Abstract
Stress causes serious illnesses and damages employee well-being. The mutual gains human resource management (HRM) framework places HRM practices as the custodian of employee well-being in an organisation. This study presents a mutual gains HRM framework which has three components. First, employees can perceive that their organisation enacts HRM practices from two benevolent intentions (a) to help employees perform better and b) to improve their well-being. Second, these benevolent HRM attributions invoke gratitude among employees. Third, gratitude reduces employees’ perceived stress and improves their engagement levels. Fourth, gratitude mediates the relationship between both benevolent HRM attributions, employee stress and engagement levels. Purposive sampling technique was deployed for the collection of data using structured questionnaire from 294 respondents, working in the telecommunications sector of Pakistan. Measurement and structural model validity were tested through structural equation modelling (SEM) using Mplus 7.0. The findings confirmed theoretical connotations among the constructs. The study contributes to the literature by introducing a new HRM framework mediated by gratitude to reduce employee stress levels and improve their engagement.
Introduction
Stress has a detrimental impact on employee well-being and performance (Kowalski & Loretto, 2017; Le Fevre et al., 2003), and stressed employees are more likely to be unhealthy, demotivated, less safe and less productive (WHO, 2007). Given the psychological and physical harm caused due to stress, scholars are increasingly calling to examine how stress can be alleviated and managed in employment settings (Cooper & Quick, 2017; Imperatori, 2017). Human resource management (HRM) practices are the custodian of employee well-being in modern organisations (Viot & Laila, 2019); thus, their role in preventing and managing stress is of paramount importance. To this end, the mutual gains HRM framework offers a win-win situation by suggesting that HRM practices should be enacted to improve employee performance and to enhance their well-being simultaneously (Guest, 2017; Kochan & Osterman, 1994).
Figure 1 shows a weak and a strong mutual gains model presented by Peccei et al. (2013, p. 22). The weak model posits that HRM makes a simultaneous win-win contribution to employee well-being and performance; the strong mutual gains model suggests that HRM practices impact performance through mediating influence of employee well-being.
Mutual Gains Conceptual Models.
Contrary to the assertions of mutual gains model, HRM research has historically demonstrated a singular focus on improving employee performance (Beer et al., 2015; Guest, 2017; Wood, 2021). Moreover, the sparse research on the topic indicates that the impact of HRM practices on workforce stress has produced mixed and inconclusive evidence. Some research shows that HRM practices are linked with lower stress levels (Boxall & Macky, 2014; Macky & Boxall, 2008), while others indicate that HRM practices improve firm productivity at the cost of increasing workforce stress (Kroon et al., 2009; Ogbonnaya & Messersmith, 2018; van de Voorde et al., 2012).
Further studies are required to examine in what way HRM practices can simultaneously and positively impact employee well-being and performance (Peccei & van de Voorde, 2019; Stankevičiūtė & Savanevičienė, 2018, 2019). Guest (2017) suggests that in developing a framework where HRM can improve well-being, it must be considered that ‘organizations are unlikely to promote well-being on ethical grounds alone’; therefore, it is ‘necessary to accommodate the interests of both employees and employers’ (Guest, 2017, p. 28). Following this, the present study attempts to develop a mutual gains framework which can benefit the employer and the employee. For this purpose, the study deploys Model 1(a) by Peccei et al. (2013).
To examine the effects of HRM on employee well-being, this study deploys the construct of perceived stress (PS; Cohen et al., 1983). PS is based on the transactional or appraisal theory of stress (Lazarus, 1966; Lazarus & Folkman, 1984). PS construct suggests that stress becomes toxic: (a) when people appraise or perceive events as threatening/demanding/stressful; and (b) when they feel they have insufficient physical/emotional resources available to deal with such events (Cohen, 2014). Measures of PS has been deployed earlier in HRM research to examine employee well-being (Gbadamosi & Ross, 2012; Ogbonnaya & Messersmith, 2018).
The second component of mutual gains model requires that HRM practices simultaneously impact employee performance in a positive manner. The present study deploys the construct of engagement as a proximal performance indicator. HRM practices do not impact performance directly, instead, they have a direct influence on employees’ proximal performance outcomes (i.e., employee attitudes and behaviours) (Dyer & Reeves, 1995; Guest, 2017; Wood, 2021). Employees with increased engagement levels invest their maximum cognitive, emotional and physical resources into work, harnessing their total selves to perform their duties (Kahn, 1990). Researchers have often deployed engagement as a proximal employee performance measure while examining its relationship with HRM practices (Albrecht et al., 2015; Saks, 2021). This is because engagement is a more firmly conceptualised and direct predictor of employee effectiveness than organisational commitment, job involvement and job satisfaction (Mackay et al., 2017). The sample organisations chosen for this study also conduct employee engagement surveys as a proximal performance measure.
