Dealer ’ s Legitimate Power and Relationship Quality in Gaunxi Distribution Channel : A Social Rule System Theory Perspective

Relative to the volume of research into B2B channel power effects on relationship outcomes and relationship quality in gaunxi distribution channel, individual channel power effects have received much less attention than plodded channel powers. To tackle this imbalance, this study examines the direct effects of dealer’s legitimate power on four dimensions of relationship quality (cooperation, relationship commitment, relationship satisfaction and trust) from the perspective of small and medium enterprise (SME) manufacturers. In order to understand how gaunxi distribution channels operate, social rule system theory is utilized. Four hypothesized relationships are tested with a sample of 400 manufacturing SMEs in Taiwan. The results suggest that dealer’s legitimate power positively influences SME manufacturers’ cooperation, relationship satisfaction, relationship commitment and trust in a significant way. Managerial implications of the findings are discussed and limitations and future research directions are indicated.


Introduction
While some researchers who examined the use and effects of channel power by firms in marketing channels indicate that non-coercive power positively influences channel relationship outcomes (e.g.Banton and Maloni, 2005;Brown, Lusch, and Nicholson 1995;Maloni and Benton, 2000;Zhao, Huo, Flynn and Yeung, 2008), the influence of individual non-coercive powers such as legitimate power on relationship quality in gaunxi typified distribution channels has not been investigated.As a result, there appears to be a limited understanding of the extent to which and how legitimate power influences relationship quality in gaunxi distribution channels.In particular, how a major dealer's legitimate power influences relationship quality with Small and Medium Enterprise (SME) manufacturers in a distribution channel governed by gaunxi social rules has not received empirical scrutiny and academic inspection.Legitimate power is important in a Chinese culture such as one in Taiwan, where high power distance is savored (Zhuang and Zhou, 2004) and the firms with the most power in gaunxi distribution channels are more venerated (Lee, 2001;Zhuang, Xi, and Tsang, 2010;Wong and Tam, 2000;Zhao, Huo, Flynn, and Yeung, 2008).For that reason, this channel power warrants further empirical investigation in relation to relationship quality in a setting dominated by Chinese culture.Besides, a rich and growing stream of studies highlight various important advantages accruing to both manufacturing and dealer firms from forging high quality business relationships for example, improved long term profits (Skarmeas, Katsikeas, Spyropoulou and Salehi-Sangari, 2008).Therefore, the need to understand how and the extent to which legitimate power influences important channel outcomes such as relationship quality in distribution channels governed by gaunxi social rules are acutely felt, hence this void is noteworthy of address.
A review of the extant marketing channels literature substantiate a paucity of empirical research on the relationship between legitimate power and relationship quality dimensions namely cooperation, relationship commitment, relationship satisfaction and trust, in developing countries and particularly in B2B SME exchange relations (Chinomona, Lin, Wang and Cheng, 2010).One of the few exceptional prior studies that examined the influence of legitimate power was by Ketilson, (1991) who delved into the impact of this power in the cooperative retailing system in Canada, a Western culture dominated setting.It is hard to find studies on the same that verify or dispel their findings on legitimate power in a setting dominated by Chinese culture (Liu and Wang 2000;Zhao, Huo, Flynn and Yeung 2008;Zhuang and Zhou, 2004).More to the point, the few studies that have examined relationship quality dimensions in a Chinese setting tend to use the plodded coercive/non-coercive powers as predictors in their empirical investigation (e.g.Lee, 2001;Zhao, Huo, Flynn and Yeung, 2008).Such aggregation can be problematic and can lead to a loss of useful information because, as power sources are combined, the result is non-additive (Lusch, 1976;Leonidou, 2005).For instance, if expert power reduces identification and referent power reduces perception of difference, then the two power sources combined may cancel each other out (Leonidou, 2005).Above and beyond, large size firms provided the domain for the majority of these previous studies on relationship quality (Skarmeas, Katsikeas, Spyropoulou and Salehi-Sangari, 2008).Despite theoretical claims and empirical evidence pointing to the critical role of relationship quality in the effective management of inter-firm relationships in B2B settings (e.g.Bruggen, Kacker, and Nieuwlaat, 2005;Dorsch, Swanson, and Kelley 1998;Dwyer, Schurr, and Oh, 1987), scant attention has been paid to relationship quality development in an SME developing country context (Chinomona, Lin, Wang and Cheng, 2010).This is unfortunate, given that SMEs are regarded as the engines for economic development and employment generation in a developing country such as Taiwan (Chang, Yang, Cheng and Sheu, 2003;Chen, 1999).
It is against this background that the current study sets out to examine the direct effects of dealers' legitimate power on SME manufacturers' cooperation, trust, relationship satisfaction and relationship commitment in gaunxi distribution channel systems governed by social norms.In addition, an attempt is made in the current study to apply the social rule system theory borrowed from sociology discipline (Burns and Dietz, 1992) in order to understand how gaunxi distribution channels operate.To the best knowledge of the authors, this is the first attempt where this theory has been applied in marketing discipline.By so doing, insights are to be gained and a contribution is made to the developing marketing literature on gaunxi distribution channels hitherto not studied extensively in developing countries predominately of Chinese culture.Based on the aforementioned arguments, this study is therefore unique.
The rest of the article is organized as follows.A theoretical background, conceptual framework and hypotheses are provided.These are followed by the discussion of methodology, the constructs and scales used, and the analysis and conclusions are outlined.Finally managerial implications, limitations and future research directions are given.

