Power Congruence Approaches to Downward Influence in Malaysia Manufacturing Industry

It has been proposed that issues of power and influence should be incorporated in the dyadic study of leadership. Building upon social exchange theory which explains how power is gained and lost in the process of influence between an agent and the target (Blau, 1964; Emerson, 1962), this paper hypothesized that the power congruence of supervisors and subordinates was anchored on 9 types of influence tactics. Three hundred and eighty-five pairs of Malaysian managers and executives working in manufacturing companies voluntarily participated in this study. The findings suggested that supervisors would use a combination of hard and soft tactics when their subordinates perceived them to have position or personal power. Implications of the findings, potential limitations of the study, and directions for future research were further discussed.


Introduction
Social power relationship occurs naturally when people with different levels of potential power interact to accomplish organizational goals (Mossholder, Kemery, Bennett, & Wesolowski, 1998).Theorists such as Barnard (1938), Blau (1964), Mechanic (1962), and Weick (1969) contended that supervisors' relationship with subordinates represented an important source of power to exert influence in their organizations.Blau's (1964) social exchange theory was built upon Gouldner's (1960) norm of reciprocity which stated that: "(i) people should help those who have helped them, and (ii) people should not injure those who have helped them" (Gouldner, 1960, p. 171).Past studies on leadership have not found conclusive evidence on Malaysian leadership style.For example, Gill (1998) suggested that Malaysian managers are found to be more direct, less delegate, and are more transactional.However, Govindan (2000) reported that Malaysian leaders lean more towards participative and consultative styles.This is in line with the assertion of Abdullah (1992) that the use of stronger tactics in Malaysian context is not likable as Malaysians generally are not in favor of overt display of anger and aggressive behavior.
Despite the enormous breadth of the literature on the relevance of power to organizational influence in general, and to an understanding of leadership in particular, research studies of power and influence are not well integrated.This is evidenced where the research on influence strategies that superiors use to translate power into actual influence is relatively recent (Ansari, 1990;Farrell & Schroder, 1999;Hinkin & Schriesheim, 1990;Somech & Drach-Zahavy, 2002;Venkatesh, Kohli, & Zaltman, 1995).In addition to that, no known researches have focused specifically on the factors that contribute to the interaction effects between the supervisors' self-perceptions of power and their subordinates' perceptions of them and vice versa, and their impact on supervisors' influence tactics.Even though the concept of fit has served as an important building block in several areas of research, there is still a lack of corresponding schemes by which fit has been tested (Venkatraman, 1989).Hence this study has two objectives in mind.First is to examine whether congruence between supervisors' perception of subordinates' power and subordinates' self-reported power predict the use of influence tactics and secondly, whether the congruence between supervisors' self-report power and subordinates' perception of supervisors' power predict the use of influence tactics.

