Zooming into paternalistic leadership: evidence from high power distance culture

Purpose – Nurses ’ turnover intention has become a major issue in developing countries with high power distance cultures. Therefore, the authors attempt to investigate how turnover intention among nurses ’ can be reduced through paternalistic leadership (PL). The authors further investigate the mediating role of job satisfaction between the associations of benevolent, moral and authoritarian dimensions of PL with turnover intention. Finally, the authors examined perceived organizational support (POS) as a conditional variable between job satisfaction and turnover intention. Design/methodology/approach – Theauthorscollecteddatafrom374nursesworkinginpublicandprivate hospitals of high power distance culture using a questionnaire-based survey on convenience basis. Findings – Structural equation modeling confirms that benevolent and moral dimensions of PL positively affect nurses ’ job satisfaction which helps them reduce their turnover intention. While the authoritarian dimension of PL negatively affects job satisfaction to further enhance their turnover intention. In addition, the authors noted POS as a conditional variable to trigger the negative effect of job satisfaction on turnover intention. Research limitations/implications – The authors used a cross-sectional design to collect responses and ensured the absence of common method variance through Harman ’ s Single factor test. Originality/value – This study identified the mechanism (job satisfaction and POS) through which benevolent, moral and authoritative dimensions of PL predict turnover intention among nurses working in high power distance culture.


Introduction
Nowadays, Asian organizations are striving to compete because of technological and procedural revolutions (Bedi, 2020). The growing competition in Asian cultures has led managers and scholars to rethink the exclusive leadership styles that help to satisfy and retain human resources (Oh and Oh, 2017). Leadership is regarded as a strong predictor of employee work-related outcomes, thereby prevails a need to ponder upon effective leadership styles Ahmad et al., 2021a). Amongst many, one of the least examined leadership styles is paternalistic leadership (PL) style (Rawat and Lyndon, 2016), which originated from Chinese ancient Confucianism ideology (Farh and Cheng, 2000) and identified as an effective leadership style in Asian cultures (Nazir et al., 2021).
PL style has recently gained researchers' attention and is still considered a new leadership style (Lau et al., 2019). This is the reason that studies have used numerous terms to define PL such as uneven power relations, unquestionable obedience and feeling of personal loyalty towards leaders (Weber, 1968). However, the latter studies opposed the authoritarian view of PL and considered it a three-dimensional construct, i.e. benevolence, morality and authoritarianism (Cheng et al., 2004;Farh and Cheng, 2000). This is because paternalism is inferred as managers' sincere and personal interest in employees' off-the-job lives to enhance their welfare (Chen et al., 2014). Paternalistic leaders are "benevolent" as they sincerely care for their subordinates' personal and professional well-being, "moral" as they exercise high personal virtues and exemplify themselves as a role model for their followers and "authoritarian" because of enforced discipline and commanding compliance from their followers on their decisions (Farh and Cheng, 2000;Cheng et al., 2004). Recent studies have identified that paternalistic leaders positively shape their followers' job-related outcomes such as job satisfaction, reduced intent to leave, creativity, self-efficacy, organizational citizenship behavior and organizational commitment (Pellegrini et al., 2010;Hongyu et al., 2012;Bedi, 2020;Nazir et al., 2021;Hawass, 2017); however, there is still need to generalize these findings in various cultures.
Paternalistic leadership (PL) style is perceived differently by western and non-western cultures. For example, Western cultures perceive paternalistic leaders as "benevolent dictators" involved in "non-coercive abuse"; whereas, non-western cultures perceive such leaders as positive figures (Pellegrini and Scandura, 2008). This is because PL is associated with highpower distance and distinctive authority between superiors and subordinates (Bedi, 2020), which embraces the norms of non-western cultures but are contradictory to the western individualistic and fair treatment beliefs (Aycan et al., 2013). Researchers have identified diverse impacts of the dimensions of PL on followers' job outcomes as benevolence and morality dimensions positively, whereas, authoritarian dimension negatively affect job-related outcomes (Nazir et al., 2021;Hawass, 2017); nevertheless, few of the studies have identified its insignificant association with job outcomes (Erben and Guneser, 2008;Nnaemeka and Onebunne, 2017). These inconsistent empirical findings are not only because of cultural differences (Lau et al., 2019) but also because of lack of sufficient construct in the knowledge repository (Wu and Xu, 2012). According to Lau et al. (2019), PL should have been studied as a three-dimensional construct as its dimensions have diverse impacts on employee outcomes.
In a recent meta-analysis, Bedi (2020) remained unable "to identify the relative importance of different dimensions of paternalistic leadership in predicting followers' outcomes (p. 998)" such as job satisfaction and turnover intention, especially in high-power distance cultures (Soomro et al., 2019). In addition, we knew little about such mechanisms and their interactional effects because of the limited empirical studies on PL (Bedi, 2020). Following these milieus, we focused on how moral, benevolent and authoritarian dimensions of PL impact employees' job satisfaction and turnover intention in Pakistan, which has a highpower distance culture (Ahmad Bodla et al., 2019;Nazir et al., 2021). Specifically, drawing upon Blau's (1964) social exchange (SET) and Sidanius et al.'s (2000) social dominance theory (SDT), we develop a model that examines job satisfaction as a mediator between PL and turnover intention and perceived organizational support (POS) as a conditional variable on the association between job satisfaction and turnover intention (see Figure 1).
According to SET, human relations are based on norms of reciprocity (Gouldner, 1960) that when a party receives something valuable from another party, it tries to reciprocate with similar or greater value (Islam et al., 2016). Suffice to say that when employees perceive their leaders in the best of their interest, in turn, they show greater confidence, loyalty (Pelligrini and Scandura, 2008) and prefer to be with their organization (Bedi, 2020). On the other hand, SDT suggests that all existing human societies are built upon group-based social hierarchies that provide one-step up status to one group to dominate over the other group (Sidanius et al., 2000). These social hierarchies underlie divisions based on age, gender and arbitrary systems (Bedi, 2020). A group might be dominating another group based on the age system; where adults are given importance and high authority over children and based on a gender system where males have more power and authority over females. In addition to this, a group might dominate over the other based on socially built classes, as formed by arbitrary set systems, such as race, ethnicity and social class (Sidanius and Pratto, 2001). Leaders enjoy dominance and power of decision-making over their subordinates due to one step-up authority provided to them (Cheng et al., 2004). The followers obey their leaders' decisions willingly because of social hierarchical systems and power assigned to leaders (Aycan, 2006).

