INTRODUCTION
The Chinese term ‘guanxi’ literally refers to connections between entities. With regard to the interpersonal relationships in Chinese society, guanxi specifically refers to a network of informal–interpersonal relationships and exchanges of favors through these relationships (Lovett, Simmons, & Kali, Reference Lovett, Simmons and Kali1999; Luo, Reference Luo2000). With favor exchange as its keep function, guanxi is created by the ongoing development of relationships through which participants can share scarce resources and cope with uncertainties (Su & Littlefield, Reference Su and Littlefield2001). Interpersonal relationships in Chinese society are characterized by high levels of interdependence and reciprocity (Tsui & Farh, Reference Tsui and Farh1997). Importantly, the concept of guanxi is deeply embedded in Chinese ethics and social norms and it defines the basic cultural rules and regulations that guide participants’ actions and coordinate the Chinese social system (Hwang, Reference Hwang1987; Yeung & Tung, Reference Yeung and Tung1996). Given the importance of guanxi, it has long been recognized as a critical antecedent of both organizational and personal success in China (Yeung & Tung, Reference Yeung and Tung1996; Tsang, Reference Tsang1998).
As a central element in Chinese society, it is not surprising that guanxi practices have been the focus of a large number of research studies examining both the phenomenon itself and its implications for organizations in China. These guanxi studies can be classified into two broad streams. The first of these, and the larger stream, focuses on the favorable consequences that top managers’ guanxi practicesFootnote 1 bring to organizations. Specifically, the use of guanxi by managers has been shown to be a shortcut for approaching the ‘right person’ and asking for favors (Luo & Chen, Reference Luo and Chen1997). Because China's economic system is volatile and institutional frameworks are in a nascent stage of development, resources secured through guanxi are often not accessible by other means. In addition, the guanxi exchange process in China is often more reliable than market-based approaches, largely because of the deeply culturally embedded system of guanxi rules and norms with sanctions for violators (Xin & Pearce, Reference Xin and Pearce1996). This stream of research has substantiated the conclusion that managers’ guanxi practices can be an important source of competitive advantage (see, e.g., Yeung & Tung, Reference Yeung and Tung1996; Tsang, Reference Tsang1998; Luo, Reference Luo2000).
In contrast, the second stream of guanxi studies, which is smaller and more recent, has emphasized the downsides of managers’ guanxi practices in business organizations (e.g., Dunfee & Warren, Reference Dunfee and Warren2001; Fan, Reference Fan2002; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). These studies highlight some important costs that arise when guanxi practices are used by managers of business organizations. Since guanxi practices effectively limit business exchanges to a privileged few in control of scarce resources, guanxi practices naturally lead to ‘benefit of the few at the expense of the many’ (Dunfee & Warren, Reference Dunfee and Warren2001: 200). In addition, these practices violate norms of justice, and can easily lead to corruption (Fan, Reference Fan2002; Su, Sirgy, & Littlefield, Reference Su, Sirgy and Littlefield2003). Finally, these practices often cause negative reactions among employees, as they feel unfairly treated or exploited. For example, because many guanxi exchanges involve the employment of friends or relatives of powerful individuals outside the firm, employees may endure the brunt of the costs associated with guanxi practices, as they may be forced to both forego promotion opportunities and work with less-than-qualified co-workers (Chen, Chen, & Xin, Reference Chen, Chen and Xin2004).
Although both streams of research provide valuable insights into the roles of managerial guanxi practices in business operations, neither approach provides a complete picture. Specifically, there are two large shortcomings in the literature on guanxi in Chinese business organizations. First, little attention has been paid to the potential connection between the positive and negative consequences (the costs and benefits) of using guanxi in business organizations. Although many prior guanxi studies have noted the co-existence of benefits and costs (e.g., Luo & Chen, Reference Luo and Chen1997; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004), the studies themselves generally focus on one or the other. Our study is designed to reveal a natural link between the positive and the negative implications: it is the expectation of benefits that motivates top managers to enact guanxi practices in the operation of their firms. However, because the rules of guanxi practice, deeply embedded in Chinese culture, emphasize mutual obligation and reciprocity (Yang, Reference Yang1993, Reference Yang1994) a favor rendered carries with it the very strong expectation that the favor will be repaid in the future (Luo, Reference Luo2000; Su, Sirgy, & Littlefield, Reference Su, Sirgy and Littlefield2003). Prior research has not taken this reciprocity into account. As a result, much prior research has emphasized the benefits brought into the firm through managerial guanxi practices without consideration of what has been promised in return. In addition, the studies of guanxi's costs have typically not been concerned with the potential benefits that might have been received or will be received in the future. Our study is designed to provide insights into the costs as well as the benefits of guanxi practices, and the mechanisms through which the two are related. Specifically, we identify two different types of managerial guanxi practices – inward guanxi practices, which refer to those guanxi practices that make use of managers’ guanxi ties and lead to resource inflows for their organizations; and outward guanxi practices – those guanxi practices that make use of managers’ guanxi ties and lead to resource outflows from the organization. We argue that these two managerial guanxi practices reflect different directions of the flow of resources involved in the guanxi-based exchanges, and their interconnection depicts the interchange of the benefits and costs of using guanxi.
The second large shortcoming is that the literature has thus far not considered the base of the guanxi practices. Our study considers two such bases – family and familiar. Family-based guanxi (jia ren), involves ties to family members and relatives while familiar-based guanxi (shou ren), involves friendship ties. As we will describe in more detail below, the cultural rules of guanxi exchange differ importantly between these two types of ties, forming the basis for our expectation that the costs associated with each guanxi base will also differ.
We strive to make four contributions with this study. First, we introduce the concept guanxi focus (inward vs. outward), defining it as the direction of the flow of resources involved in the exchange (inward to the firm or outward from the firm). Second, we explain why we expect inward (outward) guanxi practices to have a natural contagion (i.e., positive association) with outward (inward) guanxi practices. Third, we explain why we expect the reactions of employees to differ depending on focus of the guanxi practice – inward or outward. Finally, we point out that the guanxi base (family or familiar), already well known in the literature (Luo & Chen, Reference Luo and Chen1997; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004), is likely to have an important impact on both the contagion between inward and outward guanxi practices and the costs associated with those practices (whether those practices be inward or outward).