Power and downward influence
Extensive research is available in the organizational behavior literature investigating the process of interpersonal influence and social power.But the two constructs--power and influence seem to have been examined almost independently.A few researchers (e.g., Yang et al., 1998;Yang & Cervero, 2001) have used the two terms--power and influence--interchangeably.For example, Mechnic (1962) considered power, influence, and control as synonymous.In the same vein, other researchers (e.g., Salancik & Pfeffer, 1977;Tannenbaum, 1968) have defined power and influence as synonyms and used interchangeably, where the important element is control which can be exercised through formal authority.
Nonetheless, most of the researchers agree that power and influence are two distinct terms, where power refers to the potential influence, and influence as the actual use of power.It is reported that manager's effectiveness depends on his or her success in influencing others (Bass, 1990;Kotter, 1985).In other words, leadership is the ability of an individual to influence, motivate, and make others contribute towards the effectiveness and success of their organizations.Yukl and Van Fleet (1982) revealed that supervisory power is important not only for influencing subordinates, but can be used to influence peers, superiors, and also outsiders.That is, little research has been done to examine the relationship between bases of power and the use of downward influence tactics in organizations (Ansari, 1990;Hinkin & Schriesheim, 1990).This is particularly true in the Malaysian context.Kipnis (1976) contended that people who have a strong need for power and the ability to gain control of resources are likely to experience a desire to influence others.As a result, those being influenced will be gradually devalued, with the powerful individuals preferring to maintain social and psychological distance.The power and influence approach attempts to define leadership effectiveness in terms of the power as possessed by leaders, the types of power, and how power is exercised (Yukl, 2005).Similarly, leaders should employ different strategies in order to gain power through formal authority, reputation, and performance.It was found that power can be obtained through formal authority or informal coalitions with political influence (Pfeffer, 1992).Hence, this approach incorporates issues of change in the organizations and also accumulation and loss of power.Kipnis, Schmidt, and Wilkinson (1980) crafted the original taxonomy of influence tactics and this was subsequently modified by Kipnis and Schmidt (1983) and their eventual development of a commercial version of their questionnaires known as the Profile of Organizational Influence Strategies (POIS) was developed.In the course of their research on intraorganizational influence tactics, Kipnis et al. initially have identified a large number of influence tactics, which they later factor analyzed, reducing the number of subscales into seven dimensions.The tactics that were identified are (i) reason, (ii) friendliness, (iii) coalition, (iv) bargaining, (v) assertiveness, (vi) higher authority, and (vii) sanctions.Beginning with the landmark work of Kipnis et al., other experts have also begun to examine the interpersonal influence processes in organizations (Ansari & Kapoor, 1987;Erez & Rim, 1982;Erez, Kim, & Keider., 1986;Schriesheim & Hinkin, 1990).
It was found that various factors such as organizational, personal, and situational factors have contributed to the success of influence style (Kipnis, Schmidt, Swaffin-Smith, & Wilkinson, 1984).On a similar note, when influencing subordinates, supervisors were more likely to use assertiveness, friendliness, bargaining, and higher authority.Managers were found to vary their style of influence depending on the discrepancy in power between them and the target, how successfully they want to influence their subordinates, and also the number of reasons managers had for exercising influence.Another measure which is worth mentioning is the Influence Behavior Questionnaire (IBQ) designed by Yukl and Falbe (1990).The IBQ is a reconceptualization of Kipnis et al.'s (1980) earliest framework and the addition of two new tactics namely, inspirational appeals and consultation.Yukl and his associates (Falbe & Yukl, 1992;Yukl, Kim, & Falbe, 1996;Yukl & Tracey, 1992) later developed the most comprehensive taxonomy of influence tactics, consisting of nine influence tactics (Cable & Judge, 2003).The nine influence tactics are (i) pressure tactics, (ii) exchange tactics, (iii) coalition tactics, (iv) ingratiating tactics, (v) rational persuasion, (vi) inspirational appeals, (vii) consultation tactics, (viii) legitimating tactics, and lastly (ix) personal appeals.Even though Yukl's taxonomy was developed from Kipnis', it is claimed that the additional tactics included made the taxonomy more suitable for the use of managers.In fact, Yukl has further classified the outcomes of influence efforts into three qualitative groups known as commitment, compliance, and resistance.Kelman (1974) contended that this influence process relies upon the target's need for acceptance or esteem to gain compliance.
Further, other studies (Ansari, 1990;Bhal & Ansari, 2000) have discovered another three new tactics known as showing expertise, personalized help, and instrumental dependency in Indian settings.In addition to that, some researchers have extended the influence tactics into 11 influence tactics that are relevant to managers (Yukl & Seifert, 2002).According to their study, the four core tactics--rational persuasion, inspirational appeals, consultation, and collaboration are most likely to result in commitment by the target person, whereas other tactics such as exchange, pressure, or coalition were more likely to result in either compliance or resistance.
Previous studies have examined the directional differences in influence behavior (Ansari, 1990;Bennebroek & Boonstra, 1998;Bhal & Ansari, 2000;Kipnis et al., 1980;Yukl & Falbe, 1990;Yukl, Falbe, & Youn., 1993;Yukl & Tracey, 1992).Their findings have shown that the use of influence tactics is connected to the hierarchical relationship between the agent and target.The results of Burke (1986) and House, Filley, and Gujarati (1971) indicated that, downward power depended on the individual's upward power and lateral power where it could be seen as combining to influence an individual's total personal power.This implies that managers would adopt different influence tactics depending on whether they want to influence a subordinate, a colleague, or a supervisor.Kipnis and Schmidt (1985) introduced the idea of three basic intraorganizational influence dimensions known as hard, soft, and rational influence tactics.This classification of influence tactics has since been adopted by other researchers such as Deluga (1991), Falbe and Yukl (1992), and Thacker and Wayne (1995).

