Paternalistic leadership and employee voice: Does information sharing matter?

Abstract

The study examined the impact of paternalistic leadership behaviors, including authoritarian, benevolent and moral leadership, and information sharing on employee voice and whether information sharing moderates the effects. Using a sample of 286 leader–follower dyads collected from a manufacturing firm, the results indicated that authoritarian leadership was negatively, and moral leadership positively, associated with employee voice. Also, the positive relationship between moral leadership and employee voice was stronger when employees received higher levels of information sharing. Implications for theory and practice are discussed.

Keywords

employee voice information sharing paternalistic leadership

Introduction

In the last few years there has been increasing research on the paternalistic role of the leader (Chan et al., 2013; Pellegrini and Scandura, 2006, 2008; Wang and Cheng, 2010). Paternalistic leadership (PL) is defined as strong authority that is accompanied by concern and consideration (Westwood and Chan, 1992). The construct of PL is not defined as a unified construct (Aycan, 2006; Farh et al., 2006). Indeed, Farh and Cheng (2000) classified PL behaviors into three dimensions: authoritarianism, benevolence and morality. Researchers have devoted great effort to examining the relationship between PL behaviors and employee attitudes, such as commitment and satisfaction (Erben and Güneser, 2008; Farh and Cheng, 2000; Pellegrini and Scandura, 2006, 2008).

Evidence has demonstrated that a negative effect of authoritarian leadership is contradictory to the positive effect of benevolent and moral leadership on employee work behaviors (Farh et al., 2008). Although one recent PL study (Chen et al., in press) focused on affective trust in the leader as a mediator between PL behaviors and performance, there is less work considering possible explanations for the effect of PL behaviors on employee proactive behaviors, such as employee voice. Hence, the first purpose of this study is to examine the relationship between PL behaviors and employee voice.

Research has indicated that individuals are more willing to speak up when working in open organizational environments (Huang et al., 2005; Gorden et al., 1988), when sharing resources among themselves (Moye and Langred, 2004) and when receiving empowerment from their leaders (Gao et al., 2011). Individuals prefer to work in environments where they can speak up and share their thoughts openly (Milliken et al., 2003; Zhou and George, 2001). In particular, one study has explored how employees’ trust in leaders influences employee voice, moderated by the role of empowering leader behaviors, for example, informing (Gao et al., 2011). The informing of empowering leadership behavior refers to the leader’s dissemination of company-wide information, such as mission and philosophy as well as other important information (Arnold et al., 2000).

In this article I define the possible effects of this type of empowering leadership behavior on employee voice in relation to information sharing. Information sharing is an important feature of an empowering work context for employees to receive authority, power and responsibility. It allows employees to create a good understanding of what their roles and responsibilities are in organizations. In general, information sharing is the most direct way to state the affairs of the work environment to employees instead of, for example, participative decision making, which requires intense employee participation in sharing their opinions and coaching that requires face-to-face communication in the empowering process. In other words, information sharing is a one-way delivery of empowerment that enables employee suggestions. Specifically, I argue that information sharing may interact with the three components of PL (authoritarianism, benevolence and morality) in predicting employee voice. Therefore, the second purpose of this study is to examine whether information sharing moderates the effects of PL behaviors on employee voice.

To summarize, this study intends to make two major contributions. First, this study extends the leadership literature by investigating whether PL behaviors can influence employee voice (Pellegrini and Scandura, 2006, 2008). Studies have indicated a negative effect of authoritarian leadership and a positive effect of benevolent and moral leadership on employee behavioral outcomes (e.g. Pellegrini and Scandura, 2008). While employee voice has been studied in relation to western leadership behaviors, the investigation of PL behaviors can advance the understanding of the antecedents of employee voice in the Chinese context. It takes into account the impact of PL behaviors on employee voice from a cultural perspective.

Second, past studies have not taken into account moderators to explain the impact of PL behaviors on employee voice (e.g. Gao et al., 2011), such as information sharing. This study examines information sharing, and its combined effects with authoritarian, benevolent and moral leadership on employee voice. Some prior studies have revealed that information sharing does not always lead to positive employee work outcomes and voice in organizations by itself (Randel and Ranft, 2007; Renzl, 2008). By understanding the PL-voice interaction, however, this study strives to improve the knowledge on when PL behaviors may affect employee voice.

Literature review

Paternalistic leadership

PL was developed specifically as an indigenous Chinese leadership theory (e.g. Pellegrini and Scandura, 2006, 2008; Wang and Cheng, 2010). According to Farh and Cheng (2000), PL behaviors were defined as a three-dimension model: authoritarianism, benevolence and morality. Authoritarian leadership refers to leadership behavior that exerts absolute authority and demands unquestionable obedience. Benevolent leadership refers to leadership behavior that demonstrates individualized concern for personal or familial well-being beyond work relations. A benevolent leader devotes energy to take care, show concern and encourage employees when they encounter problems. Moral leadership refers to leadership behavior that demonstrates superior personal virtue, self-discipline and unselfishness (Farh and Cheng, 2000). A moral leader does not use personal relationships or improper practices to obtain illicit advantages (Gelfand et al., 2007). There is now a body of evidence suggesting that PL behaviors help to explain employee work behaviors (Pellegrini et al., 2010; Soylu, 2011).

Paternalistic leadership and related leadership theories

When reviewing the development of the PL model, researchers have raised the issue of the extent to which the dimensions of PL behaviors may overlap with existing theories, such as transformational leadership (Cheng et al., 2004; Pellegrini and Scandura, 2006, 2008). As a result, it is necessary to clarify the similarities and distinctions between PL behaviors and related leadership theories.

First, authoritarian leadership is conceptually similar to the theory of autocratic leadership (Cremer, 2006), which is defined as not taking care of the group cohesion and denying individuals control over decision making. However, the theoretical distinction is that autocratic leadership was developed in the western context, while authoritarian leadership reflects the cultural characteristics of Confucianism, with family power, paternalistic control and submission to authority (Westwood, 1997).

Second, benevolent leadership resembles the individualized consideration dimension of transformational leadership, as a benevolent leader provides support and encouragement to subordinates. A benevolent leader builds on the development and maintenance of a strong interpersonal relationship that helps employees attain their full potential. However, unlike leader–member exchange (LMX) theory, benevolent leadership goes beyond individualized concerns to include a concern about the welfare of family members.

Third, a transformational leader demonstrates high ethical standards and engages in the ‘moral uplifting of followers’ (Bass, 1997: 131). The idealized influence of transformational leadership is portrayed as the morality dimension of PL, which raises the consciousness of individuals to higher levels of morality and motivation by appealing to their ideals and moral values (Bass, 1985; Burns, 1978). However, moral leadership owes its origin to the Confucian philosophy of governance and emphasizes the ethical standard.

Employee voice

Originally, employee voice referred to any attempt to change, rather than to escape from, an objectionable state of affairs (Hirschman, 1970). According to the voice literature, employee voice has been defined in two ways (Avery and Quinones, 2002). It has been defined on the one hand as the expression of an individual’s views to decision makers (Thibaut and Walker, 1975) and, on the other hand, as an expression of job dissatisfaction (Hirschman, 1970; Milliken et al., 2003; Withey and Cooper, 1989). Employee voice has been conceptualized differently among researchers. Accordingly, Van Dyne et al. (1995) defined employee voice as ‘proactively challenging the status quo and making constructive suggestions’ (p. 266). Detert and Burris (2007) defined voice as ‘the discretionary provision of information intended to improve organizational functioning which may challenge and upset the status quo of the organization and its power holders’ (p. 869). It is a risk-taking behavior that emphasizes the expression of a constructive challenge with intent to improve, rather than merely to criticize (Van Dyne and LePine, 1998: 108).

Researchers have generally agreed that this definition of employee voice as challenging-promotive behavior is appropriate for examining the impact of, and to target the motives behind, employee behaviors (Gao et al., 2011; Van Dyne and LePine, 1998). For example, voice behaviors are about taking the initiative to respond, making comments and suggestions, and accepting responsibility to speak up, and employees emphasize flexibility, promote innovation and improve organizational performance (LePine and Van Dyne, 1998; Morrison and Milliken, 2000; Quinn and Spreitzer, 1997). In employees’ perception of leadership behavior, it is perhaps important for them to express their concerns and suggestions for improvement, which explains the importance of leadership as an antecedent of employee voice (e.g. Detert and Burris, 2007; Liu et al., 2010). In this study, employee voice is defined as a discretionary, extra-role, risk-taking behavior, which potentially challenges the status quo and seeks to make constructive change.