THEORETICAL BACKGROUND AND HYPOTHESES
Guanxi, guanxi exchange, and guanxi practices in Chinese society
Yang (Reference Yang1993) proposed three major categories of guanxi in China: (1) family guanxi (jia ren), involving relationships with family members and relatives; (2) familiar guanxi (shou ren), involving friends, acquaintances, and indirectly connected people; and (3) stranger guanxi (sheng ren), involving entirely unacquainted and disconnected individuals. Since stranger guanxi does not involve meaningful interpersonal connections in Chinese culture and thus includes little affect or mutual obligation, the exchanges in stranger guanxi are typically viewed as equivalent to open market transactions (Jiang, Reference Jiang2009). For this reason, following prior studies (e.g., Luo & Chen, Reference Luo and Chen1997; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004), we limit our discussion to family and familiar guanxi.
The cultural rules of guanxi exchange dictate that those who request favors should do so out of a genuine need for help, so favors are not to be casually requested. Whether using family or familiar guanxi, a request for help should not be made without a genuine need. One motivator for this care and concern about asking for favors is that once asked, the help that is offered must be accepted. A person who refuses the help offered after a request will cause the helper to lose face (Yang, Reference Yang1993, Reference Yang1994; Earley, Reference Earley1997). For example, Song and Werbel (Reference Song and Werbel2007) show that Chinese students seeking jobs are much less likely to use their interpersonal networks to find jobs than are students in the United States. The authors conclude that the reason for this reticence among Chinese students is that if they ask for help, they are then obligated to take whatever job their request generates. Their study showed that Chinese students who used their guanxi networks to find jobs were motivated by lower job search confidence, and those who used their guanxi networks to secure jobs tended to have lower job satisfaction in the jobs they secured.
In Chinese society, all guanxi is built upon two fundamental components: the affective dimension, which is grounded in the closeness of kinship or friendship, and the instrumental dimension, which balances the benefits received by both parties and leads to continued sharing of values and resources (Fei, Reference Fei1992). These two dimensions together determine the basic rules of guanxi.
The affective dimension dominates in family guanxi. Family guanxi ties are characterized by strong attachment, trust, and commitment (Fei, Reference Fei1992). In Chinese culture, the rules of family guanxi require that family members voluntarily help each other without expectation of return (Yang, Reference Yang1993). This is a critical distinction between family and familiar guanxi. Further, with respect to the instrumental dimension, family guanxi tends to be motivated by non-calculative and non-economic factors such as increased mutual identification among family members and enhanced family harmony and stability (Fei, Reference Fei1992). For these reasons, the norm of reciprocity in guanxi exchange is much weaker and less utilitarian for family guanxi, and family guanxi practices tend to induce few (if any) subsequent reciprocal obligations.
For familiar guanxi (friendship based), there is more balance between affective and instrumental dimensions. This balance leads to strong norms of mutual obligation and commitment, such that guanxi members have the right to ask favors from each other but also have the responsibility to help others when called upon to do so (Hwang, Reference Hwang1987; Luo, Reference Luo2000). The resulting mutual obligation requires that those who have been helped through familiar guanxi should be grateful and are obligated to repay the favor in the future (Su & Littlefield, Reference Su and Littlefield2001). Hence, ‘… once guanxi is established between two people, each can ask a favor of the other with the expectation that the debt incurred will be repaid sometime in the future’ (Yang, Reference Yang1994: 1–2). For this reason, the exercise of familiar guanxi enables a loop in which each participant can enjoy the benefits of guanxi but at the cost of taking on reciprocal obligations that must be repaid in the future (Earley, Reference Earley1997; Luo, Reference Luo2000).
Despite the common instrumental motivation shared by familiar guanxi exchange and regular open market exchange, guanxi exchange includes some distinctive mechanisms that enhance its effectiveness in the Chinese context. First, while market exchanges are driven, more or less, by economic calculation, familiar guanxi exchange involves high mutual obligation based on personal interconnection and affection. Second, the resources requested through guanxi practices are often unavailable through regular market means (e.g., preferential prices, government protection) and those that are available are accessible through a relatively underdeveloped factor market in China (Luo, Reference Luo2000). So, guanxi practices can substitute for an ineffective market environment, permitting managers to secure rare or otherwise inaccessible resources essential for their organizations (Xin & Pearce, Reference Xin and Pearce1996; Tsang, Reference Tsang1998). Third, guanxi exchange is deeply embedded in and regulated by social norms rather than by formal contracts and institutions (Yeung & Tung, Reference Yeung and Tung1996; Earley, Reference Earley1997). By combining affective and instrumental concerns, the heavy social sanction of guanxi exchange makes the sense of reciprocity in guanxi exchanges less immediate and utilitarian and more long-term oriented than market transactions (Jiang, Jin, Jiao, & Ma, Reference Jiang, Jin, Jiao and Ma2009). At the same time, the strong sense of mutual commitment and mutual obligation that are deeply embedded in Chinese culture and society also exert intense pressure on participants to follow the norms of reciprocity in guanxi exchange, thus ensuring the continuity of guanxi in long run (Yang, Reference Yang1993). As a result, guanxi exchange tends to be characterized by lower uncertainty and higher effectiveness than open market exchanges in China's transition society (Yang, Reference Yang1993; Xin & Pearce, Reference Xin and Pearce1996).
Given the benefits of guanxi exchange discussed above, it is no surprise that guanxi is often cited as a key factor behind business success and a mechanism through which firms can gain competitive advantages (Yang, Reference Yang1993; Luo & Chen, Reference Luo and Chen1997; Su, Sirgy, & Littlefield, Reference Su, Sirgy and Littlefield2003). Consequently, as agents of their organizations, top managers are frequently motivated to develop guanxi connections to powerful others (‘la guanxi,’ Luo, Reference Luo2000: 4), and to use their guanxi partners ‘to make exchanges, manufacture indebtedness, or accomplish tasks’ (Guthrie, Reference Guthrie1998: 226). In the present study, we define top managers’ business endeavors that aim at developing, maintaining, or making use of guanxi ties in the operation of their organizations as managers’ guanxi practicesFootnote 2.