Hypotheses development
Relationships among benevolent leadership, job satisfaction and turnover intention Benevolent leadership (BL) is that dimension of paternalistic leadership in which leaders demonstrate individualization and show their sincere concern for their followers' professional and personal well-being (Farh and Cheng, 2000). Benevolent leadership shapes employees' positive work attitudes which contribute to organizational overall effectiveness. According to Chinese ancient Confucian philosophy, benevolent leaders not only support and care for their followers (Farh and Cheng, 2000) but also exercise authority so that followers get to know who their boss is (King and Bond, 1985). Moreover, benevolent leaders assist as mentors (Wang and Cheng, 2010) and help subordinates when arrested in personal and/or familial crises (Farh et al., 2008). This uniqueness not only shapes followers' positive attitudes (job satisfaction and reduced turnover intention) but also creates an ongoing exchange relationship between benevolent leaders and their subordinates (Hwang, 2008).
According to SET, the relationships between two parties are based on norms of reciprocity where the receiving party is obliged to pay back with similar attitudes to the giving party (Blau, 1964). Therefore, when employees perceive their leaders as caring and supportive (benevolent leadership) they not only show satisfaction with their job but also are willing to work with the same organization. This is because benevolent leaders always work for the betterment of their followers' well-being (Nazir et al., 2021) which creates a positive workplace environment. Observing the kind concerns of their leaders, employees engender feelings of obedience, trust and loyalty (Farh et al., 2008), which become the basis for their job satisfaction and reduced turnover intention (Chen and Kao, 2009). Thus, we hypothesized: H1. Benevolent leadership positively influences employees' job satisfaction (a) and negatively influences employees' turnover intention (b).
Relationships among moral leadership, job satisfaction and turnover intention Due to tremors stemming from moral and ethical scandals in organizations, moral leadership has gained inimitable significance in leadership research (Bedi et al., 2015). Morality is regarded as an effective means of influencing followers (Yukl, 2010). Leaders' morality is considered as a part of many leadership conceptualizations such as transformational (Burns, 1978), servant (Greenleaf, 1977) and authentic (Avolio and Gardener, 2005); however, in this study, we conceptualized moral leadership as a dimension of paternalistic leadership. Moral leadership includes leaders' honest, virtuous, self-disciplined and unselfish behavior towards the followers (Erben and Guneser, 2008). Few of the researchers have argued that the moral dimension of PL overlaps the concept of ethical leadership. In actuality, both the leadership styles are different as ethical leadership focuses on two-way communication between leader and follower where both managers and subordinates are involved in decision making (Brown and Trevino, 2006). Whereas, the Chinese conceptualization of the moral dimension of PL emphasizes one-way communication between leader and subordinate where leaders make decisions and followers are expected to obey (Chen et al., 2011). Moral leaders attempt to create a positive work environment through exercising integrity and superior moral values (Bao and Li, 2019), which subordinates reciprocate with positive workplace attitudes (Islam et al., 2021c;Ahmad et al., 2021a). More specifically, literature is well documented about the associations of moral leadership with job satisfaction and turnover intention (Farh et al., 2006;Liang et al., 2007). Moral leaders reveal superior personal values and utilize their status to act in subordinates' collective interest (Cheng et al., 2004). Indeed, integrity and helping subordinates beyond self-interest are the two crucial characteristics of moral leaders. Exhibiting integrity, moral leaders avoid compromise and undue favors (Bedi, 2020), while through helping subordinates, they enhance their identification and trust , which creates a positive work environment. Subordinates, when they perceive their leader as a moral, try to follow them and show greater satisfaction (Deng and Chen, 2013) and less turnover intention (Nnaemeka and Onebunne, 2017). These arguments are in line with Blau's (1964) SET that when individuals perceive their leaders avoiding compromises and giving undue favors they try to reciprocate with greater satisfaction and staying with their organization (reduced turnover intention). Thus, we hypothesized: H2. Moral leadership positively influences employees' job satisfaction (a) and negatively influences employees' turnover intention (b).