Social Rule System Theory and Gaunxi Rules
As a theory rooted in sociology, social rule system theory asserts that most human social activities are organized and regulated by socially produced and reproduced systems of rules (Burns and Deitz, 1992).These rules are premised on societal customs, codes of conduct, norms and taboos, and have a tangible existence in social institutions such as family, community, and business enterprises (Burns and Flam, 1987;Flam and Carson, 2008).Thus, this theory posits that social rules in a system strongly influence actions and interactions between and among its members (Burns and Flam, 1987).According to Burns and Deitz (1992) social rules serve two main functions.On one hand social rules provide potential constraints or impose social sanctions on action possibilities that may be deemed negative and undesirable in a social institution.On the other hand they also generate opportunities for social actors to behave in ways that would otherwise be impossible, for instance, coordinating with others, mobilizing and gaining systematic access to strategic resources, allocating substantial human and physical resources, and solving complex social problems through collective actions.In guiding and regulating interaction, social rules give behavior recognizable, characteristic patterns, and make such patterns understandable and meaningful for those who share in the rule knowledge (Flam and Carson, 2008).Shared rules are the major basis for knowledgeable actors to derive, or to generate, similar situational expectations (Wang, Siu and Barnes, 2008).They also provide a frame of reference and categories, enabling participants to readily communicate about and to analyze social activities and events (Wang, 2007).In this regard, uncertainty is reduced and predictability is increased (Yau, Lee, Chow, Sin, and Tse, 2000).
In the context of Chinese gaunxi distribution system, social rules play a pivotal role and are adhered to for a variety of reasons.These include among others gaining mutual benefits and avoiding losses; maintaining identity, role and status; authoritative legitimacy; normative and cognitive order; social sanction avoidance; habits or routines, and uncertainty avoidance.According to Wang (2007) gaunxi social rules are accepted and adhered to because members tend to associate them with sacred principles that define their causal or symbolic relationship to members' interests and status.In other words, social rules fit into their cognitive frame for maintaining how things ought to be done.In such a case, gaunxi members react negatively to any deviance that disturbs the order, potentially destabilizes or erode it.Besides, failure by a gaunxi member to uphold the social rules and meeting gaunxi responsibilities will result in damaged prestige, loss of face, and subsequent loss of trust by other members of the gaunxi circle (Hwang, Golemon, Chen, Wang, and Hung, 2009).Thus, to save face gaunxi members abide by the social rules upon which gaunxi obligations and responsibilities are based (Wang, 2007).Yau, Lee, Chow, Sin and Tse (2000) identifies reciprocity, empathy, bonding and trust as the basic gaunxi social rules.Distribution channel members who share a gaunxi relationship in business are committed to each other by the social obligation of reciprocity in that if an exchanging partner receives a favor, he or she owes renqing to the benefactor and should be ready to pay back once circumstances permit (Hwang, 1987).Renqing, refers to one's emotional responses or a resource that one can present to another person as a gift in the social exchange process, and a set of social norms that one should follow to get along well with other people (Wang, 2007).If truth be told, what guides relational exchange behaviors in gaunxi distribution channels are the rules of reciprocal obligation and mutual assurance, which are also based on renqing (Hwang, Golemon, Chen, Wang, and Hung, 2009).In particular, gaunxi is cultivated and maintained through the exchange of renqing to attain mutual benefits.To maintain a healthy relationship punctuated by trust, cooperation, satisfaction and commitment, the Chinese people place a high value on the obligation of reciprocation of renqing (Hwang, 1987;Yang, 1994).The rule of empathy focuses on the benefactor's behavior that is guided by Confucianism principle of forgiveness, propounded by the ''do not do unto others that which you would not wish done unto you'' and by its converse: ''Do unto others as you wish done unto yourself'' (Hwang, 1987).Empathy is the ability to see a situation from another person's perspective.The greater the degree of empathy, the fewer are the barriers to the development of relationship and eventually the more the relationship satisfaction attained (Conway and Swift, 2000).As Hwang (1987) noted, a person who is versed in renqing is well equipped with empathy.In particular, a person is viewed to possess renqing if he or she has empathy towards others' emotions and feelings and is ready to help others who are in need.
The rule of bonding in gaunxi is embodied in "ganqing" which calls for emotional attachment between exchange partners within the gaunxi distribution channels.In a business relationship where no blood based attachment (qingqing) exists, one often does ganqing touzi, which means doing investment with affection to build a good gaunxi (Wang, 2007).Gaunxi works at a personal level on the basis of friendship and ganqing is a measure of the level of emotional commitment and the closeness of the parties involved (Wang, Siu and Barnes, 2008).From the business perspective, people who have ganqing care about each other and treat each other more like friends than just business partners (Hwang, Golemon, Chen, Wang, and Hung, 2009).Ganqing plays a key role in maintaining and enhancing a gaunxi relationship and makes the impersonal business relationship more personal (Yau, Lee, Chow, Sin, and Tse, 2000).Chinese people believe they can only rely on those they have ganqing to protect them in times of difficulties and to help them reach their goals in better time (Wang, 2007)."Xinyong" is another important concept of gaunxi relationship that refers to trust or imply one's credibility (Wang, Siu and Barnes, 2008).It is developed and based on a person's past history and reputation to keep his or her promise and fulfill an obligation.Those who have more connection and have stayed longer in gaunxi distribution channel are more trusted and often most end up yielding more legitimate power than others and therefore command respect.Consequently, other gaunxi channel members become committed to and are satisfied to cooperate with these gaunxi distribution channel leaders (Haley, Tan, and Haley, 1998).

