By Nathan Harter Purdue University and Donna Evanecky, Purdue University at Kokomo

Academic Citation: Nathan W. Harter and Donna Evanecky, “Fairness in Leader-Member Exchange Theory: Do We All Belong on the Inside?” Kravis Leadership Institute Leadership Review, Summer 2002.

An abridged version of this paper was presented in March, 2002, at the thirty-second annual conference of the Popular Culture Association, in Toronto.

About the Authors: Nathan Harter has taught Organizational Leadership for twelve years at Purdue University, after practicing law in Rising Sun, Indiana. Associate Professor Harter has pursued studies into the philosophical underpinnings of leadership. Donna Evanecky joined the Purdue University School of Technology at Kokomo as Assistant Professor for the Department of Organizational Leadership and Supervision in August 2001. Evanecky has eight years of supervisory, engineering, and quality experience, along with extensive education, teaching, and technical expertise. The past five years of Evanecky’s career were spent at DaimlerChrysler’s Indiana Transmission Plant in Kokomo.

INTRODUCTION

Leader-Member Exchange theory (or LMX) describes differences in the way leaders treat the in-group and the out-group. LMX then prescribes bringing all followers into the in-group. This paper questions the justice of arrangements both described and prescribed by LMX. In what respect are distinctions between in-groups and out-groups unfair? Are these distinctions eradicable in fact? Would it even be fair to try eradicating them?

THE IMPORTANCE OF FAIRNESS TO LEADERSHIP GENERALLY

A number of writers emphasize the role of fairness in the relationship between a leader and a follower, even if they do not agree on its importance or what exactly constitutes being fair. Jerald Greenberg, one of the most prominent authors in the field of organizational justice, makes the observation that people all have perceptions of fairness. People behave as though it matters (Greenberg, 1996). So, for all intents and purposes, in the leadership process, fairness matters. This is a reassuringly empirical place to begin.

In what way does it matter? The answer to this question is not so clear. Going all the way back to The Prince (Machievelli, 1531/1991), the perception of fairness matters a great deal because followers believe that it matters. It is that simple. For this reason, a prudent leader would certainly try to seem fair. The truth of the matter (Machiavelli argued) is that leader effectiveness is more important ultimately than fairness, so a leader may in fact be unfair when it serves his ends – so long as he remembers to keep up appearances. Fairness is an expectation that followers have and little else. What a leader must remember is that a follower’s perception is their reality. Seem to others to be fair, he advises, and you can literally get away with murder.

Machiavelli was responding to a series of writings about political leadership that had told leaders to practice virtue. He changed the tone of the discussion by saying that in reality, doing the right thing morally or spiritually often meant doing the wrong thing when it came to being effective. In other words, being fair does not always work. Right does not make might. Virtue can ruin leadership. Effectiveness and ethics are two separate things.

Many writers since have tried to refute the old Florentine, with varied success. The noted philosopher John Rawls insists for example, “Justice is the first virtue of social institutions…. [L]aws and institutions no matter how efficient or well-arranged must be reformed or abolished if they are unjust.” Rawls then wrote that “one may think of justice as fairness…” (Rawls, 1971, p. 111). His famous principle of fairness “holds that a person is required to do his part as defined by the rules of an institution when two conditions are met: first, the institution is just (or fair),…and second, one has voluntarily accepted the benefits of the arrangement or taken advantage of the opportunities it offers to further one’s interests” (Rawls, 1971, p. 111f). We shall return to this principle soon. The point is that not everyone agrees in what way fairness matters.

So, people all have perceptions of fairness, but they do not agree. Why should leaders try to understand and overcome these differences? The answer is that patterns of fair dealing build trust. Trust accumulates, making relationships more reliable. Reliability makes relationships more efficient, if not more pleasant, including the relationship between a leader and a follower, (e.g., Gardner, 1990, p. 33; Fukuyama, 1995, p. 21; & Young, 1991, chap. 2). This is part of the reason why issues of fairness and trust lie at the heart of leadership.

This paper questions the justice of arrangements both described and prescribed by Leader-Member Exchange Theory. In what respect are its distinctions between in-groups and out-groups unfair? Are these distinctions eradicable in fact? Would it even be fair to try eradicating them? The first step in this paper is to explain Leader-Member Exchange Theory and its ethical implications.

WHAT DOES LMX DESCRIBE?

Leader-Member Exchange Theory (also known at one time as vertical-dyad linkage) goes back over twenty-five years (Northouse, 2001, p. 111), and builds on the seminal work of the sociologist Georg Simmel, nearly one hundred years ago. In its current form, Leader-Member Exchange Theory (or LMX, for short) emphasizes that the relationships a leader has with followers can be divided into two types. These two types of relationship tend to divide followers into two groups – one group for the one type of relationship and another for the other type. The two groups are referred to as the “in-group” and the “out-group”. Again, the determining factor as to which sub-group a follower belongs to has to do with the nature of that person’s relationship with the leader, (Northouse, 2001, p. 112). This is important because it places a great deal of responsibility for the split on the leader directly and not on cliques or coalitions. Thus, the ethical implications of LMX are to be traced to the leader, which is one of the theory’s limitations.