Relationships among authoritarian leadership, job satisfaction and turnover intention
Being a significant dimension of PL style, the roots of authoritarian leadership (AL) originated from the study of enterprises in the 1970s (Cheng et al., 2000). Most of the researchers have examined AL as an independent leadership style (Kiazad et al., 2010). Authoritarian leadership is referred to as leaders' act of exercising complete control over subordinates and demand; in return, receive strict and unobjectionable obedience from them (Cheng et al., 2004). This act shows a clear superior-inferior relationship between leaders and followers (Zhang et al., 2015). Authoritarian leaders conceal important information from their followers because they do not want them to participate in decision-making, hence, such leaders prefer to be commanding to reprimand subordinates' poor performances (Pellegrini and Scandura, 2008). AL is broadly practiced in various Asian, Western and Middle Eastern countries (Harms et al., 2018;Siddique and Siddique, 2019). Specifically, this leadership style is more prevalent in collectivistic and high-power distance cultures (Chan et al., 2013). A plethora of literature has argued about the negative impacts of AL on employees' attitudes that have detrimental effects on their task performance (Chen et al., 2014). This is because leaders scoring high on the AL dimension tend to retain power asymmetry (Tsui et al., 2004) and often make unilateral decisions (Aryee et al., 2007). Such controlling behaviors from leaders are referred to as abuse of power for personal concerns (Schuh et al., 2013) as they least bother about their subordinates' well-being (Chan et al., 2013).
According to Bedi (2020), authoritarian leaders exercise their influence on subordinates mostly in two ways, i.e. personal dominance and negative social exchange. As leaders are high-level in the organizational hierarchy, therefore, they exert complete dominance in decision-making (Carson, 1969). Such controlling behaviors by the leaders are felt oppressive by the subordinates resulting in an abusive workplace environment. From a social exchange perspective, when employees perceive their leaders as dominant and controlling, they in response, show more negative attitudes (Farh and Cheng, 2000). Blau's (1964) SET posits that employees negatively reciprocate when perceiving their leaders as having controlling behavior and demanding strict obedience to their decisions. Indeed, the receiver reciprocates negatively when receiving unfavorable treatment from another party (Gouldner, 1960). Given that, it can be assumed that when employees perceive their leaders as authoritarian, they show decreased job satisfaction and increased turnover intention. Thus, we hypothesized: H3. Authoritarian leadership negatively influences employees' job satisfaction (a) and positively influences employees' turnover intention (b).