The impact of leadership behaviors on employee voice is widely examined across various studies (Morrison et al., 2011; Tangirala and Ramanujam, 2008, 2012; Venkataramani and Tangirala, 2010). For example, ethical leadership (Walumbwa et al., 2012), transformational leadership (Liu et al., 2010) and leaders’ perceived display of openness (Detert and Burris, 2007) are typical leadership behaviors that explicate their impact on employee voice. Walumbwa et al. (2012) argued that group voice mediated the relationship between ethical leadership and performance. Employee identification is an explanation for the influence of transformational leadership on how employees facilitate both the speaking out and speaking up features of voice.

Hypotheses

Paternalistic leadership and employee voice

PL behaviors have been shown to be associated with employee work performance and organizational citizenship behaviors (Chan et al., 2013; Chen et al., in press; Cheng et al., 2002). An authoritarian leader normally imposes strict controls on resources, rules and hierarchy. This behavior may discourage employee voice, as employees are required to obey leaders’ instructions completely. As explained by the notion of personal control (Tangirala and Ramanujam, 2008), when personal control of employees is such that they have low autonomy and impact, employees have limited independence to choose their work behavior (Brockner et al., 2004). An authoritarian leader may thus play a domineering role and does not allow employees to think independently, which may, in turn, undermine and depress employee voice behavior. Given the potentially high risk of voice behavior, under these circumstances, employees may prefer to remain silent rather than voice their thoughts. Alternatively, employees can reassert control by expressing their own feelings.

Benevolent leadership is expected to be positively associated with employee voice by facilitating the social exchange relationship (Blau, 1964). In a high-quality exchange relationship, a benevolent leader gives favors to employees and employees reciprocate with their constructive voice. The benevolence of a leader may encourage employee feedback by asking employees for opinions and providing solutions for work improvement. Specifically, a benevolent leader can help employees to eliminate uncertainty, which may encourage them to voice their thoughts (Schweiger and DeNisi, 1991). However, if employees perceived a leader absent of such benevolence behavior, employees may compare the perceived benefits and risks of speaking up. Employees are more likely to be engaged in a risk endeavor when they are comfortable with the benevolence of their leader.

Moral leadership has been identified as a leadership behavior that provides high ethical standards. A highly moral leader demonstrates self-discipline and unselfish behaviors and serves as a personal role model. This leadership behavior encourages employees to move beyond compliance with formal agreements, which in turn increases employee confidence in leaders’ judgments. A moral leader demonstrates integrity and is concerned with the collective interest instead of solely with self-interest. This is likely to motivate employees to engage in voice behaviors, since the leader provides a role model concerned with the common interest and improvement, and faced with such a leader employees are less likely to perceive a high risk in voicing their thoughts. As a result, employees should be more willing to speak up and voice their thoughts with the aim of making constructive comments. Thus, benevolent and moral leadership are positively related to employee voice.

Hypothesis 1: (a) Authoritarian leadership is negatively associated with employee voice, and (b) benevolent leadership and (c) moral leadership are positively associated with employee voice.

The moderating role of information sharing

Empirical studies have shown that leaders generally do share information with employees in making decisions (Gigone and Hastie, 1993; Henry, 1995; Stasser et al., 1995). Employees who have received information from their leader are more likely to emerge in the group discussion and participate in making constructive suggestions. With respect to the information exchange theory, information sharing facilitates a formal flow of communication among employees (Ling and Kellermanns, 2010; Stasser and Titus, 1987; Withey and Cooper, 1989). In one study, Lam and Schaubroeck (2011) found that when information was not completely shared among employees, employees of a low-efficacy group communicated more and had a higher chance of changing their initial decision preferences.

An authoritarian leader that ignores suggestions and avoids difficult questions implies that he/she disrespects the ability of the employees. Employees perceive authority and strict regulations as compulsory actions that limit their voice of thoughts. Specifically, Wu et al. (2002) suggested that authoritarian leadership behaviors may undermine and impair employees’ competence at work. The behavioral controls imposed by an authoritarian leader are likely to signal to employees that they are considered to be less capable and less valued individuals. An authoritarian leader may be expected to question the credibility and accuracy of employee voice. Although an authoritarian leader may be unlikely to listen, employees who have obtained more information have a higher incentive to speak up. As such, for employees who perceive higher levels of information sharing, the negative relationship between authoritarian leadership and employee voice may be weaker.

A benevolent leader who provides care, protection and guidance to employees may direct employees’ attention towards causal information. Benevolence may help employees to interpret the information for work domains; employees will show loyalty and commitment to their benevolent leader who is seen as being motivated by good intentions (Aycan, 2006). If employees are impressed by their benevolent leader, they are more likely to come up with constructive suggestions. The sharing of information may help employees to obtain greater evidence for judgmental tasks and problem solving (Stasser et al., 1995). Employees feel satisfied by the prospect of sharing their own ideas and opinions, and they are committed to fully utilizing the information provided. Employees enjoy their benevolent leader’s support of their constructive voice when they receive relevant information and resources from their organization.

When opportunities for sharing information exist and communication channels are open, employees are eager to be involved in decision making and actively participate under the supervision of a moral leader. Moral leadership behaviors that demonstrate integrity and concern for the collective interest instead of personal interest result in more employee voice. Moral leaders present a role model for employees, which enables employees to feel obligation to work and responsibility for the job. Gao et al. (2011) indicated that the positive relationship between trust in leader and employee voice is stronger when informing leadership behavior is higher. Informing would enable employees’ motivation to identify the problems and provide suggestions, such that the effect of morality on employees’ voice behaviors will be higher. On the other hand, the employees’ beliefs that their moral leader expects them to speak up would be mitigated when there is a lack of information sharing. The relationship between moral leadership and employee voice will be stronger when employees receive more information sharing. Taken together:

Hypothesis 2: The negative relationship of (a) authoritarian leadership is weaker and the positive relationship of (b) benevolent leadership and (c) moral leadership with employee voice is stronger when employees receive higher levels of information sharing.

Figure 1 presents a research model depicting the key relationship of PL behaviors with employee voice, moderated by information sharing.

Figure 1.

Research framework.

Pilot study

Previous research on PL has employed Cheng et al.’s (2000) PL scales (Aryee et al., 2007; Chan et al., 2013; Wang and Cheng, 2010). Pellegrini and Scandura (2008) have reviewed the importance of sufficient construct clarity in the measurements of PL behaviors. Given that the dimensions of PL behaviors are validated in the Chinese context (Farh and Cheng, 2000; Farh et al., 2008), Cheng et al. (2000) paid attention to the conceptual commonality among the items and found that the three dimensions of PL were statistically distinct and each subscale had a high internal consistency in two independent samples from Taiwan (a private conglomerate and a set of primary schools). The construct validity of the PL was preliminarily established. In spite of this, as yet researchers have not pointed out conceptual and methodological limitations in using Cheng et al.’s (2000) PL constructs in the existing literature.

First, in Cheng et al.’s (2004) study, the measurements of PL behaviors and subordinates’ work outcomes were all self-reported. The results may be influenced by common method biases. Second, although the PL behaviors were examined and proved to explain the unique variance in subordinates’ outcomes after controlling transformational and transactional leadership in two Chinese studies (Cheng et al., 2002, 2004), the results were reported and published in Chinese. Third, the dimensions of PL were derived inductively from the management style of the Chinese family business, which is not guided by a theoretical framework in the application of modern organizations (Cheng et al., 2004).

To increase the validity and generalizability of the constructs of PL behaviors, it is expected that PL (employees’ perceptions of PL behaviors) can explain unique variance in employee outcomes beyond that which is explained by transformational leadership. Therefore, a pilot study was conducted that includes transformational leadership as a control variable to test the relationship between PL behaviors and employee extra-role performances: (1) employee voice, (2) organizational citizenship behavior for organizations (OCBO) and (3) organizational citizenship behavior for individuals (OCBI). The direct effect of PL behaviors (i.e. authoritarianism, benevolence and morality) on employee voice, OCBO and OCBI, when controlling for transformational leadership, was tested.