Scholars agree that there is a continuing need to develop the theory on intermediary mechanisms which link HRM to employees’ psychological well-being and performance (Boon et al., 2019; Boxall et al., 2016; Guest, 2017; Wood, 2021). This study examined the impact of HRM perceptions on employee stress and HRM perceptions on employee engagement. Moreover, gratitude was examined as a novel mediator variable which links HRM practices to both employee stress and engagement at the same time.
Literature Review
A detailed review of the constructs deployed in the study and their proposed relationship is as follows.
HRM Attribution Approach
HRM research has mostly been conducted by selecting the best HRM practices (or HRM content) and examining their impact on various outcomes (Boon et al., 2019). This content-based approach is not robust enough to explain the impact of HRM practices on employee well-being or performance (Ostroff & Bowen, 2016; Sanders et al., 2014). An alternative, process-based approach focuses on the psychological processes through which employees develop an understanding of what HRM practices mean to them (Sanders et al., 2014). The meaning of HRM practices is subjectively interpreted by employees, and it is these subjective interpretations (not the content) that ultimately impact employee well-being and performance (Nishii et al., 2008). Thus, the process-based approach suggests that employee perceptions of HRM practices reveal a more substantial connection with various performance and well-being outcomes (Sanders & Yang, 2015; Wang et al., 2020).
To examine employee interpretations of HRM practices, this study deploys the construct of HRM attributions. Attribution theory postulates that a social interaction can be divided among initiators of behaviour (i.e., actors) and interpreters of behaviour (i.e., observers) (Heider, 1958; Kelley, 1967). The observers act as amateur psychologists and attempt to interpret the actor’s intent or motive behind the action or ‘why the actor has behaved in a certain manner’ (Heider, 1958; Kelley, 1967). Similarly, Nishii et al. (2008) postulated that the organisation (i.e., the actor) initiates a behaviour (through enacting HRM practices), whereas employees (i.e., the observers) develop interpretations of management’s intent, that is, ‘why management has enacted these HRM practices’; these interpretations are known as HRM attributions (Nishii et al., 2008). In line with the attribution theory, these interpretations have a significant impact on employee attitudes, emotions and behaviours.
Benevolent HRM Attributions and Employee Stress
Benevolent HRM attributions ‘reflect a belief that an HR practice was enacted to improve employees’ well-being’ (Fehr et al., 2017, p. 369). This is based on Nishii et al. (2008), who suggested that while interpreting ‘why their management has enacted certain HRM practices’, employees can attribute manipulative or benevolent intent to their organisation. Manipulative attributions include employee beliefs that HRM practices are designed to extract more work out of them, while benevolent HRM attributions reflect employee beliefs that HRM practices were designed by their organisation to help enhance their well-being. In line with Nishii and colleagues, this study considers ‘well-being HRM attributions’ (or WHRA) as the first set of benevolent HRM attributions.
This study suggests that the concept of benevolent attributions can be expanded to include employee beliefs regarding supportive HRM practices. Organisational benevolence can be reflected in many different forms, and supporting employee performance can be one of its manifestations. For instance, ‘benevolent leaders are those who create observable benefits, actions, or results for the common good’ (Karakas & Sarigollu, 2012, p. 537). Moreover, perceptions of benevolent leadership are cemented among employees when they believe their leaders display spiritual depth, ethical sensitivity, community responsiveness and engagement (Karakas & Sarigollu, 2012). Benevolence is characterised by kindness, compassion and a goodwill for other people (Brandt, 1976; Koutsouvilis, 1976), which is shown and received in many forms for example, acts of kindness, generosity, friendship and many other generally beneficial acts (Beauchamp, 2008; Csikszentmihalyi, 2019). Since there is a range of kind acts which can be perceived as a benefactor’s benevolence, this study proposes that HRM practices which are perceived by employees as helpful and supportive of their performance will also be seen as reflective of the organisation’s benevolent intent towards them.
Shantz et al. (2016) have already explored the idea of what they call ‘performance HRM attributions’ (or PHRA). They define PHRA as employee attributions of ‘HRM practices as primarily intended to support their job performance’ (Shantz et al., 2016, p. 173). Shantz et al. (2016) further suggest that PHRA is interpreted by employees as an indication of positive organisational intent behind developing and implementing HRM practices. ‘HRM performance attributions signal to employees that they are important and valuable’ (Shantz et al., 2016, p. 176). This signal can make employees feel that their organisation is supportive, provides them with job resources and emotional resources, and views their contributions as meaningful. In short, PHRA make employees feel supported, cared for and think that they are given ample resources to aid their performance (Shantz et al., 2016). In conclusion, PHRA can make employees believe that their organisation is a benevolent benefactor for choosing to enact HRM practices which support them, care for them and are a manifestation of a generous intent towards them.