Mediating role of job satisfaction
We examined job satisfaction as a mediating mechanism between PL dimensions and turnover intention to extend the literature. The association between employee job satisfaction and intent to quit the workplace is well-established in the literature and is supported by ample empirical evidence. Lambert et al. (2001) argued that the impact of job satisfaction on intent to leave the organization is natural, which has been confirmed by Griffeth et al. (2000). Griffeth et al.'s (2000) meta-analysis revealed an average correlation of about À0.19 between job satisfaction and turnover intention. The theoretical foundations regarding the satisfactionturnover relationship were derived from the work of March and Simon (1958). They proposed that job satisfaction is the primary antecedent of perceived desirability of movement, leading to employees' behavioral intent to leave the workplace (Lee et al., 1999). Indeed, the intention to quit the current job and to seek other employment opportunities in the market depends upon how much an employee is satisfied with his/her job (Jang and George, 2012). Being more specific, inferences can be drawn that higher the job satisfaction, the lower will be the intentions to quit and the lower the job satisfaction, the higher will be the intentions to quit . This inference can further be supported by SET that employees' level of job satisfaction depends upon their treatment by their organization (positive or negative leadership) which they reciprocate accordingly . Drawing upon SET, the aforementioned literature has confirmed that the dimensions of PL (i.e. moral, benevolent and authoritative) have varying impacts on followers' level of job satisfaction and intention to quit (see H1-H3). Similarly, job satisfaction has been identified as a predictor of turnover intention. Still, a recent meta-analysis by Bedi (2020) has identified less empirical evidence about the mechanism between three-dimensional PL with employees' attitudes (job satisfaction and turnover intention here). Specifically, we argue that when employees perceive their leaders as benevolence and morals show greater levels of job satisfaction which helps them reduce their intention to leave the organization. On the other hand, when employees perceive their leaders as authoritarian, they exhibit a low level of satisfaction with their job which enhances their intention to leave the organization. Thus, to examine these assumptions we hypothesize: H4. Job satisfaction negatively influences employees' turnover intention.
H5. Job satisfaction mediates the association of benevolent (a), moral (b) and authoritarian (c) leadership with employees' turnover intention.
The moderating role of perceived organizational support (POS) POS is employees' perception of being cared by their organization . According to Eisenberger et al. (1986), POS is an outcome of an exchange relationship between employees and their organizations because the extent of such relations demonstrates that employees' involvement is recognized and valued by their organization.
Employees' POS gets impacted by many factors such as money, credit, appreciation, recognition and encouragement (Eisenberger et al., 1986). According to , POS is usually fostered by two significant perceptions of employees; one is through personifying the organization and the other is through rewards as distributed by the Paternalistic leadership and turnover intention organization. Through personification, employees assign human characteristics to organizations and through rewards where employees perceive favored or unfavored treatment because of uncontrollable external organizational constraints (Eisenberger et al., 1997). In other words, it can be inferred that the aforementioned two beliefs assist to identify the resources made available by the organization for employee-organization social exchange relationships (Loi et al., 2006). POS is activated when there is a social exchange process between employee and organization through the norm of reciprocation which creates a sense of employee obligation (Galletta et al., 2011). This reciprocation might generate positive attitudinal and behavioral work-related outcomes such as high intent to stay with the organization and lower turnover (Tekleab et al., 2005;Mishra and Ghosh, 2020;Srivastava and Agrawal, 2020). Employees' positive feelings about their supervisors enhance their POS which creates a sense of obligation to reciprocate back to their organization (Jeung et al., 2017) and helps them to reduce their turnover intention (Eisenberger et al., 2002). This is because, employee consider their leaders/supervisors as the representative of their organization (Ahmad et al., 2021b) and their style help them reciprocating towards their organization (Islam et al., 2021d). According to Hur et al. (2015) "since employees in a collectivistic society are more likely to perceive themselves to be part of the organization, POS may function as an important confounding variable on the relationship between psychological state and job-related outcomes" (p. 607). Therefore, many researchers have examined POS as a moderator between employees' job-related attitudes. For example, Sher et al. (2019) identify POS buffering the effect of leader-member exchange on expectorates voice in Taiwan. Yoon et al. (2020) noted POS strengthening the association between psychological ownership and knowledge-creating in Korea. Similarly, Srivastava and Agarwal (2020) noted POS buffering the negative association between burnout and turnover intention, whereas Duke et al. (2009) noted it buffering the negative association between emotional labor and job satisfaction. However, literature is scant about how POS moderates the association between job satisfaction and turnover intention. Therefore, this study argues that employees' perception of support from their organization (in which leaders' style also contribute) may serve as a conditional variable that strengthens the negative association between job satisfaction and turnover intention. Suffice to say that, when employees feel satisfaction at their workplace, they are more likely to stay in the organization, in such situations their perception of support from their organization signals their psychological process to continue with their decision to be with their organization. Set also supports the same arguments that employees reciprocate their perceptions of being cared for through positive attitudes (Job satisfaction) which diminish their negative attitudes (turnover intention) (Blau, 1964). Thus, we hypothesize: H6. POS moderates the negative association between job satisfaction and employees' turnover intention such that employees with high POS may strengthen this relationship.