Legitimate Power
In the current study legitimate power is taken to mean traditional legitimate power.Defined, traditional legitimate power is the perception or belief that a leader has the right to exercise influence because of the leader's role or position and that such influence should be accepted (Mullins, 1999;Benton and Maloni, 2005).This power is derived from formal or informal social norms and values (Kohli, 1989;Brown et al., 1995).If other channel members feel they ought to comply with one member in deference to his or her formal position, that member has formal legitimate power (Randolph and Sashkin, 2002).If they ought to comply with a channel member because of generally accepted (informal) norms of behavior, that channel member has informal legitimate power (Mullins, 1999;Randolph and Sashkin, 2002).In a high power distance national culture like Taiwan, there is an acceptance of power inequalities (Hofstede, 1980(Hofstede, , 1991;;Wang and Clegg, 2002;Pagell, Katz and Sheu, 2005).In fact, as a social rule people expect decisions to be made by the more powerful party and may not feel comfortable otherwise (Randolph and Sashkin, 2002;Zhao, Huo, Flynn, and Yeung, 2008).Because of social norms, it is natural for one individual to rely continually on another powerful individual (Mullins, 1999).In point of fact, dependence is encouraged by the power holders (Zhuang, Xi and Tsang, 2010).Given the importance of the gaunxi network in Chinese social and business life (Yang, 1998), an individual actually strives to become dependent on the individuals who are socially recognized as more powerful (Zhuang and Zhao, 2004).
A plethora of studies on channel power indicate that legitimate power is important in understanding exchange relations in dealer-manufacturer dyad (Lee 2001;Liu and Wang, 2000;Maloni and Benton, 2000;Skinner, Gassenheimer and Kelley, 1992) and therefore merit further empirical scrutiny in relation to channel outcomes and relationship quality.