The present study postulates that both PHRA and WHRA can have a significant impact on reducing employees’ PS. An explanation for these effects and their theoretical basis is as under.
PHRA have been shown to reduce emotional exhaustion among employees (Shantz et al., 2016). PS is a subjective and global assessment of stress which arises when people perceive that their stressor is more demanding than the resources they possess to manage the stressor (Lazarus & Folkman, 1984). PHRA can lower PS levels among employees by bolstering the perception of job resources relative to the job demands they experience. The effect of PHRA on lowering PS can be explained by the job demands-resources (JD-R) model (Demerouti et al., 2001).
Job demands require sustained physical, psychological (cognitive and emotional) and social effort required to do the job, whereas job resources are the physical, psychological and social support employees are given to help achieve their work goals and stimulate their personal growth (Bakker & Demerouti, 2007). Studies based on JD-R demonstrate that managerial policies which are seen as enhancing job resources lower employee stress and improve psychological well-being (Bakker & de Vries, 2021; Bakker & Demerouti, 2016; Schaufeli, 2017). Since PHRA are based on employee perceptions that their management is benevolent, supportive of performance and provides them with sufficient job and emotional resources (Shantz et al., 2016), it is expected that such attributions will result in lower PS levels among employees.
WHRA have been shown to reduce job strain and improve satisfaction among employees (Nishii et al., 2008; van de Voorde & Beijer, 2015). The impact of WHRA on lowering stress can be explained by perceived organisational support (POS) theory. POS reflects employees’ beliefs about the extent to which their organisation values their contribution and is invested in their well-being (Eisenberger et al., 1986). POS is an ‘experience-based attribution concerning the benevolent or malevolent intent of the organisation’s policies, norms, procedures and actions as they affect employees’ (Eisenberger et al., 2001, p. 42). POS is seen by employee as a form of employer benevolence or organisational altruism (Viot & Laila, 2019), and such perceptions of employer benevolence can improve employee well-being and reduce their stress levels (Eisenberger et al., 2020). Since WHRA is based on employee attributions that their management is benevolent and cares about their well-being, it is expected that such attributions will result in lower PS levels among employees.
H1: PHRA have a significant negative effect on PS levels of employees.
H2: WHRA have a significant negative effect on the PS levels of employees.
Benevolent HRM Attributions and Engagement.
Engagement is considered a motivational state that makes employees deploy positive attitudes, behaviours and high energy in their work (Macey & Schneider, 2008; Saks & Gruman, 2014). Employee engagement can be developed and optimised through HRM practices (Rana, 2015; Saks, 2021). HRM practices can enhance engagement through various mechanisms: first when employees perceive that HRM practices help them do their job and are supportive of their performance (or PHRA attributions); and second, when employees perceive HRM practices help enhance their well-being (WHRA attributions).
JD-R can help explain how PHRA can improve employee engagement. High job demands lead to a depletion of mental and physical energy, which is detrimental to employee well-being and engagement; alternatively, job resources result in low stress, stimulate employee engagement and motivate extra role performance (Bakker & Schaufeli, 2008; Schaufeli, 2017). HRM practices can improve engagement levels when employees perceive that such practices enhance their job-level resources (Cooke et al., 2016; van de Voorde et al., 2016). Since PHRA reflect employee beliefs that management provides them support and resources to perform well (Shantz et al., 2016), it is expected that such attributions will also lead to higher levels of engagement.
POS can explain how WHRA improves employee engagement. When employees experience lack of support in their work and feel devalued or taken advantage of, they withdraw from their work roles (Kahn, 1990). Alternatively, when people perceive that their organisation values them and cares about their well-being, they become engaged by deploying their maximum cognitive, emotional and physical energies at work (Eisenberger et al., 2020; Rich et al., 2010; Saks, 2006). Since WHRA reflect employee beliefs that their employer is benevolent and invested in their well-being, it can be expected that such attributions can improve employee engagement.
H3: PHRA have a significant and positive relationship with engagement.
H4: WHRA have a significant and positive relationship with engagement.
Gratitude
Gratitude is referred to being thankful, and is characterised as an emotion, state or a moral action (McCullough et al., 2001). Attribution theory of emotions postulates that emotions are triggered (in the observer) by thoughts and beliefs about causality of action (by the actor) (Weiner, 1985, 2014). Gratitude is an attribution-dependent emotional state, which arises in an interpersonal exchange when a receiving party (i.e., the beneficiary) recognises receiving a valuable benefit from the other party (i.e., the benefactor) (McCullough et al., 2002). Such gratitude is called other-directed gratitude or benefit-triggered gratitude (Ma et al., 2017). Moreover, such gratitude is based on the sense of receiving underserved benefits (Emmons & McCullough, 2004), as it is experienced by people when they feel they have been at the receiving end of benevolence or kindness of a benefactor (Emmons et al., 2019).