Participants and data collection procedure
Past studies have identified that PL is congruent with collectivist and high-power distance cultures (Pellegrini et al., 2010;Ahmad Bodla et al., 2019) as such cultures tend to have a high level of paternalism. Literature has suggested that the effect of PL in high-power distance countries would be greater than those of low-power distance countries (Bedi, 2020). Further, there is a need to study PL in both public and private sectors (Jing-Horng et al., 2015;Chine et al., 2006). Therefore, we selected nurses working in both public and private hospitals of Pakistan (a high-power distance country) because of several reasons. According to the statistics given by the United Nations (2020), a single nurse in Pakistan has to care for twenty-five patients and five doctors, which is at a higher cost as compared to other developing countries. The Pakistani government has reported a shortage of 60,000 nurses in the country . Studies have suggested workplace bullying , brain drain, less monetary incentives and high turnover  as the reasons behind this shortage. Amongst these, turnover is the biggest reason behind this shortage that healthcare management is facing (Shahzad and Malik, 2014). As, employees are regarded as the real assets of any organization ; hence, are essential to retain. Therefore, the current study has presented a model that may assist in retaining nurses through effective leadership styles (PL here). We used item-response theory to select a sample of 660 (Islam and Tariq, 2018). Specifically, we select the criteria of twenty respondents against each item of the questionnaire (33 3 20 5 660). We received 394 responses between June to December 2020 (60% of response rate). We visited various public and private hospitals and showed them the letter (issued by the institute of business administration, University of the Punjab) to allow us to collect data. After obtaining permission from the heads, we ensure nurses about the confidentiality of their responses.
Nurses were also asked about their demographic characteristics such as gender, age and qualification. Respondents were eventually found to be evenly distributed based on gender (Male: N 5 198, 50.5%, Female: N 5 194, 49.5%). In addition, the majority of the respondents were in the age bracket of 21-25 years (50.79%), holding a master's degree (45.49%).

Measures
All the responses were collected on a five-point Likert scale ranging from 1-Strongly Disagree to 5-Strongly Agree. All the scales to measure the constructs were adopted from previous studies with proven internal consistency reliabilities.
Paternalistic leadership. We used three dimensions of paternalistic leadership namely benevolent, moral and authoritarian leadership. The measurement scales for these dimensions were adopted from Cheng et al.'s (2004) study. These scales were used to evaluate the employees' perception of how often their leaders revealed each leadership behavior. This scale has been validated in the Asian context Dedahanov et al., 2016;Chaudhary et al., 2021). Benevolent leadership was assessed through six items with 0.78 as the value of its reliability. A sample item was "My supervisor ordinarily shows a kind concern for my comfort". The current study used five items of moral leadership and noted its reliability as 0.76. A sample item was "My supervisor set him/herself a good role model to follow". The authoritarian leadership dimension was assessed through six items with a reliability of 0.71. A sample item was "My supervisor always behaves in a commanding fashion in front of employees".
Turnover intention. We used Kuvaas's (2008) five items to measure employees' intention to leave their organization. The study noted 0.82 as the value of its reliability. A sample item was "I will probably look for a new job in the next year".
Job satisfaction. A three item scale of Cammann et al. (1983) was used to measure employees' level of job satisfaction and noted 0.80 as the value of its reliability. A sample item was "I am satisfied with my current job".

Paternalistic leadership and turnover intention
Perceived organizational support. We used Eisenberger et al.'s (1986) short eight item scale of POS and noted 0.85 as the value of its reliability. A sample item was "My organization cares about my well-being".