Sample and procedure

The sample included 220 leader–follower dyads of a manufacturing factory in the People’s Republic of China. To prepare survey packages, a list of personnel, including the names of employees and their immediate supervisors, and organizational structure information were provided. The author visited the data sites and met all of the respondents in person (220 leaders and 220 employees separately), briefed them about the purposes of the study and explained the procedures for implementing the survey. The respondents were volunteer participants who filled in the surveys on site and enclosed and sealed the completed questionnaires in the provided envelopes. Each questionnaire was coded with a researcher-assigned identification number in order to match employees’ responses with their immediate leaders’ evaluations.

Two sets of questionnaires were used for employees and for their immediate leaders. The employee questionnaires asked about the perception of their immediate leader’s PL behaviors, transformational leadership behaviors and demographic questions. The immediate supervisor questionnaire asked about employee voice, OCBO, OCBI and demographic questions.

Of the 220 sets of questionnaires, 202 leader–follower dyads returned usable questionnaires, with usable response rates of 91.8 percent. All participants voluntarily joined the pilot study. For the employee sample, 56 percent were female and 63.3 percent had college education level or above. The mean age, organizational tenure and leader tenure of the employees were 30.97, 3.50 and 2.43 years, respectively. For the immediate leader sample, 47.5 percent were female and 88.2 percent had a college education or above, and the mean age was 31.83.

Measures

Paternalistic leadership (PL)

Cheng et al. (2000) scales were used to measure paternalistic leadership behaviors. This study asked employees to rate the leadership style of their immediate leaders with a 9-item scale on authoritarian leadership, an 11-item scale on benevolent leadership and a 6-item scale on moral leadership (1 = strongly disagree; 6 = strongly agree). The sample item for authoritarian leadership was ‘My leader asks me to obey his/her instructions completely.’ The sample item for benevolent leadership was ‘My leader is like a family member when he/she gets along with us.’ The sample item for moral leadership was ‘My leader employs people according to their virtues and does not envy others’ abilities and virtues.’ The alpha coefficient of authoritarian, benevolent and moral leadership were .88, .95 and .80 respectively.

Transformational leadership

Podsakoff et al.’s (1990) 23-item scale was used to measure transformational leadership along six dimensions: identifying and articulating a vision, providing an appropriate model, fostering the acceptance of group goals, high performance expectations, providing individualized support and intellectual stimulation (1 = strongly disagree; 7 = strongly agree). The alpha coefficient of identifying and articulating a vision, providing an appropriate model, fostering the acceptance of group goals, high performance expectations, providing individualized support and intellectual stimulation were .90, .83, .82, .85, .86 and .80 respectively.

Employee voice

Immediate supervisors were asked to assess employee voice by using a 6-item scale developed by Van Dyne and LePine (1998) (1 = strongly disagree; 7 = strongly agree). A sample item from this measure is ‘This particular employee develops and makes recommendations concerning issues that affect this work group.’ The alpha coefficient was .94.

Organizational citizenship behavior (OCB)

Organizational citizenship behavior was measured by using a 16-item scale developed by Lee and Allen (2002) (1 = never; 7 = always). OCB was drawn from the OCBI and OCBO scales. OCBI refers to pro-social behavior towards co-workers, and OCBO measures pro-social behavior towards the organization as a whole. There are eight items for OCBI, for example, ‘The subordinate helps others who have been absent.’ Another eight questions measured OCBO, for example, ‘The subordinate offers ideas to improve the functioning of the organization.’ The alpha coefficient for OCBI and OCBO were .93 and .80, respectively.

Control variables

As one of the most systematic streams of voice research is to examine the demographic characteristics of employees on voice behaviors (Detert and Burris, 2007; LePine and Van Dyne, 2001), characteristics of individuals may correlate with employee voice. Sex, age and education levels are the demographic characteristics that influence subordinate outcomes (McEvoy and Cascio, 1989; Sinclair et al., 1999). The organizational tenure and dyadic leader–follower tenure relationships of a leader also tend to influence employee voice (e.g. Tangirala and Ramanujam, 2008; Van Dyne and LePine, 1998). For instance, long-tenure employees are more willing to engage in voice behaviors than short-tenure employees (Stamper and Van Dyne, 2001).

Measures of gender, age, education levels, organization and leader–follower dyads tenure were obtained from employees. Gender was dummy coded (0 = male; 1 = female). A five-point scale was used to measure the education levels of the respondents (1 = primary school; 2 = secondary school; 3 = college level; 4 = university degree; 5 = Master’s degree or above). Age, organization tenure and leader–follower dyads tenure were self-reported years.

The procedures for survey translations across different languages were applied (Brislin, 1990). The questionnaire items that included transformational leadership, information sharing and employee voice were translated from English into Chinese. A back-translation was conducted, with the items being translated back to English by another bilingual academic to ensure that both English and Chinese versions of the items were comparable with a high degree of accuracy.

Preliminary analyses

Before testing the hypotheses, two confirmatory factor analyses (CFA) were conducted to assess the discriminant validity of the constructs. First, with the specified items loading on their respective dimensions, the expected three-factor model of PL behaviors including authoritarianism, benevolence and morality (χ²[177.91], χ²/d.f. = 2.96; Comparative Fit Index (CFI) = .91, Tucker Lewis Index (TLI) = .89, Root Mean Square Error of Approx-imation (RMSEA) = .09) yielded a significantly better fit than the single-factor model (χ²[493.92], χ²/d.f. = 7.84; CFI = .68, TLI = .60, RMSEA = .18). The three-factor structure of the PL scale was confirmed in the pilot study. Second, another CFA was conducted to assess the distinctiveness of the seven-factor structure: authoritarian leadership, benevolent leadership, moral leadership, transformational leadership, employee voice, OCBO and OCBI. Due to the large number of items that were used to measure all constructs, three to four manifest indicators for each construct were created by randomly assigning items to composites of the constructs. Results showed that the seven-factor model (transformational leadership, authoritarian leadership, benevolent leadership, moral leadership, employee voice, OCBO and OCBI) yielded a better fit (χ²[245.72], χ²/d.f. = 2.53; CFI = .91, TLI = .89, RMSEA = .08) than a three-factor model (transformational leadership, paternalistic leadership, employee extra-role behavior) (χ²[529.71], χ²/d.f. = 4.61; CFI = .72, TLI = .68, RMSEA = .15) and the one-factor model (χ²[1133.65], χ²/d.f. = 9.61; CFI = .43, TLI = .35, RMSEA = .21). In sum, transformational leadership, authoritarian leadership, benevolent leadership, moral leadership, employee voice, OCBO and OCBI were different constructs.

Comparison between the subscales of transformational leadership and paternalistic leadership

In order to determine how the six transformational leadership subscales and the three paternalistic leadership subscales were related to employee voice, OCBO and OCBI, a simple correlational analysis was performed. The means, standard deviations and bivariate correlations of all key variables in the pilot study are presented in Table 1.

Table 1.