Relationship Quality
In this study relationship quality (RQ) refers to SME manufacturer perceptions and evaluations of how well the relationship fulfills the expectations, predictions, goals, and desires of the manufacturing SME (c.f.Ha and Jang, 2009).High relationship quality signifies that the SME manufacturer believes in the dealer's future performance because the level of past performance has been consistently satisfactory (Wong and Sohal, 2002).Relationship quality is widely viewed as a meta-construct made up of several key relational outcomes that support, reinforce, and complement each other (Dwyer, Schurr, and Oh, 1987) and reflect the overall nature of the exchange relationship (Kumar, Scheer and Steenkamp, 1995).Although there has been considerable work on the conceptualization of relationship quality, the pertinent literature has not reached a general consensus on its constituents (Bove and Johnson, 2001;Naude and Buttle, 2000).Prior research has proposed a variety of relationship quality dimensions (e.g.Dorsch, Swanson and Kelley, 1998;Jap, Manolis and Weitz, 1999;Kumar, Scheer and Steenkamp, 1995;Lages, Lages and Lages, 2005;Leonidou, Leonidas and Talias, 2008), and cooperation, trust, relationship commitment, and relationship satisfaction are given the pride of place among the factors that constitute relationship quality (e.g., Crosby, Evans and Cowles, 1990;Henning-Thurau, Gwinner and Gremler, 2002;Hewett, Money and Sharma, 2002;Hibbard, Kumar and Stern, 2001;Kumar, Scheer and Steenkamp, 1995).Specifically, a rich epidemic of research views relationship quality as a combination of some or all of these aforementioned dimensions (e.g., Bruggen, Kacker and Nieuwlaat, 2005;Farrelly and Quester, 2005;Walter, Muller, Helfert and Ritter, 2003;Ulaga and Eggert, 2006).In the current study relationship quality consisted of four relational outcomes -cooperation, trust, relationship commitment and relationship satisfaction.
Trust, is a fundamental ingredient and lubricant and an unavoidable dimension of social interaction and has received an immense attention across disciplines in social sciences (Ke and Wei, 2008).Mayer, Davis and Schoorman, (1995) identify integrity, benevolence and competence as the dimensions of organizational trust.Integrity is one of the cornerstones of gaunxi and calls for consistent adherence to a set of principles or rules acceptable to other gaunxi distribution channel members (Wang, Siu and Barnes, 2008).The gaunxi rule of empathy carries with it the spirit of benevolence and as such a gaunxi member believes that others will want to do good to him/her as much as he/she does the same (Levin and Cross ( 2004).Within the context of ganqing, a person can join gaunxi in order to benefit on other members' skills and therefore believes that they have the competence necessary for attainment of mutual goals (Wang, Siu and Barnes, 2008).All in all, Wang, (2007) assert that xingyong is the Chinese word that describes trust in gaunxi distribution channel.
Relationship satisfaction has occupied a position of mammoth importance in marketing channels research (Abudul-Muhmin, 2005) and has often been distinguished as economic and social satisfaction (Geyskenes and Steenkamp, 2000).For instance, high performance and the resultant high revenue are expected to stimulate economic satisfaction while reliability and benevolence lead to social satisfaction (Abudul-Muhmin, 2005).Based on consistency theory (Karande, Ha and Singhapakdi, 2008) satisfied members are more likely to continue the relationship and in the same vain are less likely to conceive exit intentions.In gaunxi distribution channel, the rule of renqing and empathy are practiced in the spirit of satisfying relationships with other gaunxi members (Wang, 2007;Wang, Siu and Barnes, 2008).Renqing as a gift is likely to generate economic satisfaction while empathy can be projected to kindle social satisfaction.Indeed, distribution channels cannot function without sustained cooperation in which each party knows what to expect from its counterpart (Mehta, Larsen and Rosenbloom, 2001).Conceptualizations of cooperation in the extant marketing channels literature indicate that cooperation requires interrelated behavior by two or more parties and that it is motivated by the desire to achieve both individual and joint objectives (Mehta, Larsen and Rosenbloom, 2001).According to Sahadev (2005), cooperation is normally expected only when high levels of goal congruity are present.Gaunxi is that social glue that binds together members with commonly shared values and goals (Zhao, Huo, Flynn, and Yeung, 2008) and therefore makes cooperation inevitable.Furthermore, gaunxi channel members cooperate with each other because not doing so may result in the imposition of sanctions by others which consequentially lead to mianzi-a loss face (Lee and Dawes, 2005).No one wants to lose face in gaunxi distribution channel (Wang, 2007) and therefore the deterrent rule of sanction imposition fosters a cooperative behavior among gaunxi members.
According to Morgan and Hunt (1994), relationship commitment is at the heart of any successful long term exchange relationship.They believe commitment depends on among others relationship benefits, relationship termination costs and shared values.The rule of renqing is meant to generate relationship benefits while ganqing touzi (investments related to gaunxi cultivation) increases the termination cost (Wang, 2007).Besides, the emotional attachment emanating from ganqing between exchange partners in gaunxi distribution channel makes relationship commitment inevitable.In addition to that, the fact that gaunxi is a system of social rules based on shared norms and values also makes commitment to exchange relationships among gaunxi members unavoidable.Drawing from the extant literature on legitimate power and relationship quality dimensions this study develops a conceptual framework and research hypotheses.