Results
Initially, we test the data for missing values, outliers and normality. Further, we conducted confirmatory factor analysis (CFA) and then hypotheses were tested by examining structural model and hierarchical regression. As the data was personally administered, therefore, it was free from missing values. We applied Mahalanobis test to identify 18 outliers (p < 0.000); therefore, further analyses were conducted on 374 responses. Following Byrne (2010), we examined the values of skewness (±1) and kurtosis (±3) for the normality which were well within the range. Multicollinearity was examined following the criteria of Tabachnick and Fidell (2007) that the correlational values among variables should be less than 0.85 (See Table 1). Moreover, we conducted CFA to examine the validity, reliability and unidimensionality of the variables under study (Byrne, 2010). We examined the values of "Chi-Square/Degree of Freedom (χ 2 /df ≤ 3.0), Comparative Fit Index (CFI ≥0.90), Root Mean Square Error of Approximation (RMSEA ≤0.08) and Goodness of fit index (GFI ≥0.90)" and measurement model was found to be fit, i.e. χ 2 /df 5 2.38, GFI 5 0.91, CFI 5 0.93, RMSEA 5 0.037. We followed Hair et al. (2010) for factor loading values (i.e. FL > 0.50); Bagozzi and Yi (1988) for composite reliability (i.e. CR ≥ 0.60) and average variance extracted (i.e. AVE ≥ 0.50); Fornell and Larcker (1981) to examine whether the values of AVE are greater than the values of MSV; and Cronbach (1951) for reliability (i.e. α ≥ 0.70). The values of CR and AVE were examined for convergent validity, MSV for discriminant validity, α for reliability. Table 1 confirm the reliability and validity of the scales used in this study. We further examined Harman's Single-factor to detect common method variance (CMV) and one factor was noted to explain 36.89% variance which is less than 50% (Podsakoff et al., 2012).

Descriptive and correlation analysis
The values of descriptive statistics (i.e. mean and standard deviation) and correlations among observed variables are shown in Table 1. The mean values of benevolent (M 5 3.81) and moral (M 5 3.57) dimensions show that respondents were agreed about the presence of such leadership, however, were neutral regarding authoritarian (M 5 2.98) dimension of PL. Further, we noted that benevolent and moral dimensions positively (r 5 0.38 and 0.34, p < 0.01), whereas, authoritarian dimension negatively correlate with job satisfaction (r 5 À0.31, p < 0.01). Similarly, benevolent and moral dimensions negatively (r 5 À0.26 and À0.23, p < 0.01), whereas, authoritarian dimension positively correlate with turnover intention (r 5 0.41, p < 0.01). Finally, we noticed that satisfied employees are less likely to have turnover intention (r 5 À0.43, p < 0.01).

Descriptive statistics and correlations
Paternalistic leadership and turnover intention influence of job satisfaction on turnover intention (β 5 À0.48, CR 5 3.91, p 5 0.00), which supports H4 of the study.
To test the mediation, we assessed the significance of indirect effects which were calculated by multiplying the beta coefficients of (Independent variable → Mediating variable) with (Mediating variable → Dependent variable) (Byrne, 2010). The study noted a significant indirect effect of benevolent leadership (β 5 À0.18, CR 5 3.11, p 5 0.002), moral leadership (β 5 À0.21, CR 5 4.21, p 5 0.00) and authoritative leadership (β 5 0.19, CR 5 3.58, p 5 0.001) on turnover intention via job satisfaction. These results are in line with H5a, H5b and H5c of the study and confirmed the explanatory role of job satisfaction (see Table 3).
Finally, we examined the moderating role of POS using two-step hierarchical regression. First of all, we computed an interactional term between Job satisfaction and POS (i.e. JS 3 POS). Then in the first step of regression, job satisfaction and POS were regressed with turnover intention and both were noted to have a negative and significant effect on turnover intention (β 5 À0.39 and À0.23, p < 0.01). In the second step, the interactional term (JS 3 POS) was regressed with turnover intention and its significant effect with 18% variance was noted (β 5 À0.44, p < 0.01). This confirms the moderating role of POS (see Table 4). Further, we conducted a slope of the moderation (see Figure 2 Step-1 Job satisfaction (JS) À0.39** Perceived organizational support (POS) À0.23** R 2 0.21 Step-2 JS 3 POS À0.44** R 2 0.39 ΔR 2 0.18 Note(s): **p < 0.01 Table 2. Results of structural model Table 3. Table 4. Hierarchical regression for moderation confirm that employees who perceive a high level of support from their organization are more likely to trigger the negative effect of job satisfaction with turnover intention (H6 is accepted).