Inward−outward guanxi contagion: The cost mechanism of guanxi practices
Top managers, in their guanxi networks, simultaneously serve as individual social entities and as agents of their organizations (Xin & Pearce, Reference Xin and Pearce1996; Peng & Luo, Reference Peng and Luo2000). This ambidexterity of identity signifies two different directions of resource flow via these top managers as a result of their guanxi practices. That is, because of the instrumental nature of guanxi practices, favors and benefits are expected both by these top managers from other guanxi partners and from these top managers by other guanxi partners. As we have already noted, it has been well established that by drawing upon familiar guanxi outside the organization, top managers can secure inflows of resources into the organization (Luo & Chen, Reference Luo and Chen1997; Peng & Luo, Reference Peng and Luo2000). However, on the other hand, they must also play the role of favor provider as a result of their guanxi practices. This is particularly important because top managers have significant control over the resources of their organizations, and the favors and benefits expected from them (outward guanxi practices) are usually linked to their firms’ resources. For this reason, not only can managers’ guanxi practices provide guanxi rents for their organizations, but they can also provide outsiders with access to the resources of the organization (Jiang et al., Reference Jiang, Jin, Jiao and Ma2009). Given the distinct organizational implications of these different guanxi practices and as we explained earlier, we separate managerial guanxi practices into two types – inward guanxi practices, which refer to those guanxi practices that make use of managers’ guanxi ties and lead to resource inflows for their organizations; and outward guanxi practices – those guanxi practices that make use of managers’ guanxi ties and lead to resource outflows from the organization. Note that in this definition, inward and outward refer to the direction of resource flow involved in the particular guanxi practice.
The dichotomization of managerial guanxi practices according to the direction of resource flows, when linked with the guanxi exchange mechanism embedded in Chinese culture, leads to an important and heretofore ignored tradeoff of guanxi benefits and guanxi costs in organizations, that is, the contagion between inward and outward guanxi practices. This contagion can take place in two potential directions. On the one hand, as widely considered by prior studies, top managers may first engage in inward guanxi practices, turning to their ‘guanxi assets’ (Tsang, Reference Tsang1998) or ‘guanxi capital’ (Luo, Reference Luo2000) to secure resources and external support needed by their organizations (Luo & Chen, Reference Luo and Chen1997). By engaging in these practices, the managers activate the loop of guanxi exchange, simultaneously enjoying favors and taking on the reciprocal obligation to repay the favors when they are called upon to do so (Yang, Reference Yang1994; Earley, Reference Earley1997). Thus, by engaging in inward guanxi practices, top managers incur guanxi debts, and are likely to be called upon to use the collective wealth of their organizations to cover those debts, leading to consequent outward guanxi practices.
On the other hand, as we noted earlier, top managers largely engage in guanxi practices as agents of their organizations, so they may first be called upon to use outward guanxi practices, benefiting external guanxi partners with the resources of their organizations. As a result, with the sanctions of reciprocal norms in guanxi exchange, these top managers can secure the right to ask favors from those partners when needed. In this way, top managers may use outward guanxi practices as investments to pave the way for inward guanxi practices (Jiang, Reference Jiang2009). For instance, it is quite common in China for an executive to employ relatives of local government officials or the managers of important customer firms, and to trade these off for official protection, government concessions, or large and consistent orders (Nee, Reference Nee1992).
Taken together, both of the two potential paths of benefit−cost tradeoff in managerial guanxi practices lead to a direct linkage between inward (outward) and outward (inward) guanxi practices over time. We describe this linkage between inward and outward guanxi practices as ‘inward−outward guanxi contagion’ regardless of the direction of the initial exchange. In the tradeoff between guanxi benefits and costs, we can see that the contagion between value-enhancing inward guanxi practices and value-dissipating outward guanxi practices characterize an important but heretofore ignored cost mechanism of using guanxi in organizational contexts. We thus predict:
Hypothesis 1 : The level of inward (outward) guanxi practices is positively related to the level of outward (inward) guanxi practice.
Inward−outward guanxi contagion: The moderating effects of guanxi base
While inward guanxi practices tend to be linked to outward guanxi practices, there are several reasons to believe that not all inward guanxi practices will be equally likely to spark outward guanxi practices and vice versa. Specifically, as mentioned earlier and as noted by Yang (Reference Yang1993), the affective dimension of guanxi (reflected in tie strength, intimacy, closeness) can sometimes overpower the instrumental dimension of guanxi (reflected in the benefits involved). When this occurs, the reciprocity requirement associated with guanxi practices becomes much less acute (Yang, Reference Yang1994; Leung & Yeung, Reference Leung and Yeung1995). For this reason, we expect that the degree of affect reflected in a particular guanxi tie may alter the contagion between corresponding inward (outward) and outward (inward) guanxi practices.
As kinship distance increases, the sense of voluntary obligation decreases. Therefore, with familiar guanxi, instrumental concerns tend to play far more salient roles than in family guanxi (Fei, Reference Fei1992; Yang, Reference Yang1994). As a result, familiar guanxi exchanges are more likely to represent utilitarian transactions based on instrumental calculations and shared expectations of comparable repayment, both of which rely heavily on the strict requirement of reciprocity for favors received (Hwang, Reference Hwang1987). Therefore, managers who make use of familiar guanxi will be strongly motivated to adhere to the norm of reciprocity (Jiang, Reference Jiang2009). For these reasons, we can expect that inward (outward) guanxi practices based on familiar guanxi are more likely to incur subsequent outward (inward) guanxi practices than are family-based guanxi practices.
Hypothesis 2 : Relative to those based on family guanxi, inward (outward) familiar guanxi practices are more positively related to outward (inward) familiar guanxi practices.
