The benefits of high-quality relationships come from relational resources (Wright, et al. 2005) they create. Such resources include durable obligations (e.g., arising from feelings of gratitude, respect, and friendship), network contacts and connections (including privileged access to information and opportunities, social status, and reputation of influential others), and the ability to have open information exchanges with those around them (Valle & Halling, 1989).
Relationships that do not develop so well are considered lower quality. These relationships are not as beneficial for the individuals involved or for the organization as a whole (Gerstner & Day, 1997; Liden et al., 1997). Lower quality relationships are described as contractually defined, formal exchanges based on limited trust and in-role interactions (Luthans, 1998). These types of relationships generate management rather than leadership. They are characterized by lack of mutual respect, formal downward communications, little mutual understanding, limited support and commitment for one another, and no mutual obligation (i.e., a "stranger" relationship) (Graen & Uhl-Bien, 1991a). Findings have shown that lower quality relationships are negatively related to satisfaction, organizational citizenship behaviors, and commitment, and are positively related to turnover (Gerstner & Day, 1997). Uhl-Bien and Maslyn (2003) recently found evidence of an even more extreme case of low-quality relationship, which is characterized by negative reciprocity, or an exchange of injuries (e.g., negative social exchange, Ruehlman & Karoly, 1991).
Thus, based on relational leadership theory, effective relationships may generate mutual influence and understanding that allow leaders to more effectively perform their roles.
Assumption 3. Leadership influence to make alteration is enabled by effectual relationships.
Yet, despite the value of high-quality relationships for organizations, not all relationships are high quality. Given the findings from LMX theory, we know that low-quality relationships are not beneficial in terms of many aspects of organizational functioning, so they are not desirable in organizations, but they still are prevalent (Gerstner & Day, 1997). Moreover, in some cases we may even have negative, or dysfunctional relationships (Uhl-Bien & Maslyn, in press). Why is this, and what can we do about it? (Bass, 2005)
Beyond LMX Differentiation
The LMX literature says that LMX differentiation, in which leaders have higher quality relationships with some subordinates and lower quality relationships with others, occurs because leaders do not have time (or the need) to generate high quality relationships with everyone (Cox & Beale, 2008). Therefore, they develop a group of trusted assistants to help them perform the work of the unit. These trusted assistants would supposedly be the best or most reliable employees in the unit.
Twenty years after the inception of the theory, Graen and Uhl-Bien (1995) argued that the creation of "in-groups" and "out-groups" within work units is not beneficial, and that instead leaders should strive to develop high-quality relationships with all subordinates. They do this by "making the offer" of high-quality relationships to all and then through testing processes, different quality relationships result (Luthans, et al. 2008). This perspective allows for the fact that all relationships may not (and likely will not) reach high quality, but at least the dyad members both take part in how the relationship develops (rather than the leader determining who will be the trusted assistants) and have the opportunity to create a high-quality relationship.
This perspective also recognizes that a focus on differentiation rather than on high-quality relationships offered to all creates tremendous opportunity for lost potential in organizations. As noted by Organ (1990), when individuals are not fully committed (or are dissatisfied), they will withdraw discretionary behaviors that benefit others or the organization (e.g., helping, altruism, civic behaviors). These discretionary behaviors are beneficial to the organization (Luthans, et al. 2008) and as a result, much attention has been given in the literature to determining when and how individuals engage in these behaviors (Graen et al., 2010).
Instead of LMX differentiation, therefore, our goal should be for individuals to strive to have influence with one another (and with higher-ups). With the support of a relationship, individuals are freer to open up and provide one another with more accurate and complete information (Kotter & Cohen, 2002) so they can provide the "real" information (the "real" story). This goes both ways: with a good relationship comes reduced filtering (holding back) of information, both up and down the hierarchy (Fairhurst & Sarr, 1996). It allows individuals to share with one another the hard truth. Too many leaders do not have good information, and too many hold back in being truthful with their subordinates (House, 2003). If we extend this beyond managers to leaders more broadly, then we can argue that organizational members need to be comfortable with providing information to one another, and this comes with having effective work relationships (Luthans, 1998).
Remembering that effective leaders are defined as those who use influence to create change, individuals' abilities to be effective leaders are directly related to their ability to have influence in the organization. Since effective work relationships can extend individuals' influence networks (Drath, 1998), those who have more effective relationships with others will likely have more opportunity to gain and use influence (Kotter & Cohen, 2002). Therefore, leadership effectiveness is likely enhanced by the ability to build effective work relationships with a broader range of interdependent others (Luthans, 1998).
Proposition 1a. Leadership effectiveness is improved by the individual's aptitude to construct effective work relationships with mutually dependent others. Moreover, given the benefits of high-quality relationships and the lost potential (Organ, 1990) or harm of low-quality relationships (Northouse, 2007), more effective work situations are those with leader/managers who develop more high-quality relationships or fewer low-quality relationships.
Proposition 1b. More effectual leaders are those who are able to construct relationships with a wider range of others (instead of with only a select few).
A New Twist on Relationship Development
Both of the approaches (described previously) regarding LMX differentiation are highly manager driven: The "leader" is the primary responsible party for the quality of the relationship, either by selecting the trusted cadre (Bass, 2005) or by making offers to all (House & Podsakoff, 2003). The notion of leader control seems to be an assumption throughout LMX and leadership research (Cox & Beale, 2008). As noted by Lord and Emrich (2001), the leadership literature often assumes that causality (for leadership) originates in a leader, but this is likely an over simplification, because leadership processes always involve an interaction of leader, subordinate, and contextual qualities.
Recently, House & Podsakoff, (2003) considered manager -- subordinate relationship development from the standpoint of investigating who put effort into the relationship development: the manager or the subordinate. Consistent with reciprocity and social-exchange perspectives (House & Podsakoff, 2003), the findings showed that own effort in relationship development was not significant, but other effort was highly significant. In other words, both parties were responsible for relationship development through the amount of effort that each perceived the other to have put into the relationship (e.g., norm of reciprocity, Gouldner, 1960). Moreover, these findings were consistent across managers and subordinates. This suggests that perhaps we need to rethink some traditional notions about how relationships develop to focus more generally on the roles and responsibilities of both dyad members in relationship building.
We may also need to rethink commonly accepted conceptualizations about what constitutes a high-quality relationship. In a criticism of LMX theory, Luthans, et al. (2008) stated that high-quality relationships have been identified in the literature as having trust, respect, mutual obligation (loyalty) and influence, and wide latitude for discretion. House and Aditya (1997) pointed out, however, that these may not be universal attributes of high-quality relationships. In other words, ideas about what is considered to be a high-quality relationship may, and likely will, vary according to the members of the relationship. Moreover, it suggests that we need to more carefully consider the antecedents to relationship development relative to what each dyad member wants from the relationship.
Implicit Relational Theories
Taken together, and combined with new perspectives to be discussed here, these ideas present some very interesting possibilities regarding how we may view the development of managerial relationships in the workplace. First, considering relationship development specifically, it is possible that some relationships are easier, more "natural" to form than others. For example, in some relationships individuals may "hit it off" from the beginning, for whatever reason. This could be because they have complementary personalities, common values, congruent perspectives or interpersonal styles, similar backgrounds, and so on, such that the relationship gets off to a good start and just keeps going, with neither member really thinking about or consciously managing the process. Other relationships, however, may be much harder to develop due to personality differences, style differences, incongruent values, and so on, that make it more difficult for dyad members to build the relational components necessary for higher quality relationships. These relationships that are less compatible and require more effort to develop may be the ones that result in lower quality relationships. In contrast, those that are easier to develop may become the higher quality relationships.