Discussion
Most of the studies have examined paternalistic leadership (PL) as a construct; however, less has been focused on its dimensions, i.e. benevolent, moral and authoritative leadership (Bedi, 2020;Nazir et al., 2021). In addition, although PL shapes employees' positive job-related outcomes, still studies on its mechanisms are scant (Chaudhary et al., 2021;Lau et al., 2020). With these backdrops, we investigate how the dimensions of PL affect employees' turnover intention through job satisfaction. In addition, we examined perceived organizational support (POS) as a conditional variable on the association between job satisfaction and turnover intention. As PL style is more prevalent in high-power distance countries (Ahmad Bodla et al., 2019); therefore, we collected data from the nurses working in public and private hospitals of Pakistan.
The study noted that employees perceive their leaders' benevolent and moral styles as positive, in turn, show greater levels of satisfaction with their job which ultimately reduces their intention to leave the organization. Literature has also suggested that positive leadership enhances subordinates' positive (job satisfaction) and reduces negative (turnover intention) workplace attitudes (Islam et al., 2021c). For example, Lin and Liu (2017) conducted a study on employees working in Taiwan's banking sector and noted ethical leadership reduces their turnover intention. Ren and Chadee (2017) studied 388 professionals from eight organizations in Beijing, China and noted ethical/moral leadership enhances employees' job satisfaction. Similarly, studies have noted that benevolent leaders positively influence their subordinate's performance (Chan, 2017) andwell-being (Erkutlu andChafra, 2016). According to the findings of Bedi's (2020) meta-analysis on PL, employees show a greater level of job satisfaction and reduced turnover intention towards those leaders who care about their welfare. These findings are in line with SET that individuals reciprocate their positive leaders through positive attitudes (Blau, 1964).
On the contrary, we noted, when leaders show controlling behaviors and demand strict obedience from their subordinates (i.e. authoritative leadership), the latter tend to show more negative attitudes and behaviors such as decreased job satisfaction and increased intention to quit the workplace. These findings are in line with the findings of Schaubroeck et al. (2017) and Islam et al. (2020a) that employees respond to their abusive/authoritative leaders through greater job dissatisfaction and leave intention. Supervisors/leaders are at the higher level of organizational hierarchy and subordinates cannot directly counter them; therefore, they reciprocate to such leaders through negative attitudes (Blau, 1964) such as turnover intention and job dissatisfaction. We further noted job satisfaction explaining the associations of benevolent, moral and authoritarian dimensions of PL with turnover intention. Specifically, job satisfaction was found to negatively explain the associations of benevolent and moral dimensions with turnover intention; whereas positively explain the negative association between authoritarian dimension with turnover intention. This is because, when employees perceive the positive side of their leaders (benevolence and moral), reciprocate positively (i.e. enhanced job satisfaction and reduced turnover intention) and when perceive the negative side of their leaders (authoritarian) reciprocate negatively (decreased job satisfaction and greater turnover intention). These findings are in line with social exchange theory. Finally, we identified POS as a conditional variable that strengthens the negative association between job satisfaction and turnover intention. According to Hur et al. (2015), individuals in collectivist cultures prefer to be a part of their organization. POS creates a sense of ownership (Galletta et al., 2011) among employees that affect their psychological state. Therefore, POS helps employees to further trigger the negative association between job satisfaction and turnover intention.

Theoretical implications
The study has several theoretical implications. First, studies on paternalistic leadership have been conducted in non-Western and Western cultures (Koveshnikov et al., 2022;Chan, 2017) with varying results. House et al. (2014) noted that in the western culture paternalistic leadership (specifically authoritarian and benevolence dimensions) negative affect subordinates' attitudes. However, in high-power distance cultures (e.g. Russia and China) it positively affects subordinates' attitudes (Koveshnikov et al., 2020). Infect, studies have identified that authoritarian and benevolence styles differently affect the people of the same culture (Shen et al., 2019). This means that the dimensions of PL cannot be understood by means of societal-level differences; hence, studies need to pay more attention (Koveshnikov et al., 2022). Similarly, Nazir et al. (2021) also suggest shedding light on the consequences of PL in various high-power distance cultures. Therefore, we fulfill this existing literary gap by examining the same in Pakistan. Pakistan has a high-power distance culture (Mangi et al., 2012) with a collectivistic society (Islam et al., 2020b). According to Nazir et al. (2021), "the organizational context in Pakistan characterizes high uncertainty avoidance organizational practices" (p. 1362), including centralization, nepotism and corruption. In addition, people follow the rule of "might is right", which largely focuses on group, race, family and cultural values and ignores individual values. In this family-oriented culture, decisions are made by the head of the family which members have to accept without reservations . As individuals bring similar values in their workplace, therefore, help them to accept paternalistic leadership (Nazir et al., 2021). Koveshnikov et al. (2022) also argued that individuals response to their leaders as per their cultural orientation.
Although studies have noted employees' high tolerance level towards abusive/ authoritative leaders in high-power distance cultures (Naseer et al., 2018), still we found, such leadership is negatively associated with employees' job-related attitudes (turnover intention and job satisfaction). Therefore, our findings add to the existing literature of PL in high-power distance cultures and suggest that individuals in such cultures are not necessarily to reciprocate negative leadership style with positive attitudes. This may be because, sometimes the leader-follower relationship depends upon followers' perception of having similar personalities with their leaders (Li and Sun, 2015). This means that employees perceive their leaders' personalities through shared values and beliefs and if found similar to theirs, show similar reciprocations.
Second, our study explains the mechanism through which variation in employee turnover intention can be analyzed. The study noted that when employees perceive their leaders as moral and benevolent, they tend to be more satisfied with their job which reduces their intention to leave the organization. However, when employees perceive their leaders as authoritative, they show less satisfaction which leads to greater turnover intention. Employees with the greater intention to leave the organization become ready to change their organization at the first opportunity (Islam et al., 2013). Finally, our study investigates POS as a conditional variable that triggers the negative effect of job satisfaction on turnover intention. Most of the past studies have used POS as a predictor of job satisfaction and turnover intention, however, our study shows a new direction to the researcher by suggesting the conditional role of POS. Fourth, our findings contribute to SET that leadership is considered an important aspect of exchange relations.