Other-directed gratitude can arise from the mere perception that the benefactor intended to benefit the recipient, even if the ‘intended benefit’ did not actually take place (McCullough, 2002). This demonstrates that the receipt of the ‘real benefit’ is not required by the beneficiary. A mere perception that the benefactor is acting out of benevolent intent is sufficient for the emotion of gratitude to arise (McCullough et al., 2001). It then follows that in work settings, perceptions of the organisation as a benevolent benefactor can foster an emotion of gratitude among employees (Fehr et al., 2017). This effect occurs because employees can attribute anthropomorphic qualities to an organisation (Eisenberger & Stinglhamber, 2011) and can be seen by them as ruthless, exploitative, benevolent, etc. (Ashforth et al., 2018).
This study has identified two HRM attributions which can make employees believe that their employer has enacted HRM practices from a benevolent motive and intent. The organisation can be seen as benevolent when it is seen to enact HRM practices to help support employee performance (PHRA) (Shantz et al., 2016) or improve their well-being (WHRA) (Nishii et al., 2008). Therefore, it is logical to expect that both benevolent HRM attributions (PHRA and WHRA) will lead to improved levels of gratitude among employees.
H5: There is a positive relationship between PHRA and gratitude.
H6: There is a positive relationship between WHRA and gratitude.
Gratitude and Perceived Stress
Scholars have consistently posited that there is a positive association among gratitude and arousal of positive and joyous emotions and a negative association among gratitude and stress levels. Gratitude is a ‘universal tendency to respond positively to another’s benevolence’ (Emmons & Stern, 2013, p. 847). Gratitude arises in the beneficiary because of the recognition that he/she has been a recipient of someone’s kindness (McCullough et al., 2001). Such gratitude is experienced as a pleasant and positively valenced emotion (Fredrickson, 2004), which improves the beneficiary’s well-being and reduces stress levels (Portocarrero et al., 2020; Skrzelinska & Ferreira, 2020). More specifically, gratitude is linked with a reduction in PS in occupational settings (Lee et al., 2018; Valikhani et al., 2019). Researchers have suggested that the relationship between gratitude and stress in different organisational contexts needs to be further examined (Cortini et al., 2019; Di Fabio et al., 2017; Fehr et al., 2017). Therefore, this study proposes the following hypothesis.
H7: Gratitude has a significant negative effect on PS levels of employees.
Relationship Between Gratitude and Engagement
Perceptions of the organisation as a benevolent benefactor can foster an emotion of gratitude among employees (Fehr et al., 2017). The experience of gratitude has motivational and behavioural consequences, which makes grateful people reciprocate and act positively towards their benefactor (Ma et al., 2017; McCullough et al., 2001). Engagement is a reciprocal attitude and behaviour through which employees pay back their organisation for the care and support it gives them (Saks, 2006), by devoting their maximum cognitive, emotional and physical energies in the pursuit of organisational goals (Saks & Gruman, 2014). Thus, employee gratitude is proposed as in important antecedent of improving employee engagement at work (Schaufeli & Salanova, 2010). More specifically, ‘when employees attribute gratitude initiatives to benevolent motives, they are likely to respond with enthusiasm and engagement’ (Fehr et al., 2017, p. 369). To the best of author’s knowledge, this study aims to be the first to test this theoretical assumption in occupational settings. It is, therefore, expected that gratitude will make employees display higher levels of engagement with their work to benefit their organisation, for enacting HRM practices from benevolent motives.
H8: Gratitude has a significant relationship with the engagement level of employees.
Gratitude as a Mediator Between Benevolent HRM Attributions, Perceived Stress and Engagement
Hewett et al. (2018) noted that only three studies have empirically examined mediating variables which connect HRM attributions with various employee performance and well-being outcomes. With such scant literature on the subject, it is recommended that mediating mechanisms be examined while studying the relationship between HRM attributions and various outcomes. This study has chosen to examine two benevolent HRM attributions, that is, PHRA and WHRA. These attributions are formulated by employees while evaluating the intent and motivation of their management behind enacting HRM practices.
This study has presented the following hypotheses regarding PHRA and WHRA. First, both benevolent attributions reduce PS levels and improve employee engagement. Second, PHRA and WHRA are positively linked with employee gratitude. Third, employee gratitude is negatively linked with PS and positively associated with engagement. It is, therefore, expected that gratitude can mediate the relationship between both benevolent HRM attributions (PHRA and WHRA) and PS. Moreover, gratitude also mediates the relationship between both benevolent HRM attributions and engagement.