Means, standard deviations, correlations and reliabilities of measures in the pilot study.^{a, b, c, d}

Variables	Mean	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
1. Gender	.56	.49	–
2. Age	30.97	7.85	−.35**	–
3. Education	2.71	.60	−.05	.21**	–
4. Organization tenure	3.5	2.21	−.27**	.48**	.12	–
5. Dyad tenure	2.43	1.37	−.15*	.38**	.14*	.73**	–
6. TF – AV	5.4	.76	.14*	.00	.00	−.05	.09	.90
7. TF – PA	5.22	.78	.24**	−.11	.03	−.09	.04	.56**	.83
8. TF – FAG	5.58	.61	.19**	−.12	−.08	−.14	.00	.57**	.45**	.82
9. TF – HPE	5.48	.61	.12	−.21**	.05	−.24**	−.15*	.29**	.23**	.46**	.85
10. TF – PIS	5.17	.62	.12	−.13	.02	−.22**	−.07	.28**	.28**	.30**	.26**	.86
11. TF – INS	5.24	.65	.17*	−.18	.02	−.22**	−.06	.31**	.31**	.34**	.40**	.87**	.80
12. Authoritarian leadership	3.09	.82	−.04	.04	.19**	.01	−.05	−.18*	−.02	.02	.00	.00	−.02	.88
13. Benevolent leadership	4.38	.86	.21**	−.14	−.08	−.14*	−.08	.54**	.45**	.43**	.23**	.44**	.42**	−.12	.95
14. Moral leadership	4.65	.75	.16*	−.18*	−.11	−.08	−.03	.41**	.33**	.43**	.15*	.24**	.16*	−.14*	.58**	.80
15. Employee voice	5.62	.87	.24**	−.14*	−.09	−.05	.06	.34**	.39**	.40**	.23**	.22**	.28**	−.37**	.41**	.32**	.94
16. OCBO	5.48	.85	.25**	−.20**	−.12	−.09	−.07	.29**	.35**	.37**	.26**	.29**	.31**	−.26**	.35**	.29**	.79**	.93
17. OCBI	5.65	.57	.18**	−.06	−.14	−.02	−.03	.26**	.33**	.35**	.25**	.21**	.25**	−.25**	.36**	.23**	.70**	.88**	.80

n = 202.

The correlation coefficients are significant at ^*p < .05; ^**p < .01.

Reliability coefficients appear along the diagonal.

TF = transformational leadership, AV = identifying and articulating a vision, PA = providing an appropriate model, FAG = fostering the acceptance of group goals, HPE = high performance expectations, PIS = providing individualized support, INS = intellectual stimulation, OCBO = organizational citizenship behavior for organizations, OCBI = organizational citizenship behavior for individuals.

Dummy coded variable (Gender): 0 = male; 1 = female.

Results indicated that the majority of the correlations are strong, positive and significant at the p = 0.01 level, except the correlations of the subscales of transformational leadership and authoritarian leadership. This pattern of correlations suggests that benevolence and morality are significantly related to the subscales of transformational leadership. Some correlations of the subscales of transformational leadership and paternalistic leadership are relatively high. For example, the correlation between identifying and articulating a vision (AV) and benevolent leadership is 0.54 (i.e. the strongest one). It is clear that the level of correlation in the pilot study is appropriate. The reasonably high correlations among the leadership subscales suggest that ratings are substantially related. The findings explained that, except for authoritarian leadership, there is not a single subscale of paternalistic leadership assessing transformational leadership with a near-zero correlation. Also, it should be noted that the correlation between OCBO and OCBI is high, at 0.88.

Results of the pilot test

The pilot study indicated that PL behaviors explain additional variance in employee voice, OCBO and OCBI beyond that explained by transformational leadership. As reported in Model 2 of Table 2, three of the subscales of transformational leadership – identifying and articulating a vision, providing an appropriate model and fostering the acceptance of group goals – were positively and significantly related to employee voice (β = .22, p < .05; β = .24, p < .01; β = .29, p < .01). Providing an appropriate model and fostering the acceptance of group goals were significantly related to OCBO (β = .25, p < .01 and β = .26, p < .01), and OCBI (β = .20, p < .05 and β = .26, p < .01). These results explained 18 percent, 18 percent and 16 percent of the variance in employee voice, OCBO and OCBI by transformational leadership respectively. After controlling for transformational leadership, the three PL behaviors were tested simultaneously for each outcome variable as shown in Model 3 of Table 2. Authoritarian leadership is negatively associated with employee voice (β = −.25, p <.01), OCBO (β = −.16, p < .05) and OCBI (β = −.15, p < .05). Benevolent leadership is positively associated with employee voice (β = .27, p < .01), OCBO (β = .17, p < .05) and OCBI (β = .27, p < .01). Moral leadership is not associated with employee voice (β = .07, NS), OCBO (β = .06; NS) and OCBI (β = −.01, NS).

Table 2.

Results of regression analysis for the effect of transformational leadership, paternalistic leadership behaviors on employee voice, OCBO and OCBI in the pilot study.

Model (step) and variables entered	Employee voice			OCBO			OCBI
	1	2	3	1	2	3	1	2	3
1. Control variables
Gender	.21**	.17*	.15*	.21**	.19**	.17**	.19*	.16*	.15*
Age	−.08	−.11	−.09	−.12	−.12	−.10	.02	.03	.04
Education	−.07	−.05	001	−.09	−.07	.04	−.13	−.12	−.08
Organizational tenure	−.10	.03	.02	.02	.16	.15	.06	.18	.17
Leader tenure	.20*	.11	.12	.01	−.08	−.07	−.03	−.11	−.09
2. Transformational leadership
AV		.22*	.10		.16	.08		.13	.04
PA		.24**	.25**		.25**	.25**		.20*	.22
FAG		.29**	.13		.26**	.16		.26**	.17
HPE		.01	.06		.07	.10		.12	.15
PIS		.14	.19		.10	.06		.03	.00
INS		.24	.26		.08	.10		.10	.09
3. Independent variables
Authoritarian leadership (AL)			−.25**			−.16*			−.15*
Benevolent leadership (BL)			.27**			.17*			.27**
Moral leadership (ML)			.07			.06			−.01
Overall R²	.09	.27	.39	.09	.27	.32	.05	.21	.28
Adjusted R²	.06	.22	.34	.06	.23	.27	.03	.17	.22
Δ R ²	.09	.18	.12	.09	.18	.05	.05	.16	.06
Model F	3.64**	6.26**	8.22**	3.65**	6.37**	3.12**	2.12	4.69**	5.11**

n = 202.

p < .05; ** p < .01.

AV = identifying and articulating a vision, PA = providing an appropriate model, FAG = fostering the acceptance of group goals, HPE = high performance expectations, PIS = providing individualized support, INS = intellectual stimulation, OCBO = organizational citizenship behavior for organizations, OCBI = organizational citizenship behavior for individuals.

Dummy coded variable (Gender): 0 = male; 1= female.

Results indicated that authoritarian leadership was negatively, and benevolent leadership was positively, associated with employee voice, OCBO and OCBI. Authoritarian leadership may have a negative effect on employees’ extra-role performance as it undermines their organizational-based self-esteem (Chan et al., 2013). The findings were motivated to explain a positive relationship of employee individualized care of a benevolent leader on employee voice and OCB. In contrast, morality does not affect employee voice and OCB with the influence of a transformational leader held constant. Results indicated that a leader who demonstrates as a role model in front of his/her employees and helps employees to foster their acceptance goals is more important than one who provides high ethical standards. One possible explanation is that, conceptually, the subscale of transformational leadership (i.e. providing an appropriate model) is similar to the idea of serving as a personal role model as a moral leader. It may underscore the impact of moral leadership on employee voice and OCB by controlling the effect of transformational leadership.

To conclude, the inclusion of three PL behaviors explained the additional 12 percent, 5 percent and 6 percent of the variance on employee voice, OCBO and OCBI, respectively. Taken together, results with respect to employee voice, OCBO and OCBI provided support to the discriminant validity of PL behaviors. The theory and empirical data in the pilot study supported the notion that PL behaviors are a unique leadership theory that can extend researchers’ knowledge about leadership processes and outcomes.

Main study: Research method

The results of the pilot study supported the notion that PL behaviors (i.e. authoritarianism, benevolence and morality) explain additional variance beyond transformational leadership in employee voice and OCBs. By seeking whether PL behaviors would influence employee voice moderated by information sharing, the purpose of the main study was two-fold. First, a sample from a manufacturing firm in the People’s Republic of China was used to replicate the findings of the pilot test regarding the relationship between paternalistic leadership and employee voice. Second, the main study could test the moderating role of information sharing in the relationship of paternalistic leadership behaviors and employee voice.

Sample and procedure

The sample of this study included 286 leader–follower dyads in an office setting of a manufacturing firm in the People’s Republic of China. To prepare survey packages, a list of personnel including the names of employees and their immediate supervisors, and organizational structure information were provided. The author visited the data sites and met all of the respondents in person (groups of leaders and employees separately), briefed them about the purposes of the study and explained the procedures for implementing the survey. The respondents were volunteer participants who filled in the surveys on site, and enclosed and sealed the completed questionnaires in the provided envelopes. Each questionnaire was coded with a researcher-assigned identification number in order to match employees’ responses with their immediate leaders’ evaluations.