Conceptual Framework and Hypotheses
The proposed conceptual framework stems from the philosophy that a channel relationship built on the Chinese view of legitimate power and the gaunxi concept is likely to foster relationship quality (trust, relationship commitment, relationship satisfaction and cooperation).Figure 1 depicts the conceptual model of the direct effects of legitimate power on relationship quality dimensions and the hypotheses are postulated thereafter.

Dealer Legitimate Power and Manufacturing SME Cooperation
Cooperation in the current study means that the SME manufacturers and their dealers work together in the pursuit of mutual benefits (Anderson and Narus 1990;Sahadev, 2005).The cooperation between these exchange parties reflects the expectations of working together to achieve both mutual and individual goals jointly (Cannon and Perreault 1999).According to Powers and Reagan (2007), cooperative relationships can be thought of as ''socially contrived mechanisms for collective action, which are continually shaped and restructured by actions and symbolic interpretations of the parties involved.'Chinese people consider gaunxi to be a valuable asset for participating in social exchange activities (Gold, Guthries and Wank, 2002).Individuals use different skills to develop, improve and maintain cooperation in their gaunxi typified distribution channels with desirable individuals (Yang, 2000) especially those who have stayed long and are widely accepted as leaders (Zhuang, Xi and Tsang, 2010).Individuals regarded as the leaders in the gaunxi become the centre of attraction and naturally wield legitimate power to influence others in the distribution channel (Wong and Tam, 2000).Therefore, others will yield to this influence and cooperate with the leader (Zhuang and Xi, 2003).In the context of the current study, the manufacturing SME are conceivably expected to cooperate with dealers who wield legitimate power in the gaunxi typified distribution channel system.The links between legitimate power and cooperation have been demonstrated repeatedly in prior empirical research (e.g.Banton and Maloni, 2005;Michie and Sibley, 1985) and therefore the following hypothesis is postulated: H 1 : Levels of dealer's legitimate power are positively correlated with levels of the manufacturing SMEs' cooperation.

Dealer Legitimate Power and Manufacturing SME Relationship Commitment
In this study relationship commitment reflects the intentions and values of SME manufacturers regarding continuation of the relationship, acceptance of the joint goals and values of the partnership with dealers, and the willingness to invest resources on behalf of the relationship (Sheu and Hu, 2009).Relationship commitment is espoused to consist of three facets and these are affective, calculative/ continuance or instrumental and moral or normative (Cater and Zabker, 2009).These three facets befit and conform to the gaunxi norms.Calculative commitment rests largely on a rational calculation of reciprocal benefits "renqing" (Lee and Dawes, 2005).Affective commitment involves acceptance and internalization of the channel's shared goals and values and a strong identification with both the channel and channel leader due to ties "yuan" (Bond, 1991).Normative commitment means that partners stay in the relationships because they feel doing otherwise will result in a loss of face "mianzi" (Tung and Yeuang, 1996).Based on these Chinese psycho-cultures of "yuan", "renqing" and "mianzi", channel members are more likely to be committed to a relationship with a channel leader who wields legitimate power in the "gaunxi" typified distribution channel system.Conceivably, in the current study context, the manufacturing SME are expected to be committed to a relationship with dealers who wield legitimate power in their "gaunxi" distribution channel system.The linkages between legitimate power and relationship commitment have been empirically validated in previous studies (e.g.Zhao, Huo, Flynn, and Yeung, 2008;Brown, Lusch, and Nicholson 1995) and therefore, the following hypothesis is postulated: H 2 : Levels of dealer's legitimate power are positively correlated with levels of the manufacturing SMEs' relationship commitment.

Dealer Legitimate Power and Manufacturing SME Relationship Satisfaction
Extant empirical studies suggest relationship satisfaction refers to social as well as economic aspects of the exchange (Geyskens, Steenkamp and Kumar, 1999).In line with prior research on relationship quality, the current study focuses on overall relationship satisfaction (e.g.Dorsch, Swanson and Kelley, 1998;Ulaga and Eggert, 2006;Skarmeas, Katsikeas, Spyropoulou and Salehi-Sangari, 2008).It defines relationship satisfaction as a positive affective state resulting from the appraisal of all aspects of a dealer's working relationship with the SME manufacturers (Cater and Zabker, 2009).A channel leader in a gaunxi has an obligation to care for other members and even actively seek for such dependence (Zhuang, Xi and Tsang, 2010).Other members' dependence consolidates the holder's power "zhao kao shan -looking for a reliable mountain to back you up" (Zhuang and Zhou, 2004).Furthermore, the gaunxi norms of renqing and mianzi require other members and the channel leader to assist those in need (Lee and Dawes, 2005) and to give face to others and to protect the face that they have (Bond, 1991).Such assistance and protection is likely to generate good feeling in the dependent member and eventually social satisfaction with the channel leader.In addition, reciprocated benefits through "renqing" norm from channel leaders will earn economic satisfaction to the dependent member in the distribution channel.Conceivably, in the current study context, the manufacturing SMEs are expected to be satisfied with their relationship with dealers who wield legitimate power in the "gaunxi" typified distribution channel system.Prior studies have empirically supported the linkage between legitimate power and relationship satisfaction (Brown, Lusch, and Nicholson 1995;Banton and Maloni, 2005) and therefore the following hypothesis is postulated: H 3 : Levels of dealer's legitimate power are positively correlated with levels of the manufacturing SMEs' relationship satisfaction.