Employee responses to managerial guanxi practices
Business ethics research has pointed out that as a social existence beyond the governance of formal institutional frameworks (Yang, Reference Yang1994; Xin & Pearce, Reference Xin and Pearce1996; Luo, Reference Luo2000), managers’ guanxi practices signify ‘steering clear of the structural constraints of policy through the shortcut of exclusive, discriminatory relationships’ and ‘fulfilling the gratification of somebody or some groups at the cost of collective benefits of the society’ (Yang, Reference Yang1993; Bian, Reference Bian1997; Dunfee & Warren, Reference Dunfee and Warren2001; Fan, Reference Fan2002). Therefore, managerial guanxi practices in business intrinsically go against fundamental principles of fairness and thus tend to be deemed as unjust by observers (Su & Littlefield, Reference Su and Littlefield2001; Fan, Reference Fan2002). Furthermore, according to Leventhal, Karuza, and Fry (Reference Leventhal, Karuza and Fry1980), a key principle of justice is bias suppression, which requires that managers suppress their self-interests as well as any demands to serve the interests of their personal relations (Leventhal, Karuza, & Fry, Reference Leventhal, Karuza and Fry1980; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). As a result, when managers use guanxi practices, their actions may be interpreted as unfair, unjust, and/or unethical by employees. Other authors have described the concern about employee reactions to managers’ use of guanxi practices (Lind & Tyler, Reference Lind and Tyler1988; Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). In our study, we consider the negative reactions of employees as one of the implicit (and frequently unaccounted for) costs that are incurred when managers use guanxi practices.
Although managers’ use of guanxi practices necessarily involves some violations of basic principles of justice (Fan, Reference Fan2002), some conditions may attenuate this guanxi cost. For example, according to the self-interest model of justice (Lind & Tyler, Reference Lind and Tyler1988; Ambrose, Harland, & Kulik, Reference Ambrose, Harland and Kulik1991), favorable outcomes can sometimes justify certain procedures that would otherwise be perceived as unjust (see also Tyler & Blader, Reference Tyler and Blader2000). For this reason, we expect that the direct outcome implications of different guanxi practices will play a key role in accounting for employees’ responses.
As noted earlier, managers’ use of inward guanxi practices have long been identified as important to enhancing performance in Chinese business organizations (Xin & Pearce, Reference Xin and Pearce1996; Luo, Reference Luo2000; Peng & Luo, Reference Peng and Luo2000). Further, there is some suggestive evidence that the benefits associated with inward guanxi practices leads to a level of acceptance of those practices among employees (Hwang, Reference Hwang1987; Guthrie, Reference Guthrie1998). As a result, employees may interpret inward guanxi practices as advantageous to the performance of their organizations, which is an important antecedent of their personal gains. According to the self-interest model of organizational justice, these favorable perceptions should help to justify the use of inward guanxi practices.
Outward guanxi practices, in contrast, have less positive outcome implications for organizations. With managers playing the role of favor provider, outward guanxi practices involve outflows of organizational resources. Accordingly, these practices cannot be justified by advantageous consequences, inclining employees’ to perceive them as unjust. Moreover, outward guanxi practices often involve uses of organizational resources that are visible and salient to employees. For example, when top managers enact outward guanxi practices that involve the employment of relatives of powerful outside individuals, qualified employees may directly suffer through lost opportunities or income (Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). In addition, some costs of outward guanxi practices can be observed by employees through both formal and informal means (e.g., intra-firm networks, records of business entertainment costs, see Nee, Reference Nee1992). The self-interest model of justice predicts that the immediacy of the costs associated with outward guanxi practices and their direct relevance to employees will exacerbate employees’ negative interpretations of and reactions to those practices.
Hypothesis 3 : Outward guanxi practices tend to induce more negative justice perceptions among employees than inward guanxi practices.
Employee trust and organizational commitment
In organizations, when employees’ perceive guanxi practices negatively, it may lead to several forms of reactions (Porter & Steers, Reference Porter and Steers1973; Shore & Wayne, Reference Shore and Wayne1993). An employee's perception that certain managerial actions are unjust may then contribute to other factors such as job satisfaction or organizational commitment. In this study, we consider two such factors – trust in management and organizational commitment – to represent employees’ reactions to guanxi practices.
Employees’ trust in management is linked to factors such as the perceived integrity of management and the favorability of expected outcomes (Clark & Payne, Reference Clark and Payne1997). As mentioned before, given their discriminatory nature, managers’ use of guanxi practices inevitably violates rules of fair competition (Dunfee & Warren, Reference Dunfee and Warren2001; Fan, Reference Fan2002). For this reason, using guanxi practices may naturally affect the perceptions of employees about the manager's integrity. Nevertheless, based on the contrasting outcome implications of inward and outward guanxi practices, we believe that the expected advantageous consequences will help to justify the use of inward guanxi practices, whereas a similar justification is not available for outward guanxi practices. Correspondingly, we expect that using inward guanxi practices will impact employees’ perceptions of managerial integrity less negatively than using outward guanxi practices. Moreover, it is also clear that the widely observed beneficial consequences of inward guanxi practices will lead to more favorable outcome expectations, which may further enhance employee trust in management. By contrast, value-reducing outward guanxi practices tend to lower employees’ outcome expectations and thus attenuate their trust in management.
Prior studies have also pointed out that the impacts of perceived injustice associated with managerial actions tend to go beyond the interpersonal level and attenuate employees’ perceptions of organizational justice (Leventhal, Karuza, & Fry, Reference Leventhal, Karuza and Fry1980). In particular, perceived inequity and injustice in organizations may arouse tensions among employees regarding procedural fairness and reduce their job satisfaction as well as their organizational commitment (Allen & Meyer, Reference Allen and Meyer1990). Therefore, since some guanxi practices are naturally associated with perceptions of injustice (Fan, Reference Fan2002), an organization whose managers frequently engage in guanxi practices may suffer from lower organizational commitment among employees. However, given the different injustice implications of inward and outward guanxi practices, we can similarly expect that outward guanxi practices will more sharply reduce employees’ organizational commitment than will inward guanxi practices.
In addition, the private nature of guanxi sometimes tends to block employees from observing the actual guanxi exchange process (Yang, Reference Yang1994; Fan, Reference Fan2002). For this reason, compared with inward guanxi practices targeting to the operation of the focal firm, the real target and purpose of outward guanxi practices may not be clear to employees. Particularly, since outward guanxi practices include the utilization of collective organizational resources, any opacity may sharpen employees’ concerns about managers’ use of organizational resources for personal gain, therefore intensifying the negative impacts of employees’ negative reactions toward outward guanxi practices (Chen, Chen, & Xin, Reference Chen, Chen and Xin2004).