Gratitude’s mediating effect on the impact of benevolent HRM attributions (PHRA and WHRA) on lowering PS and the impact of PHRA and WHRA on improving employee engagement can explained by social exchange theory (SET) (Gouldner, 1960). SET describes how social transactions between an employer and employee evolve into mutually beneficial and rewarding relationships (Cropanzano & Mitchell, 2005). SET is based on the norm of reciprocity, which states that people (i.e., beneficiaries) have an inherent desire to respond positively to their benefactors and return their favours or acts of kindness (Gouldner, 1960). In order for a social exchange to arise, employees must perceive that their employers are kind and meet their emotional needs through policies which make them feel valued and cared for (Cropanzano et al., 2017; Cropanzano et al., 2001). A social exchange-based employment relationship improves employee well-being and makes them perform better, as a perception of being valued and cared for meets their socio-emotional needs (Cropanzano & Mitchell, 2005; Eisenberger et al., 2020).
Gratitude builds and reinforces social relationships between the benefactor and beneficiary (Algoe et al., 2008; Gouldner, 1960) and is instrumental in enacting a social exchange between the two parties (Blau, 1964). PHRA and WHRA signal to the employees certain ideas: the ideas that the organisation values their effort, helps them perform well and cares about their well-being. In line with SET, when HRM practices become an indication of organisational care and support for employees (a) it can engender gratitude (Cropanzano & Mitchell, 2005; Fehr et al., 2017) and (b) such gratitude can have a positive impact on their well-being, specifically reducing PS (Fehr et al., 2017; Lee et al., 2018; Valikhani et al., 2019). In line with SET, this gratitude can act as an intermediary in the relationship between PHRA/WHRA and employee PS levels.
H9: Gratitude mediates the relationship between PHRA and PS.
H10: Gratitude mediates the relationship between WHRA and PS.
SET can also help explain why grateful employees return the benevolence of their employer with enhanced levels of engagement. Both PHRA and WHRA engender gratitude; moreover, grateful people try to return favour of a much greater value than the initial favour received from the benefactor, even exceeding the basic requirements of reciprocity (Algoe et al., 2008; Ma et al., 2017). Engagement is one way in which employees pay back their employer for its benevolence (Saks, 2006). Summing up, (a) PHRA and WHRA stimulate employee gratitude because of seemingly supportive and caring policies; and (b) since gratitude engenders social exchanges, such gratitude can invoke a desire within employees to respond to their employer’s benevolence with improved levels of engagement. Hence, gratitude can play a mediating role between both benevolent HRM attributions and engagement (Figure 2).
Theoretical Framework.
H11: Gratitude mediates the relationship between PHRA and engagement.
H12: Gratitude mediates the relationship between WHRA and engagement.
Methodology
Primary data has been collected from the telecommunication sector in Pakistan. Mansoor et al. (2011) observed that intense competition in the telecommunication sector makes PkTelcos exert increased pressures on their employees to perform, which caused increased job demands and elevated levels of stress among their employees. Other studies have also demonstrated that employees in PkTelcos experience uncertain job prospects and excessive work demands, which lead to much higher incidence of employee stress in PkTelcos (Ahmad et al., 2012; Ramay et al., 2017). Therefore, this research is conducted on the telecom sector of Pakistan, which is one of the high stress-inducing employment sectors in the country (Khan & Imtiaz, 2015). Purposive sampling technique was deployed for the collection of data as the stress levels among the sales staff of telecommunication sector is considerable (Khan & Imtiaz, 2015; Ramay et al., 2017). Structural equation modelling (SEM) technique was proposed for the testing of hypothesised relationship. Mplus 7.0 was used for statistical estimations. Sample size of this study were 294 respondents. The data was collected through offline questionnaire. The questionnaire of this study was adapted in which PHRA was measured using a 5-item scale by Shantz et al. (2016). The scale measures employee interpretations of 5 HRM practices (training and development, selection processes, reward systems, performance appraisal and employee participation) on a 5-point Likert scale. WHRA were measured for the same 5 HRM practices on a scale developed by Nishii et al. (2008). Gratitude is measured by deploying the three-item Gratitude Adjective Checklist (GAC) by McCullough et al. (2002). GAC is a robust scale instrument which measures other-directed or benefit-triggered gratitude as a sum of three adjectives: grateful, thankful and appreciative. Engagement was measured by using the 5-item job engagement scale developed by Saks (2006). A 10-item PS scale (Cohen et al., 1983; Cohen & Williamson, 1988) was deployed to measure PS levels among employees.
Results
Covariance-based SEM (CB-SEM) technique was deployed for the testing of research model. The sample size of 294 respondents was found to be suitable for data analysis. Before conducting SEM analysis, data normality was determined. Furthermore, exploratory factor analysis, confirmatory factor analysis (CFA), discriminant validity and convergent validity were examined. After the affirmation of adequate results, path analysis was conducted.