Dealer Legitimate Power and Manufacturing SME Trust
A cross examination of the existing spate of relationship marketing literature indicate that trust is probably the most widely studied and accepted construct (e.g.Geyskens, Steenkamp and Kumar, 1998;Madhok, 2006;Morgan and Hunt, 1994).Consistent with this wave of literature the current study defines trust as the willingness by SME manufacturers to rely on the dealers whom they have confidence in (Moorman, Zaltman and Deshpande, 1992).Trust is critical in dealer-manufacturer relationships because it lessens the transaction costs of the exchange process and concurrently augments transaction value via the creation of a positive working environment (e.g., Wren, Simpson and Paul, 1998;Aulakh, Kotabe and Sahay,1996;Johnson, Cullen, Sakano and Takenouchi, 1996;Zhuang, Herndon and Zhou, 2003).In Chinese business communities, trust is a crucial factor in sustaining the complex business connections (gaunxi) and contributes to the success of a firm (Zhuang, Xi and Tsang, 2010).Trappey and Lai (1996) stressed that the prevailing culture in Taiwan emphasizes trust at personal level.Chinese business persons deem the trustworthiness and xingyong (i.e. a person's credibility) of their trade partners important in business dealings.Xingyong generally refers to the "integrity, credibility, trustworthiness, or the reputation and character of a person," and "a person's credit rating" in business circles (Kiong and Kee 1998, p. 85).A channel leader who wields legitimate power is obviously expected to have "xingyong" (Trappey and Lai, 1996).On the basis of this "xingyong" and "yuan" the channel members in the gaunxi can trust their leader without proviso (Zhuang and Zhou, 2004).Conceivably, in the context of the current study, the manufacturing SMEs are expected to trust dealers who wield legitimate power in their "gaunxi" typified distribution channel system.Evidence from empirical previous studies supports the linkage between legitimate power and trust (e.g.Yukl, Kim and Falbe, 1996;Hunt, Mentzer and Danes, 1987) and therefore the following hypothesis is postulated: H 4 : Levels of dealer's legitimate power are positively correlated with levels of the manufacturing SMEs' trust.

Sample and Data Collection
The Small and Medium Enterprise Association of Taiwan (SMEAT) was selected as the sampling frame.By cooperating with the SMEAT Association, 1000 SME manufacturers were randomly selected from the association membership database and the questionnaire was distributed to the SME manufacturing owners or the manufacturers' officials who were in charge of either sales or marketing.These participants were considered competent to evaluate their firms' relationships with their dealers.Prior to the distribution of the questionnaire in person, these potential participants were contacted and informed of the significance of the current study by telephone.A total of 400 usable questionnaires were randomly retrieved for the final data analysis, representing a response rate of 40 percent.

Sample Description
The profile of the participants is presented in Table 1.The sample showed that more than three quarters of the participating SME manufacturing firms employ less than 10 workers and the majority of the remainder has a workforce between 11-21 employees.Half of the participating manufacturing SMEs (50.8%) had less than 5 years experience, followed by twenty-six per cent, seventeen per cent, and about six per cent with 6-10 years, 11-20 years, more than 20 years experience respectively.A majority of the participants had revenues between USD30-90 millions, and the reminder made revenues between US90-$190 (12.5%),US$190-$300 (3.5%) and more than US$300 million (5%).The analysis also indicated that consumable and non-consumable goods manufacturers almost equally occupied the research sample, although the former shared slightly higher proportion (56.6%).