Hypothesis 4 : Outward guanxi practices are more negatively related to employees’ (a) trust in management and (b) organizational commitment than inward guanxi practices.
METHOD
Pilot study
Two of the constructs that our study develops – inward and outward guanxi practices – have not been used in prior research. For that reason, we conducted a pilot study to develop measures for them. We first held four focus group meetings in a university in Beijing with 28 MBA students (nine in the first group, seven in the second and fourth groups, and five in the third group), all of whom were middle-level or first-line managers before their admission into the MBA program. These meetings covered a series of discussions about guanxi, including the notions and content of inward and outward guanxi practices, the utility of different guanxi bases, and so forth. The focus group leaders were careful to emphasize that the concept of guanxi being discussed involved informal and interpersonal social ties.
Using information captured in these meetings, we developed 12 items designed to capture inward and outward guanxi practices (see Table 1). Then, we constructed a survey consisting of the newly developed measures together with well-developed items for capturing procedural justice perceptions (Tyler & Blader, Reference Tyler and Blader2000), trust in management (Mayer & Davis, Reference Mayer and Davis1999), organizational commitment (Becker, Reference Becker1992), and outcome favorability (Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). This instrument was tested in a pilot survey with 132 employees from four firms. The results of exploratory factor analysis obtained eight factors with eigenvalues >1 and accounting for 67.2% of total variance. All items in the survey loaded as expected. In addition, coefficient αs (reported in Table 1) indicated that all of the measures had good internal reliability (all of the αs in Table 1 are above or very near 0.8).
Results of factor analyses (EFA and CFA)
Note. The first alpha coefficient is obtained in the pilot study and the second in regular study.
CFA = confirmatory factor analysis; EFA = exploratory factor analysis.
Sample
Data collection was done as part of a large research program sponsored by the National Social Science Foundation of China. With the help of local governments and officially authorized industrial guilds in China, we chose 57 sample firms in nine Chinese provinces. We excluded newly founded firms (those younger than 3 years) to avoid the confounding effects of liability of newness (Xin & Pearce, Reference Xin and Pearce1996). We also eliminated small organizations (with annual sales <30 million RMB or assets smaller than 40 million RMB) so as to avoid effects related to the liability of smallness (Xin & Pearce, Reference Xin and Pearce1996; Jiang, Reference Jiang2009). Our sample firms averaged 6.21 years since founding.
Our informants consisted of two basic sets from whom the data were collected, respectively: top managers and employees (e.g., workers, first-line managers, supervisors, middle-level managers, etc.) in the 57 sample firms. Namely, for the sample top managers, we conducted semi-structured interviews to measure their inward guanxi practices, the guanxi bases in use, and information about their firms. These interviews generated 75 responses from the 57 firms (at least one for each firm).
For the sample employees, the data collection was accomplished with an online survey system established by the National Social Science Foundation of China project. In order to avoid collecting both independent and dependent variables with the same survey instrument (Podsakoff & Organ, Reference Podsakoff and Organ1986; Shadish, Cook, & Campbell, Reference Shadish, Cook and Campbell2002), we used a temporally lagged survey design including two phases. In the first phase, we gathered their assessment about the outward guanxi practices in their firms. In the second stage (∼12 weeks later), we captured these employees’ interpretations and reactions. At Stage 1, we posted our web address on each sample firms’ information system and asked employees to respond within 4 weeks. This Stage yielded 569 responses, of which 389 were complete. In addition, we dropped those respondents who were recruited by current employers within the past year, as we were concerned that they would not be familiar enough with their organizations to respond accurately. This led us to delete 31 responses. We followed up with the 180 respondents who provided incomplete responses, and asked them to complete the survey. This process yielded 48 more complete responses. T-tests of demographic variables showed no significant differences between first-round and second-round respondents, suggesting no serious selection bias. Six weeks after our follow-up request (about 10 weeks after the start of Stage 1), we initiated Stage 2, requesting all 406 complete respondents from Stage 1 to report their interpretations and reactions to managerial guanxi practices in their organizations. Stage 2 yielded 364 complete responses and this comprised our final sample. A total of 67% of our informants were male.
Variables and measures
Inward and outward guanxi practices
As mentioned before, inward guanxi practices typically align with the concept of managerial social capital (Luo, Reference Luo2000; Peng & Luo, Reference Peng and Luo2000). Therefore, we adapted the widely used six-dimension measurement of managerial social capital, capturing both business and political guanxi practices (Peng & Luo, Reference Peng and Luo2000; Acquaah, Reference Acquaah2007), in the semi-structured interviews with sample top managers to measure their inward guanxi practices. Similarly, we also used six items from the pilot study to measure outward guanxi practices for employees (see Table 1). Importantly, for our predictions about the implications of guanxi practices for employees’ reactions, both independent and dependent variables came from a single survey. To control for any potential common method variance problems (Graham & Collins, Reference Graham and Collins1991; Fowler, Reference Fowler1995) we adopted the build-out-self-test approach. That is, for each sample employee, we used the average score of all other employees in the same organization as the dependent variable in tests of employee reactions toward managers’ guanxi practices (Hypothesis 4). According to Howard (Reference Howard1994), this process significantly attenuates mono-method bias in organization studies.
Guanxi bases
We adopted the guanxi classification of Yang (Reference Yang1993) and Luo (Reference Luo2000) to measure different guanxi bases, with slight adjustments in some items based on suggestions received from our pilot study meetings with executives. Specifically, we designated family guanxi ties as lineal kin, near kin, and spouse's relatives, while familiar guanxi was designated as ordinary friends, acquaintances (classmates, colleagues, hometown fellows, etc.), and mediator-based guanxi (indirectly connected persons that are introduced by others).