Two sets of questionnaires were used, one for employees (they were all office staff) and one for their immediate leaders. The employee questionnaires asked about their immediate leader’s PL behaviors, information sharing and demographic questions. The immediate leader questionnaire asked about employee voice and demographic questions.

Of the 309 sets of questionnaires, 286 usable questionnaires (286 leader–employee dyads) were returned, with a usable response rate of 92.5 percent. These 286 employees were supervised by 131 immediate leaders. The maximum number of surveys completed by a single leader was three. For the employee sample, 66.0 percent were female and 50.0 percent had a college education or above. The mean age and organizational tenure of the employees were 32.4 and 4.36 years, respectively. For the immediate leader sample, 49.8 percent were female and 97.0 percent had a college education or above. The mean age and organizational tenure of the leaders were 39.55 and 7.87 years, respectively.

Measures

The key measures, PL behaviors (the independent variable) and employee voice (the independent variable), were the same as in the pilot test.

Information sharing

Arnold et al. (2000) scales were used to measure information sharing. Information sharing was measured with six items answered by employees (1 = strongly disagree; 7 = strongly agree). Sample items from this measure include ‘My leader explains company goals,’ and ‘My leader explains his/her decisions and actions to my workgroup.’ The alpha coefficient was .94.

Control variables

Measures of the demographic variables were obtained from the employees, such as gender, age, organization and leader–follower tenure, as in the pilot study. A two-point-scale measured the education levels of the respondents (0 = below college; 1 = college or above). Age, organization tenure and leader–follower dyad tenure were self-reported years. Location refers to the geographic location at which employees carried out their work activities. As the location of work may affect employee attitudes, location was self-reported and dummy coded in the analysis (0 = non-HK region; 1 = HK region).

Main study: Results

Preliminary analyses

A confirmatory factor analysis (CFA) was conducted to assess discriminant validity. A five-factor model was assessed, including authoritarianism, benevolence, morality, information sharing and employee voice, as shown in Table 3. The null model allowed all items to load on one factor. Results showed that the five-factor model (Five factors – authoritarian leadership, benevolent leadership, moral leadership, information sharing, employee voice) yielded a better fit (χ²[669.99], χ²/d.f. = 2.52; CFI = .90, TLI = .88, RMSEA = .07) than a four-factor model (Model 1 − benevolent leadership and moral leadership were combined into one factor) (χ²[1087.41], χ²/d.f. = 4.04; CFI = .80, TLI = .76, RMSEA = .10), and a three-factor model (Model 2 − authoritarian leadership, benevolent leadership and moral leadership were combined into one factor) (χ²[1286.34], χ²/d.f. = 4.73; CFI = .76, TLI = .71, RMSEA = .11). As a result, the items for authoritarian leadership, benevolent leadership, moral leadership, information sharing and employee voice were measuring distinct constructs. The PL construct exists at the same level as the dimensions (i.e. authoritarianism, benevolence, morality) and these dimensions are in effect different leadership behaviors potentially with distinct impacts on employee voice.

Table 3.

Comparison of measurement models for constructs in the main study.^a

Samples	Models	Model comparison
		χ²	d.f.	Δχ2	CFI	TLI	RMSEA	Comparison
(N = 286)	Null model(All the indicators are independent)Baseline model (Five factors – authoritarian leadership, benevolent leadership, moral leadership, information sharing, employee voice)	669.99	265	–	.90	.88	.07
	Model 1(Four factors; benevolent leadership, and moral leadership were combined into one factor)	1087.41	269	417.42**	.80	.76	.10	Model 1 vs Null Model
	Model 2(Three factors; authoritarian leadership, benevolent leadership, and moral leadership were combined into one factor)	1286.34	272	616.35**	.76	.71	.11	Model 2 vs Null Model

Δχ² = Chi-square change between the five-factor model and the other alterative models.

CFI = comparative fit index; TLI = Tucker-Lewis coefficient; RMSEA = root mean square error of approximation.

* p < .05; ** p < .01.

The means, standard deviations and bivariate correlations of all key variables are presented in Table 4.

Table 4.

Means, standard deviations, correlations and reliabilities of measures in the main study.^{a, b, c, d, e}

Variables	Mean	SD	1	2	3	4	5	6	7	8	9	10	11
1. Gender	.66	.47	–
2. Age	32.40	8.72	−.27**	–
3. Education	.50	.50	−.12*	−.17**	–
4. Organization tenure	4.36	1.31	−.05	.55**	−.17**	–
5. Dyad tenure	2.86	.19	−.17**	.52**	−.12*	.76**	–
6. Location	.87	.26	−.05	.43**	−.07	.32**	.29**	–
7. Authoritarian leadership	3.49	.96	−.19**	.19**	.02	.08	.13*	.23**	.75
8. Benevolent leadership	3.62	1.02	−.09	−.07	.01	−.07	.01	−.01	−.40**	.90
9. Moral leadership	3.26	1.09	.11	−.13*	.01	−.07	−.09	−.05	−.27**	.42**	.89
10. Information sharing	3.83	.94	−.07	.01	−.07	.04	.09	−.19**	−.26**	.42**	.26**	.94
11. Employee voice	4.30	.86	−.07	.23**	.10	.14	.21**	.14*	−.14*	.16**	.17**	.11*	.82

n = 286.

The correlation coefficients are significant at ^*p < .05; ^**p < .01.

Reliability coefficients appear along the diagonal.

Dummy coded variable (Gender): 0 = male; 1 = female.

Dummy coded variable (Location): 0 = non-HK region; 1 = HK region. Location refers to the geographic location at which employees carried out their work activities.

Regression analysis

Given the nested data structure of 286 employees by 131 immediate supervisors, there is a possibility that employees under the same leader share common influences. Ordinary least squares regression might violate the assumption of independent observations in the statistical analysis. Two-level analyses using linear mixed models in SPSS to decompose the total observed variance into individual- and group-level residual variances were conducted in order to control for possibly the same leader effects. As shown in Table 5, the effect coefficients analyzed by multilevel analyses are comparable to the unstandardized regression coefficients in ordinary regression analysis. The goodness-of-fit of the paternalistic leadership model is indicated by a -2log likelihood statistic, whereby the difference between the statistics of two subsequent models (D-2log-likelihood) indicates the increase in model fit. The level of analysis is the individual, examining the interactive effects of PL behaviors and information sharing on employee voice, while controlling for the same leader effect (Janssen et al., 2010). All the variables were centered for moderated regression.

Table 5.

Results of regression analysis for the interactive effect of paternalistic leadership behaviors and information sharing on employee voice in the main study.^{a, b, c, d, e}

Model (step) and variables entered	Employee voice
	1	2	3	4
1. Control variables
Gender	.06	.00	.00	.00
Age	.02*	.02**	.02**	.02**
Education	.11*	.12**	.13**	.11**
Organizational tenure	−.02	−.03	−.03	−.03
Leader tenure	.05*	.06*	.06*	.06*
Location	−.13	−.17	−.18	−.20
2. Independent variables
Authoritarian leadership (AL)		−.15**	−.15**	−.15**
Benevolent leadership (BL)		.04	.03	.04
Moral leadership (ML)		.13*	.13*	.13*
3. Moderating variable
Information sharing (IS)			.03	.09
4. Interactive effects
AL x IS				−.06
BL x IS				.03
ML x IS				.08*
Overall R²	.09	.16	.16	.20
Adjusted R²	.07	.14	.14	.16
Δ R²	.09	.07	.00	.04
Model F	5.15**	6.32**	5.60**	5.31**
-2log likelihood	711.86	687.58	687.28	675.10
D-2log likelihood	711.86	24.28	24.58	36.76

n for hierarchical linear modeling = 131 supervisors and 286 employees.

p < .05; ^**p < .01.

Unstandardized coefficients are reported.

Estimate of the residual variances between leaders.

Dummy coded variable (Gender): 0 = male; 1 = female.

Dummy coded variable (Location): 0 = non-HK region; 1 = HK region. Location refers to the geographic location at which employees carried out their work activities.