Sample
With an aim to generalize on supervisors and subordinates in Malaysia, the population of the present study consists of supervisors and subordinates currently employed in large scale manufacturing companies in Malaysia.Malaysia is gradually becoming a major destination for foreign investment, particularly in the manufacturing sector.Since the 1980's, Malaysia has started to diversify and transform its economy into a more industrialized one.This endeavor has been seeing continuing success, with manufacturing and high-tech industries now playing a significant role in shaping the economic expansion of the state.Currently, the manufacturing sector is considered as one of the cornerstone of Malaysia's economic diversification strategy.As the aim of this research is to study the relationship between supervisors' and subordinates' power, and their choice of influence tactics for influencing their subordinates, the unit of analysis is at individual level that is the individuals that make up the supervisor-subordinate dyads.1,300 pairs of questionnaires were distributed to from executives working at 82 manufacturing companies that participated in this study, only 385 pairs were usable for analysis.43 companies (53%) from the industrial sector formed the majority of participating companies, followed by the consumer sector with 27 (32.5%)companies, and the construction sector with 12 (14.5%)companies.Of the sample, most of the companies were predominantly Malaysian companies, 52 (63.4%).Others were American companies, 17 (20.7%);Japanese companies, 6 (7.3%);European companies, 2 (2.5%); and others (consisting of Taiwanese, Korean, and Singaporean based companies), 5 (6.1%).

Measures
Forty-four items arranged in random order in the questionnaires were used to measure the 9 influence tactics employed by supervisors in their attempts to influence their subordinates.Those items used as influence tactics were not gathered from one source, but were taken from different researches (Ansari, 1990;Ansari & Kapoor, 1987;Bhal & Ansari, 2000;Kipnis et al., 1980;Schriesheim & Hinkin, 1990) and were grouped into three known as hard, soft, and rational tactics.Subjects were specifically cautioned not to make any judgment about the desirability or undesirability of the behaviors described and to merely report the frequency of attempts made by their supervisors to change their mind or opinion in the workplace.The subordinates were asked to describe on a 7-point Likert scale with the following anchors: never (1), almost never (2), seldom (3), sometimes (4), usually (5), almost always (6), always (7).
The five bases of power taxonomy as proposed by French and Raven, namely, reward, coercive, legitimate, referent, and expert power were used in this study as it still appears to be fairly representative and popular in application (Ansari, 1990;Frost & Stahelski, 1988;Rahim, 1989;Rahim, Antonioni, Krumov, & Ilieva., 2000).In addition, the items for another two power bases, namely information and connection, were adapted from the works by Ansari (1990) and Bhal and Ansari (2000).For the purpose of this study, the seven bases of power were further grouped under position and personal power.There were all together 35 items in the power bases section.Though these 35 items measured the same power dimensions, two versions had to be developed, one for the supervisors and one for the subordinates.Both versions had the same items but minor changes in wording were needed to make the questionnaire suitable for respondents in describing their own and their targets' power.Respondents were asked to rate their perception using the 7-point Likert scale with: 7 = strongly agree, 6 = agree, 5 = slightly agree, 4 = neutral, 3 = slightly disagree, 2 = disagree, and 1 = strongly disagree, for all the 35 items.A higher score indicates greater bases of the supervisors' or subordinates' power.

Analysis and results
The reliability of the scales was assessed through Cronbach's coefficient alpha.The reliability coefficients and zero-order correlations among study variables are contained in Table 1.
Generally, the values indicating intercorrelations among the predictors variables were low, ranging from .13 to .43.Only 17 out of 28 intercorrelation factors were statistically significant, thus indicating the independence of the scales used for measuring the predictors.In addition to that, a number of the predictor variables were noted to be significantly correlated to the criterion variables ranging from -.27 to .52, which were considered as low intercorrelation values.
In order to test the formulated hypotheses, a series of hierarchical multiple regression were conducted.The hierarchical regression equations consisted of three models.In the first model on each equation gender was entered as the control variable.It was necessary to control gender to ensure clearer, credible results as previous studies on the effect of gender the criterion variable had been inconsistent.The equation for the second model consists of the four perspectives (two each from both supervisors and subordinates) for the independent variables, namely position power and personal power.This is followed by the interaction terms of position and personal power from each perspective for simultaneous equations.Three separate hierarchical multiple regressions analyses were conducted for each of the dependent variables namely, hard, soft, and rational influence tactics.Results for the interaction effects are contained in Table 2.
The results have indicated that hard tactic will be used when subordinates' perception of personal power congruent with supervisors' perception of subordinates personal power ( = 1.06, p < .05).On the other hand, when supervisors self reported position power which interacted with subordinates' perception of supervisors' position power ( = 1.05, p < .01),soft tactics will be applied.
As illustrated in Figure 1, when supervisors perceived themselves as having low and high position power, and at the same time subordinates perceived the supervisors as having low to moderate power, then there is minimal impact on the use of soft tactics.In contrast, supervisors would apply more soft tactics to subordinates who perceived them as having high position power.This indicates that supervisors who perceived themselves to have high position power would use soft influence tactics more frequently compared to supervisors who viewed themselves to have low or moderate position power.
In the cases where supervisors' perceptions were congruent with subordinates self-report that subordinates only possessed low to moderate personal power, supervisors use of hard tactics registered negligible changes.On the other hand, the use of hard tactics was more prominent when supervisors' perceived subordinates as possessing high personal power.The statistically significant interaction effect ( = 1.06, p < .05)are depicted in Figure 2.
Nevertheless, when subordinates' self-reported personal power reached above moderate levels, the use of hard influence tactics decreased drastically if supervisors' perceived their subordinates as having personal power regardless of whether it was low, moderate, or high.Thus, it can be concluded that if subordinates were perceived to hold high personal power, supervisors would apply hard tactics on them when at the same time; subordinates viewed themselves as possessing moderate and below personal power.