Measurement Instrument
Research scales were operationalised mainly on the basis of previous work.Proper modifications were made in order to fit the current research context and purpose.Some five-item scales adapted from Lee's (2001) previous work were used to measure traditional legitimate power."Trust" used a six-item scale measure adopted from Kabadayi and Ryu (2007), while a five-item scale to measure "cooperation" was adopted from Cannon and Perreault (1999)."Relationship satisfaction" was measured by a six-item scale which was also adapted from Cannon and Perreault's (1999) work.Finally, "relationship commitment" used a five-item scale from Morgan and Hunt (1994).All the measurement items were measured on a 5-point Likert-type scales that was anchored by 1= strongly disagree to 5= strongly agree to express the degree of agreement.Individual scale items are listed in Appendix

Measure Validation
In accordance with the two-step procedure suggested by Anderson and Gerbing (1988), prior to testing the hypotheses, confirmatory factor analysis (CFA) was performed to examine reliability, convergent and discriminant validity of the multi-item construct measures using AMOS 5. Initial specification search led to the deletion of some of the items in the constructs scale in order to provide acceptable fit.Recommended statistics for the final overall-model assessment show acceptable fit of the measurement model to the data: chi-square value over degrees = 858.289(242), p<.01; Goodness-of-Fit Index (GFI) = 0.847; Adjusted Goodness-of-Fit Index (AGFI) = 0.810; Incremental Index of Fit (IFI) = 0.916; Tucker and Lewis Index (ITL) = 0.904; Comparative Fit Index (CFI) = 0.915; Root Mean Square Error of Approximation (RMSEA) = 0.080.Loadings of individual items on their respective constructs are shown in Table 2, while the scale construct correlations are presented in Table 3.The individual item loadings are all above the recommended 0.5 (Anderson and Gerbing, 1988), indicating acceptable individual item reliabilities as more than 50% of each item's variance is shared with its respective construct.
Composite reliabilities (CR) and average variance extracted (AVE) for each construct were also computed using the formulae proposed by Fornell and Lacker (1981).The results are shown in Table 2, and descriptive statistics and correlations among the study constructs are presented in Table 3.The composite reliabilities (CR) are above 0.8 therefore, exceeding the recommended value of 0.7 suggested by Hulland (1999).This indicates an excellent and acceptable internal consistency and reliability of the respective measures.All average variance explained (AVE) values are above 0.5 and thus are also acceptable according to the literature (Fraering and Minor, 2006).By and large, these results provided evidence for acceptable levels of research scale reliability.Discriminant validity was established by ensuring that the average variance extracted (AVE) for each multi-item construct was greater than the shared variance between constructs (Fornell and Larcker, 1981;Gerbing and Anderson, 1988;Nunnally and Bernstein, 1994).As such, all pairs of constructs reveal an adequate level of discriminant validity (see Table 3).

Data Analysis and Results
Structural Equation Modeling (Maximum Likelihood method in AMOS 5) was used for the data analysis based on the conceptual framework (Figure 1).The results are reported in  1990;Browne and Cudeck, 1993;Marsh, Balla and Hau, 1996).This suggests that the model converged well and could be a plausible representation of underlying empirical data structures collected in Taiwan.The results in Table 4 provide support for all the proposed four hypotheses.The first core issue was the relationship between dealer's legitimate power and SME manufacturer's cooperation.Consistent with hypothesis one (H1), results indicate that higher levels of dealer's legitimate power are associated with higher levels of the SME manufacturer's cooperation with the dealer.The second core issue was the relationship between dealer's legitimate power and SME manufacturer's relationship commitment.Also in support of hypothesis two (H2), the results indicate that higher levels of legitimate power the dealer possesses are associated with higher levels of relationship commitment the SME manufacturer has with the dealer.The third core issue was the relationship between dealer's legitimate power and SME manufacturer's relationship satisfaction.The standardized coefficient of legitimate power for relationship satisfaction is positive and significant.This implies that hypothesis three (H3) is consistent with the current study prediction and is supported.Thus, higher levels of dealer's legitimate power are associated with higher levels of relationship satisfaction the SME manufacturer has with the dealer.The last core issue was the relationship between dealer's legitimate power and SME manufacturer's trust.The current study empirical results are in line with the proposed hypothesis four (H4) and support the reasoning that the higher the level of legitimate power the dealer possesses, the higher the SME manufacturer's trust in the dealer.