Data Normality
Before testing model using SEM, it was necessary to determine data normality. In this regard, multivariate normality of the constructs was determined. Performance human resource attributions was denoted by PHRA, well-being human resource attributions by WBHRA, engagement by ENG, gratitude by GRTD and perceived occupational stress by PS. According to Khwaja and Zaman (2020), standard deviation (SD) should be between ±2, while skewness and kurtosis should be between ±2 and ±3, respectively. Descriptive statistics results are reported in Table 1 which depicts that kurtosis, skewness and SD outcomes are in the permissible range. Thus, normality issues are not prevailing in the data.
Data Normality (N = 294).
Measurement Model
Exploratory factor analysis (EFA) is essential once data normality is attained. Exploratory factor analysis indicate that the items have been loaded on their respective factors and there have been no cross loadings. EFA values must be greater than 0.4 and less than 1 (Tabassum et al., 2020). After the attainment of precise EFA pattern, it is mandatory to conduct CFA. CFA confirms that the established pattern is precise, and it is confirmed through model goodness of fit measures. According to Lowry and Gaskin (2014), factor loadings of CFA must be between 0.3 and 1. Table 2 illustrates validity, reliability and factor loadings of the measurement model, in which EFA is denoted by (δ), CFA by (λ), average variance extracted by (AVE) and composite reliability by (CR). The results indicate that both EFA and CFA values are in the permissible range. According to Bashir et al. (2020), composite reliability (CR) values should be greater than 0.7 and less than 1. Similarly, Fornell and Larcker (1981) argued that average AVE values should be between 0.4 and 1. CR and AVE values in Table 2 indicate that there have been no reliability and validity concerns. Model fit measures were also reported in which absolute fit measures—namely chi-square (χ2), degree of freedom (df), standardised root mean square residual (SRMR), adjusted goodness of fit index (AGFI), root mean square error of approximation (RMSEA) and goodness of fit index (GFI)—were determined. Khwaja and Zaman stated for affirming model fitness, χ2/df value must be between 1 and 5, GFI and AGFI values must be less than 0.95, RMSEA and SRMR values must be less than 1. Table 2 reports that χ2/df = 2.009, GFI = 0.869, AGFI = 0.839, RMSEA = 0.062 and SRMR = 0.075. Similarly, incremental fit indices were estimated through confirmatory fit index (CFI), Tucker–Lewis index (TLI) and incremental fit index (IFI). CFI = 0.903, IFI = 0.915 and TLI = 0.902 were reported, which are all in the permissible range (Lowry & Gaskin, 2014). Measurement model was therefore precise with no statistical significance concerns.
Measurement Model Outcomes: Reliability, Factor Loadings and Validity (N = 294).
Table 3 reports discriminant validity and multicollinearity results. The square of correlations between the factors were less than AVE. Moreover, there were significant standardised factor loadings for all items (Khwaja et al., 2019; Mahmood et al., 2019).
Multicollinearity and Discriminant Validity (N = 294).
Structural Model
Path analysis was conducted to determine the causal relationship among the constructs. There were 12 hypothesised relationships among the constructs. For the acceptance of hypothesis, significance value (p-value) must be less than .05 and t-stats must be greater than 1.96. Hypotheses results are reported in Table 4. H1 argued about the PHRA association with PS. H1 of the study has been accepted, as the path coefficients (β) has been 0.229, t-stats has been 3.881 and p-value has been .00. Similarly, H2 determined the effects of well-being HRM attributions (WBHRA) with PS. The hypothesis has been accepted, as β = 0.417, p-value = .00 and t-stats = 8.510. H3 examined PHRA impact on engagement (ENG) which emerged to be significance too, as β value = 0.254 and t-stats = 4.096. H4 examined the effects of WBHRA with ENG, and the results revealed path coefficients value to be 0.277 and t-stats as 4.540.
Results of Hypotheses.
H5 proposed the relationship between PHRA with gratitude (GRTD); and upon testing, it was specified that effect size = 0.230, p-value = .00 and t-stats = 3.833. H6 conjectured the significant relationship among WBHRA and GRTD, and the results indicate β value = 0.327 and t-value = 6.727. H7 argued about positive and significant relationship between GRTD and PS, and the results reflect β value = 0.114 and t-stats = 2.035. H8 stated significant and positive relationship between GRTD and ENG, and the outcomes illustrate path coefficients to be 0.132, p-value to be .00 and t-stats to be 2.163.
H9 of the study argued about the mediating role of GRTD among PHRA and PS. The results highlighted that the beta value (effect size) with mediator emerged to be 0.223, t-value was 4.288 and sig value was 0.00. Similarly, H10 elucidated that GRTD mediates between WBHRA and PS, and the results indicated effect size of 0.389 and t-value of 7.701. Thus, the hypothesis was accepted. H11 stated the mediating effect of GRTD among PHRA and ENG, and the results portrayed beta value of 0.271 and t-value of 4.442. H12 proposed that GRTD mediates among WBHRA and ENG, and the results depicted beta value of 0.284, p-value of .000 and t-value of 4.732. Hence, all the hypotheses of the study were accepted, signifying strongly knitted theoretical associations.