Our measurement of guanxi practices is described in Table 1. We asked sample top managers the same set of questions regarding friendship-based inward guanxi practices, friendship-based outward guanxi practices, family-based inward guanxi practices, and family-based outward guanxi practices. The anchors for inward guanxi practice items were 1 = ‘almost none’ to 5 = ‘very extensive;’ and for outward guanxi practices the anchors were 1 = ‘totally disagree’ to 5 = ‘totally agree.’ As such, our survey design resulted in family-based, familiar-based, and overall inward guanxi practices, but only overall outward guanxi practices. Overall was the average across family- and familiar-based practices.
Employees’ interpretations and reactions
We adopted existing measures to capture employees’ perceived procedural justice (Tyler & Blader, Reference Tyler and Blader2000), trust in top managers (Mayer & Davis, Reference Mayer and Davis1999), and organizational commitment (Becker, Reference Becker1992), with slight adjustments in the Chinese translation. All items were measured with 5-point scales anchored by 1 = ‘totally disagree’ to 5 = ‘totally agree.’
Our study is designed to uncover the potential effects of managerial guanxi practices – a firm-level construct – on employees’ responses – an individual-level construct. If our measures are reliable, participants from the same company should have a high degree of convergence in their observations about managerial guanxi practices. To gauge this reliability, we used two within-group agreement measures – the intraclass correlation coefficient (see Kenny & La Voie, Reference Kenny and La Voie1985) and the interrater reliability coefficient (IRR – see James, Demaree, & Wolf, Reference James, Demaree and Wolf1984). The intraclass correlation aims at ensuring that the variance between different sample groups is significantly greater than the variance within groups, and uses a one-way analysis of variance approach to gauge the difference. The IRR indicates the proportion of variance explained by group membership (the organization) relative to the expected variance explained with a uniform null distribution. As such, the IRR reflects the degree to which participants from the same group agree in their assessments. Generally, an IRR coefficient above 0.50 signifies moderate agreement, while a coefficient above 0.70 indicates substantial agreement (James, Demaree, & Wolf, Reference James, Demaree and Wolf1984). According to this criterion, the results reported in Table 2 indicate significant intraclass correlation values for all 10 organization-level measures, with coefficients ranging from 0.63 to 0.76 (p < .001). Furthermore, median IRR for all organization-level measures ranged from 0.62 to 0.74, which, according to the aforementioned criterion, indicate moderate to high levels of IRR. Taken together, these results suggest good within-group agreement for our self-report data, and good reliability for the measures in our survey.
Means, standard deviations, and correlationsFootnote a
Note. a r > |0.013| are significant at p < .05.
ICC = intraclass correlation; IGP = inward guanxi practice; IRR = interrater reliability; OGP = outward guanxi practice.
Control variables
As discussed earlier, information about control variables were collected from top managers. As mentioned before, prior organization studies have pointed out that highly favorable outcomes can justify otherwise distasteful processes, thus increasing employees’ perceptions of procedural justice (Lind & Tyler, Reference Lind and Tyler1988; Ambrose, Harland, & Kulik, Reference Ambrose, Harland and Kulik1991) and evaluations of managers (Clark & Payne, Reference Clark and Payne1997). In addition, favorable outcomes can significantly enhance employees’ overall satisfaction, therefore amplifying their organizational commitment. Therefore, we adopted outcome favorability as a control variable in our model. This variable was measured by two existing items developed by Tyler and Blader (Reference Tyler and Blader2000) and reported in Table 1.
We also included age and firm ownership as two control variables in our model. Prior research indicates that employees’ perceptions of justice and organizational commitment tend to rise with employee age, such that older employees tend to have higher tolerance for and acceptance of managerial guanxi practices (Su, Sirgy, & Littlefield, Reference Su, Sirgy and Littlefield2003). Regarding firm ownership, Chen, Chen, and Xin (Reference Chen, Chen and Xin2004) and Jiang (Reference Jiang2009) both concluded that guanxi practices tend to be interpreted more negatively by employees in state-owned firms relative to those in non-state-owned firms. We measured firm ownership with a dummy variable in which state-owned firms were designated as 1 and non-state-owned were designated as 0. Preliminary results showed no statistically significant difference between the intensities of the overall inward and outward guanxi practices in state-owned firms and non-state-owned firms.
RESULTS
Means, standard deviations, and correlation coefficients are reported in Table 2. As expected, the correlations between inward and outward guanxi practices are basically positive and significant, and the correlations between guanxi practices and employee responses tend to vary across inward and outward guanxi practices.
Regarding our hypotheses, correlation coefficients indicate that outward guanxi practices are significantly and negatively related to perceptions of procedural justice (r = −0.31, p < .01) and trust in management (r = −0.22, p < .01), while the correlations between inward guanxi practices and employees’ interpretation and reactions are often positive. Moreover, as expected, inward and outward guanxi practices are positively related to each other (r = 0.47, p < .01). Generally, these preliminary results provide suggestive support for our predictions.
We adopted a hierarchical multiple regression method to test our four hypotheses. Testing results are reported in Tables 3 and 4. Hypothesis 1 predicted the existence of inward−outward guanxi contagion. We used two regression models to test this hypothesis, in accordance with the two potential directions of the inward−outward guanxi contagion mechanism. Outward guanxi practice is regressed on inward guanxi practice first, and then inward guanxi practice regressed on outward guanxi practice. The results in Table 3 provide strong support for Hypothesis 1. For example, inward guanxi practices have a significant and positive impact on outward guanxi practices (γ = 0.210, p < .001) and vice versa (γ = 0.194, p < .001). When the complete model is compared with the controls-only model, there is significant improvement in the R 2 of the inward−outward contagion model (ΔR 2 = 0.040, p < .001), and for the outward−inward contagion model (ΔR 2 = 0.037, p < .001).
Hierarchical regression results of inward−outward guanxi contagion
Note. IGP = inward guanxi practices.
†p < .10; *p < .05; **p < .01; ***p < .001.
Hierarchical regression results of overall guanxi practices and employees’ reactions
Note. †p < .10; *p < .05; **p < .01; ***p < .001.
Hypothesis 2 extended the basic inward−outward guanxi contagion model by predicting guanxi base as a moderator of the positive relationship between inward and outward guanxi practices. Since it includes the comparison of the effect sizes of two independent variables, we adopt a three-step hierarchical multiple regression model to test the hypothesis. In Step 1, only control variables were entered into the model. In Step 2, family-based inward guanxi practices are added. In Step 3 familiar-based guanxi practices are added. Then, the order of the last two steps was reversed to check the change in effect sizes because of order of entry.