Discussion and Conclusions
This research investigated the direct influence of traditional legitimate power on relationship quality dimensions -cooperation, relationship commitment, relationship satisfaction and trust in Taiwanese distribution channel system as perceived by the SME manufacturers.In order to test the hypotheses, data were collected from manufacturing SMEs in Taiwan.All the proposed four hypotheses were empirically supported indicating that dealers' legitimate powers positively influence manufacturing SMEs' trust, relationship satisfaction, relationship commitment and channel cooperation in a significant way.This finding empirically verifies that legitimate power influences relationship quality in gaunxi typified distribution channel system.
Interesting to note from the results is the fact that legitimate power has the most significant impact on cooperation than relationship satisfaction, trust and relationship commitment respectively.By implication, this finding indicates that manufacturing SMEs in Taiwan are more likely to cooperate with a dealer who wields legitimate power in their gaunxi than they do to relationship satisfaction, trust and relationship commitment respectively.This seemingly robust relationship between legitimate power and cooperation should be anticipated given that cooperation is normally expected only when high levels of goal congruity are present (Sahadev, 2005) and also that legitimate power is derived from societal or institutional acceptance of commonly shared norms and values (Brown, Lusch and Nicholson, 1995).Besides, this finding perhaps also conforms to Wong and Tam (2000) gaunxi networking four Ts procedure: testing, trial, trust, and teaming.Likewise, in the current study context, cooperation forms the basis for testing a relationship in gaunxi typified distribution channel system (Yang, 2000).The cooperation between the SME manufacturer and dealer put the relationship survival and continuity to trial.If the SME manufacturer's expected satisfaction with the relationship is attained, e.g.relationship benefits or renqing are realized, then trust in the dealer or xingyong increases and consequently teaming heighten the SME manufacturer's commitment to the relationship.In a nutshell, the current study findings provide tentative support to the proposition that dealer's legitimate power in a gaunxi typified distribution channel should be recognized as a significant facilitator of relationship quality among manufacturing firms in the SME setting.This study therefore, serves to confirm the above assertion and hereby proceeds to provide managerial implications of such finding.

Managerial Implications
To the best knowledge of the authors, the current study is the first to investigate the direct effects of legitimate power on relationship quality dimensions in gaunxi distribution channel using data collected from SME manufacturers in Taiwan.Given that Taiwan is dominated by Chinese culture our findings potentially provide fruitful implications for both academicians and marketing practitioners in other East Asian countries such as Singapore, Hong Kong, Korea and mainland China all that share in and are dominated by this Chinese culture of gaunxi.The managerial implications are that an appropriate use of legitimate power by the major dealers can significantly enhance relationship quality with SME manufacturers in gaunxi distribution channel.Therefore, managers in the major dealer firms can utilize their legitimate powers in order to attain channel cooperation, trust, relationship satisfaction and relationship commitment with manufacturing SMEs in their gaunxi typified distribution channel.In this regard, they should encourage their boundary personnel to develop a good gaunxi with the boundary personnel of the SME manufacturers.
By extension, further global implications drawn from the current research findings are that, managers in foreign major dealer firms seeking to do business with SME manufacturers in Taiwan aught to cultivate gaunxi first and observe the social rules of the game in order to develop relationship quality, survive and succeed in doing business in a Chinese culture dominated setting such as Taiwan.Abiding by gaunxi social rules is paramount for the foreign managers since this is more important to gaunxi distribution channel members than a formal written contract.
In addition, an academic contribution is made to the existing body of business to business (B2B) marketing literature related to major dealer and SME manufacturer relations by systematically exploring the influence of powers drawn from social norms or institutional acceptance -legitimate power on relationship quality dimensions in the context of gaunxi -a Chinese way of doing business.Besides, a successful attempt made to apply social rule system theory -a sociology theory in this respect for the first time in order to aid our understanding of the linkage between traditional legitimate power and relationship quality dimensions in gaunxi distribution channel, contributes theoretical impetus to the existing B2B marketing channels literature.

Limitations and Future Research
This study has some limitations.This study was limited to a random sample of SMEs in the manufacturing industry in Taiwan.Subsequent research could replicate this study in broader sampling contexts and also extend our conceptual framework by studying the effects of a larger set of variables.For instance, the influence of other channel powers such as referent, expert and information power could be investigated.Likewise, studying the indirect effects of these channel powers on the studied relationship outcomes (cooperation, trust, relationship satisfaction and relationship commitment) could represent further avenues for developing the relationships revealed in this research.Above and beyond, this will provide added insights and immensely contribute new knowledge to the existing body of distribution channels literature on small business.Finally, the data were gathered from the SME manufacturer's side.The results would be more informative if data from both sides of the channel dyad were compared.Future studies may be conducted by using paired data.

Figure
Figure 1.Conceptual Framework

Table 2 .
Accuracy analysis statistics

Table 4 .
The model is acceptable in terms of overall goodness of fit.Acceptable model fit are indicated by GFI ≧ 0.80; AGFI ≧ 0.80; RMSEA values ≦

Table 5 .
Results of structural equation model analysis