The distinction between the two groups is not arbitrary. According to LMX, the in-group can be distinguished from the out-group in the following manner. Members of the in-group will have more responsibility and decision-making influence than those in the out-group. Quite probably, the in-group member will have higher job satisfaction and access to valuable resources, such as money, time, and authority because of their status with the leader. Along with these, the leader will demonstrate trust, support, collective goals, as well as initiative beyond the scope of the everyday job toward the member of the in-group, in exchange for more effort and enthusiasm (Northouse, 2001, p. 114-115, 118).

In contrast, members of the out-group are given little support or responsibility, along with minimal influence in decision-making. Overall, lack of interaction with out-group members by the leader is apparent. For example, perhaps the leader disagrees with a member’s views and work style. The leader may interact very little with the member because of this dissimilarity. Thus, it is likely that the member and the leader will develop a low quality relationship, ultimately driving the member into the out-group. He or she becomes marginalized.

This distinction between groups goes all the way back to biblical times. In the Old Testament, II Kings 14:28 refers to King Jeroboam’s administration ignoring policies of justice and fairness. As a result, the rich became richer, and the poor, poorer. There was an in-group and an out-group created by the different relationships they had with the king. This historical anecdote correlates with organizations today with respect to the in-group getting more than the out-group. People in the out-group are so oppressed or invisible that it is hard for the leader to notice their plight. Instead, the leader continues to give more challenge, freedom, and responsibility to the in-group members. They in turn do more to justify their status, which puts them at the forefront of the leader’s mind the next time a job needs to be done and the next time a perquisite comes along, and so it continues in a spiral: the rich do get richer.

In other words, LMX describes a disparity that can be traced to the leader. Something the leader says or does over time contributes to the creation of two unequal factions. On the surface, this seems unfair. Disparities of any kind are highly suspect, especially if one begins from an egalitarian viewpoint that everyone ought to be treated the same. Making matters worse is the tendency for the disparity to become more pronounced and permanent, so that people become fixed in their groups, notwithstanding their talent or efforts. Once the groups take shape, they tend to stay that way and rigidify. This too seems unfair, because now members begin each new day on an uneven playing field. The out-group would have to go to extraordinary lengths to reach the status and power already being enjoyed by the in-group, and it becomes simpler to give up.

The disparity supposedly affects productivity, and that makes sense (Northouse, 2001, p. 115 & 119). Once the disparity strikes participants as unfair (Northouse, 2001, p. 120) and they conclude that the leader either caused it or at least has not repaired it, they are likely to question the leadership itself. That is not good. So what does LMX recommend that the leader do?

WHAT DOES LMX PRESCRIBE?

LMX proceeds as though the division into in-groups and out-groups is both unproductive and unfair. Consequently, it offers to a leader certain prescriptions to prevent or fix the situation (Northouse, 2001, p. 115-118).

LMX prescribes first that leaders become cognizant of their attitudes toward all of their members. Certainly a leader may not like some people as much as others, but a leader must separate personal feelings from the task at hand. The task is leadership. If a member sees that a leader treats members differently, especially over time, he or she may perceive inequity. As the Equity Theory attributed to J. Stacey Adams (1965) states, a member perceives what they can get from a situation (outcomes) in relation to what they put into it (inputs), and then compare their input-outcome ratio with the input-outcome ratio of others. If an inequity is perceived, the member will attempt to correct it, perhaps by reducing effort or quitting altogether.

Say for example that six people work for a given manager. All but one of these employees is male. As time passes the female begins to realize the boss is not treating her as he does her co-workers, even though she works longer hours and completes more successful projects than do her co-workers. She is likewise not given extra responsibility, freedom, or special perks. She is in the out-group. Or, in this instance, she is the out-group. As time goes on, she begins to do things to make the situation equitable -- things like discontinuing her long hours, losing dedication, and ultimately resigning. Although this is an illustration, it is not fiction; it actually happened to one of the authors in a previous career.

Can and will a leader stop this segregation of in-group and out-group from occurring? That is the question LMX expects a conscientious leader to answer. Warren Bennis has been quoted as saying, “Good leaders make people feel that they're at the very heart of things, not at the periphery. Everyone feels that he or she makes a difference to the success of the organization. When that happens people feel centered and that gives their work meaning.” (Edu-Leadership.com) This is the first step under LMX that a leader must take toward ending the segregation of in-group and out-group members. The leader’s success depends upon grasping this truth. It has been said that for a leader to be successful, they must surround themselves with competent people. More likely than not, a leader will have very competent people within the out-group, but if he or she is not using these valuable resources fully, then they cannot contribute to the leader’s success. At the end of the day, if not everyone is a member of the in-group, then the leader has failed and has jeopardized the potential for success. This at least is the prescription for managers.

In the end, LMX advises a leader to avoid the temptation to participate in creating disparities and to become especially sensitive to the perceptions of followers, so that as a consequence all followers might be brought within the in-group.

QUESTIONS OF FAIRNESS

LMX is coherent, and it takes workplace justice seriously. For these two reasons, if for no other, it deserves further study and respect. Our purpose in this section is to raise questions about the ethics of LMX: Is it fair? By the same token, is the situation it hopes to remedy unfair?