Discussion
This research has made various theoretical, practical and contextual contributions to the literature.
Scholars are increasingly calling for studies that demonstrate how HRM practices can help alleviate employee stress and improve their overall well-being (Kowalski & Loretto, 2017; Ogbonnaya & Messersmith, 2019). Employers are more likely to prioritise employee well-being if research can be done through a mutual gains framework, that is, where HRM practices can be enacted to improve employee performance and enhance their well-being simultaneously (Guest, 2017; Kochan & Osterman, 1994). The present research fills this theoretical gap by developing a novel mutual gains frameworks based on the model suggested by Peccei et al. (2013).
Process HRM approach suggests that employee attributions and interpretations of HRM practices have a direct impact on their performance and well-being outcomes (Nishii et al., 2008; Sanders & Yang, 2015). The present study was based on attribution theory (Heider, 1958; Kelley, 1967), HRM attributions framework (Nishii et al., 2008) and benevolent HRM attributions (Fehr et al., 2017). Based on previous theoretical assertations, this study postulated that employees could perceive that their organisation enacts HRM practices with a benevolent motive and intent; such beliefs are called benevolent HRM attributions (Fehr et al., 2017). The present study demonstrated that two such benevolent HRM attributions include employee perceptions that HRM practices are supportive of their performance (PHRA) and employee perceptions that HRM practices are enacted in part to help improve their well-being (WHRA). The novel mutual gains framework developed in the present study shows that both PHRA and WHRA lower employee stress and improve their engagement levels.
The impact of PHRA and WHRA on PS and engagement can be explained through JD-Rs model and POS framework. PHRA makes employees believe that their organisation supports their performance and provides them the resources to do their job well. Such perceptions of improving job resources are theorised to improve well-being and employee engagement (Bakker & de Vries, 2021; Bakker & Demerouti, 2016; Schaufeli, 2017).
WHRA–well-being link is relatively simpler to explain. WHRA are based on employee beliefs that their employer is a benevolent benefactor who cares about their well-being, and such attributions are postulated to have positive effects on their cognitions, emotions and overall well-being (Nishii et al., 2008; Shantz et al., 2016). However, WHRA–engagement link is even more interesting in some ways. HRM practices perceived by employees as ‘efficiency oriented’ or designed to extract maximum work (Nishii et al., 2008) are harmful to employee performance and well-being. Conversely, HRM practices seen by employees as ‘well-being oriented’ improve employee engagement. This shows that when employers are seen by employees as caring and mindful of their well-being, they devote extra energy and effort to reciprocate their employer’s kindness (Eisenberger et al., 1986; Eisenberger et al., 2020). Thus, employers can benefit with improved employee engagement not just by enacting state of the art HRM practices (Sanders & Yang, 2015) but by making sure that these practices are ‘visibly seen’ by employees as caring and well-being oriented.
Since the link between HRM and various outcomes is not direct (Boxall et al., 2016; Guest, 2017; Hewett et al., 2018; Wood, 2021), gratitude was chosen as an intermediary variable to examine the relationship between PHRA/WHRA and PS and PHRA/WHRA and engagement. Both benevolent HRM attributions (PHRA and WHRA) are grounded in employee beliefs that their employer is a supportive and caring benefactor. Other-directed gratitude is a state of thankfulness which arises in the beneficiary as a result of the kindness or benevolence of a benefactor (Emmons, 2004).
The present study made another theoretical contribution by demonstrating that PHRA and WHRA arouse employee gratitude, which in turn lowers employees’ PS and improves their engagement by acting as a mediator. Studies on gratitude in organisational psychology are very rare (Cortini et al., 2019). Examining how gratitude arises in the workplace and can become an essential feature of the organisation can be extremely beneficial to employee well-being and performance (Fehr et al., 2017). While gratitude can be fostered by HRM practices, such studies are non-existent in literature (Cortini et al., 2019). Gratitude is instrumental for healthy organisations and research needs to examine how it can improve employee success, productivity and well-being (Di Fabio et al., 2017). This study is the first empirical examination of its kind which has shown the conditions in which gratitude arises at work and how such gratitude can improve employee well-being and engagement simultaneously.