Practical implications
Our findings have several implications for the management. First, employee turnover costs an organization ten times greater than retaining an existing one; therefore, our study suggests management not to ignore leadership. Because, employees perceive their leaders/ supervisors as representative of their organization, hence reciprocate accordingly. More specifically, when employees perceive their leaders' morals and benevolence, they not only prefer to continue with the same organization but also exhibit greater performance and vice versa. If organizations fail to engage their employees through positive gestures (effective leadership) may face serious consequences (employee turnover). Our findings let the managers understand that the social dominance culture in Pakistan seems to be productive up to some extent but if authority is not exercised rightly at the right time, organizations can have more chances to lose talented employees (Chen and Kao, 2009). Therefore, our study suggests management to train their leaders/supervisors regarding benevolent and moral values.
Our findings also suggest that positive leadership is not sufficient to retain employees, the role of other organizational factors (such as POS) is equally important. Paternalistic leaders have special characteristics (like caring for employees' well-being) which employees consider their leader's gestures and not the gestures by their organization, thereby more reciprocal towards leaders rather than organizational goals. In such situations, POS serves as a conditional variable that helps them continue with their organization. Management needs to understand that employees build POS based on mutual belief that if they would be cared by their organization, in return, they will serve their organization. In addition, employees perceive their organizational support positively when their leaders'/supervisors would treat them with positive styles (benevolence and moral). Rhoades and Eisenberger (2002) also emphasized that management by working on POS could get the desired results from the employees. Thus, management should take initiatives towards "employee care" which is possible through employee-centric policies and positive leadership in the organization. Working on positive leadership (moral and benevolence styles) and employee care (POS) management can better build a retention strategy.

Limitations and future recommendations
Our study is not free from limitations. First, we used a cross-section design to collect data. Cross-section design is associated with CMV and restricts causality. Although, we applied Harman's Single factor test where CMV was not an issue in this study, still, future researchers are suggested to use cross-section time-lagged design to better understand how employees' Paternalistic leadership and turnover intention perception and attitudes vary (Islam and Irfan, 2022). In addition, we suggest future researchers conduct longitudinal designs to better predict causality. Second, the benevolent, moral and authoritative dimensions of PL may produce social approbation effect (Chan, 2017); therefore, future researchers can use other data collection methods (i.e. observation or interview). Third, our study noted positive outcomes of benevolent and moral leadership; nevertheless, literature has identified their negative outcomes (Rui and Xinqi. 2020). Therefore, future studies are suggested to examine how and when benevolent and moral dimensions negatively affect employees' work-related outcomes. Fourth, we noted that most of the respondents were aged between 21 and 35 years (millennials). Millennials are more sensitive and stimulated to various organizational factors; therefore, examining generation X (mature and hard worker) may produce different results. Finally, as the empirical evidence on the mechanism between the dimensions of PL with employees' job-related outcomes are limited, therefore, the future researcher can use psychological ownership, well-being as mediating variables and psychological contract breach, person-job fit as moderating variables.