Hypothesis test

Hierarchical multiple regression was used to test Hypothesis 1 following the steps suggested by Muller et al. (2005). After regressing employee voice on the control variables in step 1 of Table 5, authoritarian leadership, benevolent leadership and moral leadership were entered in step 2. Authoritarian leadership was negatively significantly related to employee voice (β = -.15, p < .01) and moral leadership (β = .13, p < .05) was positively significantly related to employee voice, lending support to Hypotheses 1(a) and 1(c), which predict that authoritarian leadership was negatively associated, and moral leadership was positively associated with employee voice. However, benevolent leadership was not significantly related to employee voice (β = .04, NS), and, therefore, Hypothesis 1(b) was not supported.

To test Hypothesis 2, information sharing and interactions of PL behaviors and information sharing were entered in step 3 and step 4 of Table 5. The results indicated that the interactive effect of authoritarian leadership and information sharing on employee voice (β = −.06, NS) and benevolent leadership and information sharing on employee voice (β = .03, NS) were not significant, respectively. Thus, Hypotheses 2(a) and 2(b) were not supported. The interactive effect of moral leadership on employee voice (β =.08, p < .05) was significant, which is plotted in Figure 2. Therefore, Hypothesis 2(c) was supported.

Figure 2.

The moderating effect of information sharing on the link between moral leadership.

Also, an additional analyses using multilevel modeling that takes into account both individual-level and group-level variability in the variables of interest was performed. The results of the multilevel analyses and the regression analyses were identical, providing very similar results. Thus, the results of the linear regression analyses were reported in the main text.

Discussion

This study examined how the three dimensions of PL behavior (authoritarianism, benevolence and morality) act as antecedents of employee voice and considered whether information sharing moderated these effects. In the present study, the demographic variables, age, educational background and leader tenure, included as control variables, have significant effects on employee voice. As hypothesized, the results suggested that authoritarian leadership was negatively, and moral leadership positively, related to employee voice. This suggests that employees are concerned about voicing their thoughts under the authoritarian leadership style. An authoritarian leader may view employee voice as ‘trouble making’ behavior by ‘rebellious’ employees (Van Dyne and LePine, 1998). Employees may fear sharing their thoughts and feel it is not safe to risk their personal interests (Detert and Burris, 2007). In other words, it implies that authoritarian leadership controls or suppresses employee voice behaviors.

In contrast, and again as expected, moral leadership had direct effect on employee voice. Moral leadership encourages employees to voice their concerns, perhaps imitating their leader in seeing involvement as a kind of responsible participation in the workplace. This positive association between moral leadership and employee voice is consistent with the findings in the literature. It suggests that leader morality encourages employees to give their opinions and suggestions. Unexpectedly, the results indicated that benevolent leadership had no effect on employee voice, suggesting that there is no evidence of a social exchange relationship linking the benevolence of a leader and employee voice behaviors. One possible explanation may come from the ideology of human relationships in Chinese society (Farh et al., 1997). Under the individualized care and support of a benevolent leader, employees may feel a need to maintain interpersonal harmony and so avoid voice for fear of damaging the relationship with their leader. The nonsignificant role of benevolence may thus not reflect the close relationship between leader and employees, which may prohibit the offer even of constructive suggestions and recommendations.

This study demonstrated a significant moderating effect of information sharing on the association between moral leadership and employee voice, although there was no moderation in the case of authoritarianism and benevolence. Information sharing did not play a critical intervening role in accounting for the negative effect of authoritarian leadership. A possible explanation is that authoritarian leadership imposes strict work standards, insists on following work rules and sets high performance standards. This pattern suggests that authoritarian leaders may enact other organizational policy or practice instead of employee voice. Also, this study did not indicate the positive effect of benevolent leadership behaviors on employee voice was stronger when employees received higher levels of information sharing. This type of leadership behavior may not help employees to promote their thoughts when a leader is willing to share the information with them.

The results indicated that the positive relationship between moral leadership and employee voice was stronger when employees received higher levels of information sharing. Consistent with this, the knowledge of organizational goals, policies and practices would induce a higher level of employee voice if there is high trust in the leader (Gao et al., 2011). Morality may also help to encourage constructive employee voice in fully utilizing the information, such that employees are more likely to accept the informing style of leadership. Employees can continue their voice behavior with a moral leader who enables employees to be responsible for the job duties and disseminate information.

Theoretical implications

Our findings have several theoretical implications. First, drawing on extant leadership theory, this study provides a PL framework for mapping employee voice. This is important because it adds value to the existing PL literature by examining the effects of the three-dimensional model of PL behavior in relation to employee voice (Farh and Cheng, 2000). The findings suggest that authoritarian leadership has a negative effect on employee voice, most likely because this leadership style inhibits employees’ tendency to display behaviors that might be interpreted by the leaders as challenging authority. The finding of a positive association between moral leadership and employee voice are to some extent consistent with other findings on the association between leadership behaviors and employee voice (e.g. Liu et al., 2010; Milliken et al., 2003). The perceived morality of a leader plays a critical role in shaping the effect of employee voice. Moral leadership behavior may help employees to develop a stronger commitment to their leader and more positive interpretations of their leaders’ responses to their opinions, suggesting to employees that there is little risk in speaking up, as a moral leader would be expected to be receptive to voice and would not be expected to respond negatively.

Second, this study responds to the call by Farh et al. (2008) for studies to investigate the potential boundary conditions that may amplify or neutralize the effects of PL behaviors. Information sharing may help to explain the effectiveness of PL behaviors in engendering employee voice activities. The results suggest that moral leadership is more likely to encourage employee voice under conditions of high information sharing. This suggests that informing employees helps to enhance their motivation to identify problems encountered and develop suggestions for improvement, and makes them more responsive to moral leadership. Information sharing is an important boundary condition for a high moral leader to encourage employees to input and share their opinions.

Third, this study extends the cultural perspective on leadership and voice literature in a non-western context. Although the influence of Confucian ideology on Chinese leadership is well-established, researchers have questioned whether this leadership style works well in modern business organizations (Farh and Cheng, 2000). This study examines the impact of PL behaviors on employee voice, with the expectation that this may be different from the impact of western leadership behaviors, such as transformation leadership (Detert and Burris, 2007; Liu et al., 2010). Friedman et al. (2006) suggested that high power distance in a society is significant in determining the general level of employee voice behaviors. It contributes to the understanding of employee voice issues by examining the impact of an indigenous Chinese leadership style – PL on employee voice.

Managerial implications

Although it is evident that many managers have a PL style when managing their employees, it was not known how this leadership style works in modern organizations and how it influences voice behaviors. To create an environment that is open to voice, an appropriate leadership behavior is the key consideration. This study fills this gap in our knowledge and provides new insights on the likely effects of PL in the workplace. The results of the present study suggest that to encourage employee voice a manager should avoid the style of authoritarianism and maintain a high level of morality.

A good working environment is one where employees are willing to express their views and offer their ideas and suggestions openly (Withey and Cooper, 1989). The findings show that authoritarian leadership may hinder and moral leadership may stimulate routine employee voice in the workplace. The results provide insights for managers who are eager to receive suggestions from employees by demonstrating the influence of leadership style. Authoritarian leadership has long been prevalent in Chinese society (Cheng et al., 2002; Wu et al., 2002) and we now know that it has a negative effect on employee voice. At the same time, it appears that moral leadership works to facilitate employee voice behaviors, suggesting that a high degree of moral leadership is generally welcomed by employees. Thus, it is desirable for managers to set a high ethical culture and maintain a high moral standard in dealing with their employees.

In addition, managers may encourage voice activities by sharing information with employees to create an open communication channel. Practices such as holding open discussions about specific work tasks may help employees to share their thoughts. Managers may act as a moral leader in providing constructive feedback and maintaining a high integrity perception of the workplace to support the openness of receiving information, opinions and expectations. The willingness of employees to exercise voice may be more likely under moral styles of managers when information is provided. Managers are accountable to be good role models as well as to practice information sharing, and as such, with these actions, employee voice behavior is encouraged.