According to Table 3, in the complete model, both familiar-based inward (γ1 = 0.261, p < .001) and family-based inward guanxi practice (γ2 = 0.118, p < .05) have significant and positive effects on outward guanxi practice. Also, when all other four predictors are controlled, entering familiar-based inward guanxi practices significantly improves the predictive power of the model (ΔR 2 = 0.075, p < .001). However, the power gained from adding family-based inward guanxi practices is insignificant (ΔR 2 = 0.008, ns). In addition, by comparing the complete model to a restricted model that requires r 1 = r 2, we find that the complete model is statistically different from the restricted model (ΔF = 36.88, p < .001), further suggesting that familiar-based inward guanxi practice has stronger positive effects on outward guanxi practices than family-based guanxi practice. Taken together, this evidence strongly supports Hypothesis 2.
Hypothesis 3 predicted that outward guanxi practices tend to incur more negative perceptions among employees than inward guanxi practices do. As can be seen in Table 4, after all other independent variables are controlled, entering inward and outward guanxi practices in Step 3 substantially adds to the explanatory power of the regression model (for inward guanxi practice, ΔR 2 = 0.020, p < .01; for outward guanxi practice, ΔR 2 = 0.0710, p < .001). But, the coefficient for outward guanxi practice is negative (γ2 = −0.276, p < .001), while the coefficient for inward guanxi practice is positive (γ1 = 0.17, p < .01). Moreover, in order to compare the effects of inward and outward guanxi practices, we also introduce a restricted model that requires r 1−r 2 to be negative. The comparison of the restricted model and the unrestricted model suggests that these two model are significantly different from each other (ΔF = 41.12, p < .001), further supporting Hypothesis 3 and bolstering the notion that outward guanxi practices can more strongly and negatively affect employees’ justice perceptions than inward guanxi practices.
Using similar logic, Hypothesis 4 predicted that outward guanxi practices would have more negative effects on employees’ trust in management and organizational commitment than would inward guanxi practices. According to Table 4, outward guanxi practices significantly attenuate employees’ trust in management (γ2 = −0.227, p < .001), and inward guanxi practices marginally enhance employees’ trust in management (γ1 = 0.110, p < .10). Moreover, when entered in the regression model in Step 3, outward guanxi practice added substantially to the predictive power of the model (ΔR 2 = 0.049, p < .001), and while the improvement obtained by entering inward guanxi practice in Step 3 is also (marginally) statistically significant (ΔR 2 = 0.010, p < .10) it is less so. A direct model comparison indicates significant differences between the complete model and the restricted model designating r 1−r 2 to be negative (ΔF = 14.37, p < .001). These results together provide strong support for Hypothesis 4a.
However, Hypothesis 4b is not supported by the evidence reported here. The complete model indicates that neither inward nor outward guanxi practices affect employees’ organizational commitment. Further, neither inward nor outward guanxi practices significantly improves the R 2 of the model with when entered in Step 3 (see Table 4).
We also conducted an ancillary test in which inward guanxi practices with different guanxi bases were tested as two separate variables. Results (see Table 5) indicate that after we controlled the effects of three control variables and outward guanxi practice, familiar-based inward guanxi practice does significantly improve employees’ perceived procedural justice (γ = 0.142, p < .05) and trust in management (γ = 0.115, p < .05), while family-based inward guanxi practice has only insignificant effects (for perceived procedural justice, γ = −0.024, ns; for trust in management, γ=0.047, ns). However, neither of these two inward guanxi practices significantly impacted employees’ organizational commitment. Therefore, Hypothesis 4b is not supported.
Hierarchical regression results of family-/familiar-based guanxi practices and employees’ reactions
Note. IGP = inward guanxi practices.
†p < .10; *p < .05; **p < .01; ***p < .001.
DISCUSSION
In our study, we tried to develop a new perspective on guanxi research by separating managerial guanxi practices into inward-focused and outward-focused depending on the direction of the flow of resources. Inward guanxi practices secure benefits for the focal organizations, whereas outward guanxi practices involve providing firm resources to outside players. We described a natural linkage, or contagion, between these two practices, arguing that the use of one naturally tends to induce the use of the other. We further described how the contagion between inward (outward) and outward (inward) guanxi practices is moderated by the base of the guanxi – family or familiar. This new perspective helps to integrate the prior research on the benefits of guanxi practices in organizations with the prior research on the costs of such practices.
Generally, our results support the existence of contagion between inward and outward guanxi practices. Further, our evidence on the moderating effect of family-based guanxi practices supports our contention that the contagion between inward (outward) and outward (inward) guanxi practices derives from very strong norms of mutual obligation and reciprocity that characterize familiar guanxi, but are much weaker or non-existent for family guanxi. Guided by the norms of reciprocity, familiar guanxi participants are motivated to maintain the balance of favors exchanged through their guanxi relationships so as to ensure the sustainability of guanxi exchange in the long run. Because of these strong norms, much guanxi involves a tradeoff of value flows into and out of organizations through managers’ guanxi relations. In effect, outward guanxi practices can serve either as prepaid guanxi investments to secure access to future inward guanxi-based resource flows or as the consequence or repayment of earlier inward guanxi practices.
Our study also considered some costs associated with the use of guanxi practices in organizations, but specifically linked to outward guanxi practices. We noted that the nature of guanxi practices of any sort inherently involves benefiting guanxi participants at some cost to others – including costs to society (Luo, Reference Luo2000; Dunfee & Warren, Reference Dunfee and Warren2001), and that these practices generally violate norms of fairness and justice. More specifically, using managerial guanxi practices in organizations may also incur some very direct costs, that is, employees’ negative interpretations and reactions (e.g., Chen, Chen, & Xin, Reference Chen, Chen and Xin2004). Our study partially confirmed the existence of these implicit guanxi costs, supporting our contention that not all managerial guanxi practices are associated with negative responses among employees. In fact, our evidence suggested that only outward guanxi practices significantly attenuate employees’ justice perceptions and their trust in top management. Inward guanxi practices, by contrast, tend to strengthen employees’ perceived justice and trust in management. Of course, the fact that inward-focused and outward-focused guanxi practices are intrinsically linked lends caution to any recommendations made to managers regarding the use of guanxi practices in organizations. Specifically, while employees may applaud inward-focused guanxi practices, the adoption of those practices leads naturally to the use of outward-focused guanxi practices, so employees are likely to eventually be upset if guanxi practices are adopted in organizations.
