the study highlights the role of employee-to-employee interaction in environmental proactivity, demonstrating that voice-related employee-to-employee interaction could be instrumental in environmental management performance. In addition, prior studies have largely shown a positive association between GHRM and green employee outcomes (Paulet et al., 2021). However, the study demonstrates that the connection between GHRM and green outcomes is not always so straightforward. In certain cases, contingent on the boundary conditions, the relationship could turn positive or negative (as seen in hypothesis 3).

Further, the current study serves as a link for bridging the literature on employee silence and green work behaviors, thus expanding the nomological network of employee silence which has mostly nuanced the motives for silence but overlooked other nuances. It narrows down focus from overall silence at work to silence about environmental issues which should presently be a critical issue of concern, particularly in the hospitality industry. It provides a nuanced perspective on this employees’ voice choice, which is also a participative form of green work outcome that could proceed from green human resource practices but has rarely been researched in literature. Also, building on findings from hypothesis 5, a significant contribution is made by showing that SBLM may not always have negative relations to desired outcomes. As literature has mostly shown SBLM’s negative effects and few positive outcomes, we suggest that it sometimes might not significantly influence workplace outcomes especially as it relates to green initiative-taking because of the latter’s long-term focus and persistence in the face of obstacles. Again, more studies are required to substantiate this. We suggest that future studies test the effect of SBLM on other proactivity constructs such as green role breadth self-efficacy and green taking charge.

Practical contributions

The outcomes of the study have significant practical implications. It demonstrates that engaging employees through practical GHRM is instrumental in hospitality employees’ eco-silence and green initiative-taking behavior.

First, given that GHRM seems not to be enough in preventing eco-silence in the absence of co-worker voice, efforts made at preventing employees’ silence on environmental issues in the hospitality sector should, in addition to GHRM, seek to promote a workplace that impedes general employee silence. Thus, it is suggested that employee silence/voice be incorporated into the indicators of sustainability performance measurement system, which provides hotels with information relevant to the short and long-term management, controlling, planning, and performance of environmental activities. This is because, according to this study, if general employee voice is not witnessed often, ideas about the environment will be suppressed which could negatively impact hotels’ environmental management. Second, as diffusion of responsibility and bystander effect could be undermining the voice and initiative-taking related outcomes of GHRM, organizations should seek to promote collaborative decision making and collective actions rather than merely soliciting individual opinions and actions on green issues. More practically, this can be achieved by keeping groups small, avoiding over-dependence on experts, providing safe spaces for employees to share ideas, and having shared responsibility for outcomes of decisions (Emmerling & Rooders, 2020).

Finally, the results revealed that SBLM could moderate the indirect relationship between GHRM and green initiative-taking. Consequently, to lessen the restraints of green initiative-taking in the hospitality industry, managers should be more aware of the bottom-line mentality they exhibit and seek to reduce an overt bottom-line mentality. More specifically, the bottom-line mentality exhibited in the study’s context may not just be financially related, it could be bottom-line mentality related to the organization’s “uncertainty norms” (Deichmann & van den Ende, 2014). These supervisors might be overly sensitive to uncertainty, apparent riskiness, and potential for failure of green initiative-taking, and thus be operating with a standard procedure bottom-line mentality. Especially when employees’ green initiatives are radical and disruptive, managers may tend to discourage it in favor of standard green organizational protocols to maintain order within the organization. Thus, managers need to be aware that though they might not be having financial considerations to the neglect of competing priorities such as the environment, they could have a fixation on standard organizational procedures to the neglect of competing priorities. Relevant training and capacity-building may help supervisors in this regard.

Limitations and suggestions for future studies

While our sample size is theoretically sufficient, we, however, acknowledge that this could be a potential limitation of the current study. In addition, we recognize that though our data was obtained after the COVID-19 pandemic, the pandemic heightened the uncertainty within the hospitality industry which could have residual effects on initiative-taking behaviors. Further, while this study examined the role of employee eco-silence, future studies can examine the role of eco-voice in similar relationships. This is because the decline of silence does not essentially infer the increase in voice, as they are not diametrically opposite (Brinsfield, 2013). Also, to further nuance the dynamics of eco-silence, it is recommended that future studies examine the motive for eco-silence (e.g. acquiescent, prosocial, opportunistic, and quiescent) because of their differential effects on employee outcomes. In addition, future studies may consider the potential role firm size can play in similar relationships, as smaller-sized firms could potentially profit more from the decline of employee silence (Della Torre et al., 2021).

In future studies, theories such as the uncertainty management theory could be employed to examine how employees use the information on their supervisor to reduce feelings of uncertainty and to assess their supervisor’s trustworthiness before engaging in eco-silence and proactivity. The uncertainty management theory can be very valuable in the integration of concepts such as employee voice/silence, employee initiative, and the role of managers (Takeuchi et al., 2012).

In addition, following the study’s findings as regards green initiative-taking, future studies may consider team (collective) green initiative-taking as an outcome, to determine if the relationship with green human resource policies would be different particularly considering the diffusion of responsibility and bystander effect observed in this study. In addition, while personality or individual factors (such as environmental concern and proactive personality) and others such as leader-member exchange and CSR (He et al., 2021), have not emerged from the qualitative study, future studies could consider their role in promoting green initiative-taking. It may also be suggested that variables such as role ambiguity, task complexity, and interdependence (Crant, 2000) may provide additional insight into the impetus for green initiative-taking. Further, considering the study’s findings regarding witnessing co-worker voice, the role of other forms of employee-to-employee interaction (such as humor) on environmental behavior may be explored.

Following Babalola et al. (2020) we call on future studies to investigate the implications of bottom-line mentality towards diverse bottom-line outcomes, such as prestige (Babalola et al., 2020) and standard procedure (as proposed in this study). Lastly, considering that bottom-line mentality is an evolving construct in literature, it would be valuable to study its antecedents to uncover why leaders are inclined towards a BLM. In particular, further studies can be done to investigate whether certain industries in which leaders operate (such as the hospitality industry) specially motivate them to espouse BLMs (Babalola et al., 2022).

Disclosure statement

No potential conflict of interest was reported by the authors.