Discussion and conclusions
Contrary to expectation, soft tactics appeared to be the most commonly used tactics when supervisors' perceived themselves and were also perceived by subordinates to have position power.The rationale could be due to the fact that supervisors were aware that the attribution of success using the softer influence tactics such as personalized help and ingratiation could contribute to better interpersonal relations and help gain the support of members (Tandon, Ansari, & Kapoor, 1991).Subordinates were often found to respect the hierarchy of social positions in effect in the organization and hence were more receptive to instructions, thus guaranteeing compliance (Chambon, 2005) without the need to use hard influence tactics.In addition, using the power-dependence theory, Sell, Lovaglia, Mannix, Samuelson, and Wilson (2004) explained that the more dependent an individual is on a social relationship, the less power that individual has.
As noted by Somech and Drach-Zahavy (2002), previous studies have shown that subordinates responded more favorably to personal power rather than positional power, as supervisors who used personal power were more likely to make the target feel useful and appreciated.On a similar note, Abdullah (1996) endorsed that Malaysian managers like to create a sense of "family atmosphere" among their employees as this will instill in them a sense of commitment to the organization.
Surprisingly, in the case of personal power, when the subordinates' self perception co-incided with the supervisors' perceptions of the subordinates', supervisors would resort to the usage of hard influence tactics.This could probably be due to the fact that supervisors were reluctant to admit or recognize the referent or expertise power bases of their subordinates, and in such situations those power cannot be translated into influence on the supervisors (Rodrigues & Lloyd, 1998).As noted by Greene and Podsakoff (1981), when the leaders were under pressure to maintain high-productivity deadline, they would tend to disregard their subordinates' power and judged the targets as unworthy and hence use more hard tactics (Raven, 1993).
Rational appeal did not seem to be related to any power congruence variable.According to Yukl and Tracey (1992), studies found that rational persuasion and consultation are the most effective tactics for influencing peers.This is further supported by Brass and Burkhardt (1993) who stated that this influence tactic appeared to be "universal" and were unaffected by structural factors.Besides, others researchers such as Tepper, Brown, and Hunt (1998) indicated that subordinates regarded managers' use of rational influence tactics to be more objectionable only if it were used together with soft influence tactics.

Limitations and future research
As with any research, this study has several limitations.This current study has relied primarily on samples drawn specifically from the workers in the manufacturing sectors.Besides, the data were collected from both the subordinates and their supervisors at a single point in time, hence the direction of causality cannot be determined.Clearly, a longitudinal approach would have placed researcher in a better position to draw causal conclusions.Therefore, only conclusions or discussions of the general relationships between the variables of interest could be drawn.

Practical considerations
The study has important implications for practicing managers.It is believed that by understanding their subordinates' power bases, supervisors when they become targets themselves can gain greater insights into the tactics used by their subordinates and thus avoid negative outcomes.In other words, supervisors would experience fewer conflicts and misunderstandings if they realized that some of the "confrontational" influence styles used by some subordinates are actually the results of power differences rather than personal hostility.It is believed that by sharing power with subordinates, it may be possible to expand the total amount of power available to the organization.More importantly, it allows management to control and accurately predict employees' behaviors and attitudes by using the appropriate influence tactic or combination tactics in tandem with the power bases for successful targeted outcomes such as compliance or commitment.This suggests that increasing a leader's awareness of how his or her self-perceptions compared to the subordinates' perceptions can lead to greater agreement.Note.Per_lm=supervisors report on subordinates' personal power; per_mm=subordinates self assessment of personal power.

Figure 1 .
Figure 1.Interaction between supervisors' self-reported and subordinates' perception of supervisors' position power for soft influence tactics.Note.Pos_ml=subordinates report on supervisors' position power; pos_ll=supervisors self assessment of position power.

Figure 2 .
Figure 2. Interaction between subordinates' self-reported and supervisors' perception of subordinates' personal power on hard tactics.