Limitations and future research

This study has several limitations. First, given that the design of this study is cross-sectional in nature, this study does not allow conclusions regarding causality between PL behaviors and employee voice. As the relationships found are correlational and provide no evidence of the direction of the relationships, this study cannot rule out the possibility of reverse causation. One may argue that employee voice may reversely influence leadership behaviors, such as authoritarian leadership. Future study could perform a cross-lagged panel analysis in a longitudinal study to replicate the findings.

Second, the data were collected from a convenience sample from Chinese employees, which may raise concerns about the generalizability of the findings (Farh et al., 1987). Employees tend to place more emphasis on the relational aspects of PL behaviors in cultures with high power distance and collectivism (Hofstede, 2001). For example, in China, authoritarianism may not always be seen as unacceptable by employees, as they may respect an authoritarian leader due to their traditional value of obedience. This suggests that Chinese employees may be more willing than employees in western cultures to tolerate an authoritarian leader. Of course, these findings may be culture specific, so that it would be beneficial for future researchers to extend this line of research in other cultural contexts.

Third, the findings of PL behaviors on employee voice may be overstated if there are unmeasured variables that should have been controlled. Future research should pay attention to control variables that are missing to examine the impact on employee voice, for example, leader–member exchange (Burris et al., 2008; Van Dyne et al., 2008).

Fourth, Van Dyne and LePine (1998) suggested that the best condition for investigating voice is where the work is ‘dynamic and new ideas facilitate continuous improvement’ (p. 109). The results of this study may be limited by the nature of the business sample. For example, the effect of leadership behaviors on employee voice is highly valued in a strong context for leadership. Future research should extend the results to other types of employees and work, such as those in changing technology and complex situations; these might be even more likely to be engaged in employee voice (Brockner et al., 1998).

Lastly, this study did not provide a comprehensive model of all antecedents of employee voice. There is a research gap to strengthen the mediating mechanism of PL behaviors on employee voice, such as conflict and personal control (e.g. Tangirala and Ramanujam, 2008) and organization-based self-esteem (Chan et al., 2013). Also, it is plausible that there are other boundary conditions that may trigger the impact of PL behaviors on employee voice. The effects for antecedents of employee voice, such as trust in leader and empowering leadership behaviors, are complex and need the attention of researchers. Future research might continue to explore the dynamic relationships among the contextual variables, PL behaviors and employee voice.

To conclude, research on PL behaviors and employee voice needs to be continued and to provide guidelines for future work. This study provides new insights for the impact of PL behaviors on employee voice. Authoritarian leadership was negatively, and moral leadership positively, related to employee voice. This study also adds to the literature by examining whether information sharing enhances the effects. The positive relationship between moral leadership and employee voice was stronger when employees received higher levels of information sharing. It highlights the fact that more research work can be done to contribute to the PL literature.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Simon CH Chan is a Teaching Fellow at the Department of Management and Marketing, Faculty of Business, The Hong Kong Polytechnic University. His research interests focus on leadership and organizational behavior. He has published research papers in Journal of Organizational Behavior, Asia Pacific Journal of Management, International Journal of Human Resources Management, as well as other management journals. [Email: simon.ch.chan@polyu.edu.hk]

References

Arnold

Arad

Rhoades

Drasgow

(2000) The empowering leadership questionnaires: The construction and validation of a new scale for measuring leader behaviors. Journal of Organizational Behavior 21(3): 249–269.

Aryee

Chen

Sun

Debrah

(2007) Antecedents and outcomes of abusive supervision: Test of a trickle-down model. Journal of Applied Psychology 92(1): 191–201.

Avery

Quinones

(2002) Disentangling the effects of voice: The incremental roles of opportunity, behavior, and instrumentality in predicting procedural fairness. Journal of Applied Psychology 87(1): 81–86.

Aycan

(2006) Paternalism: Towards conceptual refinement and operationalization. In: Yang

Hwang

Kim

(eds) Indigenous and Cultural Psychology: Understanding People in Context. New York: Springer, 445–466.

Bass

(1985) Leadership and Performance Beyond Expectations. New York: Free Press.

Bass

(1997) Does the transactional-transformational leadership paradigm transcend organizational and national boundaries? America Psychologist 52(2): 130–139.

Blau

(1964) Exchange and Power in Social Life. New York: Wiley.

Brislin

(1990) Applied cross-cultural psychology: An introduction. In: Brislin

(ed.) Applied Cross-Cultural Psychology. Newbury Park, CA: SAGE, 9–33.

Brockner

Heuer

Siegel

Wiesenfeld

Martin

Grover

Reed

Bjorgvinsson

(1998) The moderating effect of self-esteem in reaction to voice: Converging evidence from five studies. Journal of Personality and Social Psychology 75(2): 394–407.

10.

Brockner

Spreitzer

Mishra

Hochwarter

Pepper

Weinberg

(2004) Perceived control as an antidote to the negative effects of layoffs on survivors’ organizational commitment and job performance. Administrative Science Quarterly 49(1): 76–88.

11.

Burns

(1978) Leadership. New York: Harper & Row.

12.

Burris

Detert

Chiaburu

(2008) Quitting before leaving: The mediating effects of psychological attachment and detachment on voice. Journal of Applied Psychology 93(4): 912−922.

13.

Chan

Simon, Huang

Snape

Lam

(2013) The Janus face of paternalistic leaders: Authoritarianism, benevolence, subordinates’ organization-based self-esteem, and performance. Journal of Organizational Behavior 34(1): 108–128.

14.

Chen

Eberly

Chiang

Farh

Cheng

(in press) Affective trust in Chinese leaders: Linking paternalistic leadership to employee performance. Journal of Management.

15.

Cheng

Chou

Farh

(2000) A triad model of paternalistic leadership: The constructs and measurement. Indigenous Psychological Research in Chinese Societies 14: 3–64. (In Chinese.)

16.

Cheng

Chou

Huang

Farh

(2004) Paternalistic leadership and subordinate reverence: Establishing a leadership model in Chinese organizations. Asian Journal of Social Psychology 7(1): 89–117. (In Chinese)

17.

Cheng

Huang

Chou

(2002) Paternalistic leadership and its effectiveness: Evidence from Chinese organizational teams. Journal of Psychology in Chinese Societies 3(1): 85–112.

18.

Cheng

Shieh

Chou

(2002) The principal’s leadership, leader–follower exchange quality, and the teacher’s extra-role behavior: The effects of transformational and paternalistic leadership. Indigenous Psychological Research in Chinese Societies 17: 105–161. (In Chinese)

19.

Cremer

(2006) Affective and motivational consequences of leader self-sacrifice: The moderating effect of autocratic leadership. The Leadership Quarterly 17(1): 79–93.

20.

Detert

Burris

(2007) Leadership behaviour and employee voice: Is the door really open? Academy of Management Journal 50(4): 869–884.

21.

Erben

Güneser

(2008) The relationship between paternalistic leadership and organizational commitment: Investigating the role of climate regarding ethics. Journal of Business Ethics 82(4): 955–996.

22.

Farh

Cheng

(2000) The paternalistic leadership of Chinese organization: An analysis of culture perspective. Local Psychology Research 13(1): 1–54.

23.

Farh

Cheng

Chou

Chu

(2006) Authority and benevolence: Employees’ responses to paternalistic leadership in China. In: Tsui

Bian

Cheng

(eds) China’s Domestic Private Firms: Multidisciplinary Perspectives on Management and Performance. New York: Sharpe, 230–260.

24.

Farh

Earley

Lin

(1997) Impetus for action: A cultural analysis of justice and organizational citizenship behavior in Chinese society. Administrative Science Quarterly 42(3): 421–444.

25.

Farh

Liang

Chou

Cheng

(2008) Paternalistic leadership in Chinese organizations: Research progress and future research directions. In: Chen

Lee

(eds) Business Leadership in China: Philosophies, Theories, and Practices. Cambridge: Cambridge University Press, 171–205.

26.

Farh

Podsakoff

Cheng

(1987) Culture-free leadership effectiveness versus moderators of leadership behavior: An extension and test of Kerr and Jermier’s substitutes for leadership model in Taiwan. Journal of International Business Studies 18: 43–60.

27.