Implications for Practitioners
The present findings can also be of use to the industry practitioners. This research has demonstrated that promoting gratitude through benevolent HRM attributions lowers employee stress levels and improves their engagement. This provides a realistic, win-win, mutual gains solution which can be implemented by HRM practitioners. To engender gratitude among employees, HRM departments can convey the idea that HRM practices have been enacted to support and improve their performance and enhance their well-being. To bolster the perception that the management supports employee performance and cares about their well-being, HRM practitioners can deploy a comprehensive employer branding strategy. Employer branding is a set of intra-firm communication policies designed to convey the perception that employee contributions are valued and the organisation is invested in their well-being (Kryger Aggerholm et al., 2011). Such intra-firm communications can invoke gratitude among employees, which can further improve employee engagement and well-being.
Research in gratitude is advantageous because people can be taught to be grateful. A considerable amount of literature has shown that gratitude can be invoked, cultivated and nurtured among people through simple exercises and activities (Ma et al., 2017; Skrzelinska & Ferreira, 2020). In the same vein, gratitude interventions at work can enhance ‘the expression, recognition, and perception of gratitude in the workplace’ (Cortini et al., 2019, p. 8). Considering the win-win situation gratitude can create for both employer and employee, gratitude needs to be embedded within the identity and culture of an organisation (Fehr et al., 2017). Some activities and interventions through which gratitude is enhanced among people is counting one’s blessings in life (or at work), making gratitude lists or diaries, expressing gratitude and focusing on a benefactor’s benevolence (Skrzelinska & Ferreira, 2020).
Sustainable HRM practices (Stankevičiūtė & Savanevičienė, 2018), well-being oriented HRM practices, comprehensive employee wellness programmes (Ongori & Agolla, 2008) and stress management techniques (SMTs) (Kröll et al., 2017) can improve employees’ job performance and psychological health. These programmes are beneficial in improving employee well-being in two ways. First, because of the inherent benefits they offer vis-à-vis improving employee well-being. Second, they can also bolster the perception that the organisation is a benevolent benefactor who cares about employee well-being. Such gratitude can further enhance the benefits of the programme and have a positive impact on employee well-being and performance.
In short, the present study provides a sound theoretical framework to achieve a win-win situation for both the employer and employee, as postulated by the mutual gains model (Guest, 2017; Kochan & Osterman, 1994; Peccei et al., 2013). The employees benefit through a reduction in their stress levels, which can contribute to an improvement in their overall well-being. On the other hand, employers benefit through an improvement in employee engagement, which has shown to be a robust predictor of employee performance (Mackay et al., 2017). A meta-analysis of 339 studies, based on 230 organisations and 82,248 business units across 49 industries showed that employee engagement was a robust predictor of firm performance.
Limitations and Future Recommendations
The study employed cross-sectional methodology to establish a relationship between different constructs. Future studies can deploy a longitudinal research methodology with mixed-method research design. This study was conducted in the mobile phone telecommunication sector of Pakistan because of the high stress inducing environment in which employees operate. Choosing a single industry allows for a closer comparison and controls for between-industry differences. However, choosing a single industry may restrict the generalizability of the study across sectors. The respective study has scrutinised two internal benevolent HRM attributions based on the framework developed by Shantz et al. (2016) and Nishii et al. (2008). Future studies may explore other benevolent attributions that employees can make regarding their employer or management, for example, benevolent leadership (Karakas & Sarigollu, 2010) and caring HRM practices (Saks, 2021). Further studies can examine dispositional gratitude among employees, which is the general tendency/predilection of people to experience gratitude and repay the benevolence of a benefactor (McCullough et al., 2002).
Conclusion
This study demonstrated that gratitude arising from a perceived benevolence of HRM practices can secure mutual gains for the employer and the employee. In this vein, this study showed that employee gratitude reduces employees’ PS levels. This is in line with prior findings on the relationship between gratitude and PS (Lee et al., 2018; Valikhani et al., 2019). This effect is grounded in the broaden and build theory of positive emotions (Fredrickson, 2004). This theory states that gratitude is a positive emotion which enhances people’s well-being and reduces stress through bolstering their resilience and enhancing personal resources (Fredrickson, 2004). Second, this study demonstrated that gratitude enhances employee engagement levels. This effect occurs because the experience of gratitude has motivational and behavioural consequences, which makes grateful people reciprocate and act positively towards their benefactor (McCullough et al., 2001). Similarly, engagement is a reciprocal attitude and behaviour through which employees pay back their organisation for its benevolence, by devoting their maximum energies to work roles (Saks & Gruman, 2014; Saks, 2006).
A contextual contribution comes in the form of conducting this study in Pakistan. Research from South Asian countries is encouraged as it can add new perspectives to the advancement of HRM practices (Budhwar & Debrah, 2013). Research in HRM and organisational psychology is in its nascent stage in Pakistan, and it requires focus and theoretical development (Ali & Brandl, 2017). This study has opened new research avenues in HRM literature with a framework for reducing employee stress, which is built and empirically examined in a developing South Asian country.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