We examined the different employees’ responses to inward and outward guanxi practices through a behavioral theory lens. That is, practices that are associated with value inflow into an organization will naturally be perceived as more just than parallel practices that are associated with value outflow from the organization. Put differently, the self-interest model of justice (Lind & Tyler, Reference Lind and Tyler1988; Ambrose, Harland, & Kulik, Reference Ambrose, Harland and Kulik1991) claims that favorable outcomes help to justify actions, and tend to enhance employees’ justice perceptions. Our study represents an extension of the self-interest model of justice to a new setting, and a new level of analysis, as we argue that these practices are an organization-level phenomenon.
Our study supports the argument that value consequences tend to surpass the effects of the neutrality principle in accounting for employees’ responses toward managerial guanxi practices. Nevertheless, the neutrality principle and the rule of bias suppression are both likely to play important roles. According to the results of our ancillary analysis, inward guanxi practices based on familiar guanxi are more significantly and positively related to employees’ perceived justice and trust in management than family-based inward guanxi practices, suggesting that with comparable value implications, the more intense the affective concern involved in guanxi bases, the more negative the employee responses incurred.
One unexpected result in the present study is that our hypothesis regarding the effects of guanxi practices on employees’ organizational commitment was not supported. According to our analyses, none of the managerial guanxi practices had significant effects on employees’ organizational commitment. Instead, two control variables, outcome favorability and participants’ age, showed significant influences on employees’ organizational commitment. We believe that our results may be due to the different foci of employees’ commitment. According to Becker (Reference Becker1992), employees can be committed to either their organizations or their supervisors. Therefore, it is possible that managerial guanxi practices, as actions of managers, might mainly affect employees’ perception and reactions toward managers and the procedures under their control instead of the organization. In other words, employees may mainly attribute the negative implications of managerial guanxi practices to their managers instead of to their organizations. In addition, the norms of employment in China tend toward the long term in that individuals do not change employers frequently in China.
Taken together, we can see that these results of our study depicted an organizational-level cost mechanism of managerial social capital with a specific instance in Chinese society, namely, inward guanxi practices. First, given the intrinsic instrumental nature of guanxi, using inward guanxi practices to benefit the focal organizations, which are widely referred to as the utilization of ‘managerial social capital’ (Peng & Luo, Reference Peng and Luo2000; Acquaah, Reference Acquaah2007) or ‘guanxi assets’ (Tsang, Reference Tsang1998; Luo, Reference Luo2000), may inherently require certain organizational wealth for exchange, thus leading to the direct or explicit cost mechanism – inward−outward guanxi contagion. Moreover, although inward guanxi practices per se may not incur the ethical dissatisfaction and negative responses of employees, their direct consequence – outward guanxi practices – tend to significantly attenuate employees’ interpretations and reactions, thus resulting in an indirect cost of inward guanxi practices. In this regard, our study provides a new perspective for considering the costs and benefits of using managerial social capital.
Taking a broader perspective, we can see that the inward−outward guanxi contagion mechanism could be valuable for business operations and the overall market system in China, despite some negative implications for employees and individual organizations. That is, through guanxi exchanges involved in inward and outward guanxi practices, a manager can serve as the bridge through which the firm can effectively exchange resources with outside networks with certain managerial guanxi practices (Luo & Chen, Reference Luo and Chen1997). Top managers can use guanxi practices to bridge structural holes and broker opportunities embedded in their guanxi networks by conducting outward guanxi practices in exchange for the access to advantageous inward guanxi practices (Burt, Reference Burt1992, Reference Burt1997). As such, the inward−outward guanxi contagion can serve as an informal substitute for more formal approaches to resource and value exchange. Particularly, in the emerging market in China with relatively low factor fluidity and weak institutional support, this substitutive exchange channel may be of higher value to business organizations (Xin & Pearce, Reference Xin and Pearce1996).
Limitations and future extensions
Our study has some important limitations. First, we focused exclusively on employees’ subjective consequences of guanxi practices in Chinese organizations. It would be good for future studies to incorporate both objective and subjective measures of managerial guanxi practices. We did take steps to ameliorate any common-method biases from our methodology by separating the collection of independent and dependent variables and using the build-out-self-test methodology. But, more objective measurement would have been beneficial.
Second, although we focused directly on guanxi practices in our research, we did not actually map out guanxi networks in Chinese organizations. For that reason, our observed managerial guanxi practices could only be measured in a static way and through the general perceptions of top managers and employees about the overall intensity of managers’ guanxi practices. This may lead to potential problems. For example, one may argue that despite its importance, family guanxi practice is much less frequent than familiar guanxi largely due to limited number of ties among the family members. Accordingly, the relatively weak effects of family-based inward guanxi practice may arise from the lack of practice. However, since in this study we did not measure the frequencies of inward guanxi practices on different guanxi bases, we cannot effectively rule out such possible alternative explanation. Likewise, based on such static information, we may not be able to effectively compare the favorable and unfavorable implications of a specific guanxi practice over time and determine the actual implications of this managerial practice. In-depth studies using a dynamic way and capturing both overall guanxi networks and network-level guanxi practices would allow a more rigorous test of the independent and interaction effects of structural and behavioral aspects of guanxi.
Moreover, as mentioned earlier, our study provides a new perspective on managerial social capital, namely, simultaneously examining its costs and benefits from both inside and outside the organization. However, given the cross-level nature of this research question, our actual sample size was limited, potentially limited the robustness of our results. In such a manner, it is important for future studies to examine our conclusions with bigger samples and in different contexts.
Acknowledgements
This study is supported by Chinese National Natural Science Foundation NO. 71072009.