Friedman

Chi

Liu

(2006) An expectancy model of Chinese-American differences in conflict-avoiding. Journal of International Business Studies 37(1): 76–91.

28.

Gao

Janssen

Shi

(2011) Leader trust and employee voice: The moderating role of empowering leader behaviors. Leadership Quarterly 22(4): 787–798.

29.

Gelfand

Erez

Aycan

(2007) Cross-cultural organizational behavior. Annual Review of Psychology 58(1): 479–514.

30.

Gigone

Hastie

(1993) The common knowledge effect: Information sharing and group judgment. Journal of Personality and Social Psychology 65(5): 959−974.

31.

Gorden

Infante

Graham

(1988) Corporate conditions conducive to employee voice: A subordinate perspective. Employee Responsibilities and Rights Journal 1(2): 101–111.

32.

Henry

(1995) Improving group judgement accuracy: Information sharing and determining the best member. Organizational Behavior and Human Decision Processes 62(2): 190–197.

33.

Hirschman

(1970) Exit, Voice, and Loyalty: Responses to Decline in Forms, Organizations, and States. Cambridge, MA: Harvard University Press.

34.

Hofstede

(2001) Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations. Thousand Oaks, CA: SAGE.

35.

Huang

Van

Vliert

Van der Vegt

(2005) Breaking the silence cultures: Stimulation of participation and employee opinion withholding cross-nationally. Management and Organization Review 1(3): 459–482.

36.

Janssen

Lam

Huang

(2010) Emotional exhaustion and job performance: The moderating roles of distributive justice and positive affect. Journal of Organizational Behavior 31(6): 787–809.

37.

Lam

Schaubroeck

(2011) Information sharing and group efficacy influences on communication and decision quality. Asia Pacific Journal of Management 28(3): 509–528.

38.

Lee

Allen

(2002) Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. Journal of Applied Psychology 87(1): 131–142.

39.

LePine

Van Dyne

(1998) Predicting voice behaviors in work groups. Journal of Applied Psychology 83(6): 853–868.

40.

LePine

Van Dyne

(2001) Voice and cooperative behavior as contrasting forms of contextual performance: Evidence of differential relationship with Big Five personality characteristics and cognitive ability. Journal of Applied Psychology 86(2): 326–336.

41.

Ling

Kellermanns

(2010) The effects of family firm specific sources of TMT diversity: The moderating role of information exchange frequency. Journal of Management Studies 47(2): 322–344.

42.

Liu

Zhu

Yang

(2010) I warn you because I like you: Voice behavior, employee identifications, and transformational leadership. Leadership Quarterly 21(1): 189–202.

43.

McEvoy

Cascio

(1989) Cumulative evidence of the relationship between employee age and job performance. Journal of Applied Psychology 74(1): 11–17.

44.

Milliken

Morrison

Hewlin

(2003) An exploratory study of employee silence: Issues that employees don’t communicate upward and why. Journal of Management Studies 40(6): 1453–1476.

45.

Morrison

Milliken

(2000) Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Review 25(4): 705–725.

46.

Morrison

Wheeler-Smith

Kamdar

(2011) Speaking up in groups: A cross-level study of group voice climate and voice. Journal of Applied Psychology 96(1): 183–191.

47.

Moye

Langfred

(2004) Information sharing and group conflict: Going beyond decision making to understand the effects of information sharing on group performance. The International Journal of Conflict Management 15(4): 381−410.

48.

Muller

Judd

Yzerbyt

(2005) When moderation is mediated and mediation is moderated. Journal of Personality and Social Psychology 89(6): 852–863.

49.

Pellegrini

Scandura

(2006) Leader-member exchange (LMX), paternalism and delgration in the Turkish business culture: An empirical investigation. Journal of International Business Studies 37(2): 264–279.

50.

Pellegrini

Scandura

(2008) Paternalistic leadership: A review and agenda for future research. Journal of Management 34(3): 566–593.

51.

Pellegrini

Scandura

Jayaraman

(2010) Cross-cultural generalizability of paternalistic leadership: An expansion of leader-member exchange theory (LMX). Group and Organization Management 35(4): 391–420.

52.

Podsakoff

MacKenzie

Moorman

Fetter

(1990) Transformational leader behaviors and their effects on followers’ trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly 1(2): 107–142.

53.

Quinn

Spreitzer

(1997) The road to empowerment: Seven questions every leader should consider. Organizational Dynamics 26(2): 37–49.

54.

Randel

Ranft

(2007) Motivations to maintain social ties with coworkers: The moderating role of turnover intentions on information exchange. Group & Organization Management 32(2): 208–232.

55.

Renzl

(2008) Trust in management and knowledge sharing: The mediating effects of fear and knowledge documentation. Omega 36(2): 206–220.

56.

Schweiger

DeNisi

(1991) Communication with employees following a merger: A longitudinal field experiment. Academy of Management Journal 34(1): 110–135.

57.

Sinclair

Martin

Michel

(1999) Full-time and part-time subgroup differences in job attitudes and demographic characteristics. Journal of Vocational Behavior 55(3): 337–357.

58.

Soylu

(2011) Creating a family or loyalty-based framework: The effects of paternalistic leadership on workplace bullying. Journal of Business Ethics 99(2): 217–231.

59.

Stamper

Van Dyne

(2001) Work status and organizational citizenship behavior: A field of restaurant employees. Journal of Organizational Behavior 22(5): 517–536.

60.

Stasser

Stewart

Wittenbaum

(1995) Expert roles and information exchange during discussion: The importance of knowing who knows what. Journal of Experimental Social Psychology 31(3): 244–265.

61.

Stasser

Titus

(1987) Effects of information load and percentage of shared information on the dissemination of unshared information during group discussion. Journal of Personality and Social Psychology 53(1): 81–93.

62.

Tangirala

Ramanujam

(2008) Exploring nonlinearity in employee voice: The effects of personal control and organizational identification. Academy of Management Journal 51(6): 1189–1203.

63.

Tangirala

Ramanujam

(2012) Ask and you shall hear (But not always): An examination of the relationship between manager consultation and employee voice. Personnel Psychology 65(2): 251–282.

64.

Thibaut

Walker

(1975) Procedural Justice: A Psychological Analysis. Hillsdale, NJ: Erlbaum.

65.

Van Dyne

LePine

(1998) Helping and voice extra-role behaviour: Evidence of construct and predictive validity. Academy of Management Journal 41(1): 108–119.

66.

Van Dyne

Cummings

McLean Parks

(1995) Extra-role behaviors: In pursuit of construct and definitional clarity. Research in Organizational Behavior 17: 215–285.

67.

Van Dyne

Kamdar

Joireman

(2008) In-role perceptions buffer the negative impact of low LMX on helping and enhance the positive impact of high LMX on voice. Journal of Applied Psychology 93(6): 1195–1207.

68.

Venkataramani

Tangirala

(2010) When and why do central employees speak up? An examination of mediating and moderating variables. Journal of Applied Psychology 95(3): 582–591.

69.

Walumbwa

Morrison

Christensen

(2012) Ethical leadership and group in-role performance: The mediating roles of group conscientiousness and group voice. Leadership Quarterly 23(5): 953–964.

70.

Wang

Cheng

(2010) When does benevolent leadership lead to creativity? The moderating role of creative role identity and job autonomy. Journal of Organizational Behavior 31(1): 106–121.

71.

Westwood

(1997) Harmony and patriarchy: Then cultural basis for ‘paternalistic headship’ among the overseas Chinese. Organization Studies 18(3): 445−480.

72.

Westwood

Chan

(1992) Headship and leadership. In: Westwood

(ed.) Organizational Behaviour: A Southeast Asian Perspective. Hong Kong: Longman, 123–139.

73.

Withey

Cooper

(1989) Predicting exit, voice, loyalty, and neglect. Administrative Science Quarterly 34(4): 521–539.

74.

Hsu

Cheng

(2002) Expressing or suppressing anger: Subordinate’s anger responses to supervisors’ authoritarian behaviors in a Taiwan enterprise. Indigenous Psychological Research in Chinese Societies 18: 3–49.

75.

Zhou

George

(2001) When job dissatisfaction leads to creativity: Encouraging the expression of voice. Academy of Management Journal 44(4): 682–696.