1. As a preliminary matter, we would ask whether LMX is overly simplistic, offering an either/or schema (one is a member either of the in-group or of the out-group) rather than a more subtle continuum of being more or less “in” as a matter of degree, so that clear groups are really the product of one’s imagination or interpretation, like polar extremes on a graduated scale? Could that not be possible at least sometimes? Andy is more favored than Betty, who in turn is more favored than Clark, and so on. The assumption that there are in fact two distinct groups, as opposed to degrees of being “in”, creates the illusion of exclusivity. We shall have perceived as solid (i.e. groups) what is in reality quite fluid. In addition, we would ask whether a person might be “in” with regard to one thing and “out” with regard to another? Or does LMX hold that it is an all-or-nothing deal; one is either “in” all the way or “out” all the way? That would seem unnecessarily stark as well (See Northouse, 2001, p. 121, citing Graen & Uhl-Bien, 1995). Finally, it was Simmel who wrote about “multiple group-affiliations within a single group” to the effect that persons constituting a group might also find themselves in competition and forming sub-groups for selected purposes, so that the in-group (if there is such a thing) might be further subdivided into factions struggling against each other, even though they prove unified against the out-group (Simmel, 1955, p. 155f). What does the presence of these smaller sub-groups signify? That is a set of three questions right there, even before we turn to questions of fairness.

2. Even if LMX does lead to a clear split into two groups, is it truly preventable? This is an important question, because theories of ethics rarely blame someone who is without sufficient means to have done otherwise. It makes no sense to blame anyone for what is inevitable. Can a group (or leader) avoid the condition LMX describes? One could argue that such splits are inevitable, that in-groups will start to form and tend to congeal into elites. (See e.g. Levine, 1995, chap. 11, p. 281 regarding what is known in sociology as elite theory, citing Mosca, Pareto, Michels, Lasswell, and Deutsch).

   The Italian Vilfredo Pareto, for example, once wrote the following: “Except during short intervals of time, peoples are always governed by an elite…meaning the strongest, the most energetic, and most capable…” (Paredo, 1991, p. 36). He was part of a much longer tradition about the tendency of groups to stratify, going back to Aristotle. One could ask therefore whether LMX correctly detects a pattern as a matter of sociological law, but then it goes wrong trying to convince leaders they can do something about it. Here it would be helpful to point out, in defense of LMX, that for Pareto at least, “elites do not last.” Instead, they participate in a pattern of “continuous replacement” (Pareto, 1991, p. 36). According to this principle, the rich will not keep getting rich forever; they will gradually lose control and fall from favor, to be replaced by leaders from the out-group. In this respect, Pareto disagrees strongly with the idea that divisions between in-groups and out-groups are permanent. Of course, permanence is relative. Pareto spoke in terms of decades and centuries for the transition to occur. LMX would seem to take the position that the out-group should not have to languish for decades until social forces gather enough pressure to change things. Such would be an example of “justice delayed” being “justice denied.” LMX has greater faith in the efficacy of leaders apparently than elite theorists would have.

3. The question whether elites are inevitable deserves a second look. Can it be said that LMX describes one stage in the evolution of groups into hierarchy or elites? What LMX is describing is just a micro version of something that all firms experience as they differentiate and organize. As groups grow in size and complexity, they have to subdivide somehow. That is part of what we mean by becoming organized. Taking this a step further, don’t larger organizations require hierarchy and elites? Or at least, is not the existence of hierarchies and elites ethically neutral, when it comes to organizational structure?

   We rarely condemn large corporations for having layers of authority, managers, because that is a necessary piece of the bureaucratic model. If anything, it is part of the division of labor, a step toward greater clarity and efficiency, when firms break out in this way. By this logic, LMX condemns what amounts to a necessary phase in the evolution of all firms. To prevent or undo this phase could retard the maturity of the firm. Having said that, LMX does fit the recent emphasis in business and industry on “flattening” the organization.

4. Assuming LMX is correct that groups tend to move toward an unfair split between the in-group and out-group, whose fault is that? We might assume it is the leader, and LMX clearly assumes it, but if each follower engages in an exchange with the leader -- if in other words the two participate in the relationship together -- does the follower in the out-group share some of the responsibility for what they negotiate or accept out of that exchange? In other words, is it possible that followers are also somewhat to blame for their predicament?

   We normally do not believe that one person is so powerful in a relationship as to determine the other person’s status wholly without some tacit consent or approval on the part of the other. To believe otherwise is to inflate the “victim” status of rational, autonomous adults. They are big boys and girls. Northouse observes, on the contrary, that LMX “does not elaborate on strategies for how [a follower] gains access to the in-group if one chooses” (Northouse, 2001, p. 120). Presumably, it is not within their control. Plus, we would ask, is it not possible that because of group dynamics, the OTHER followers (the peers of those in the out-group) help to create and enforce the split, finding ways to keep themselves distinct? The in-group is likely to want to keep itself in favor. That is, LMX seems to neglect the interaction of group members as a factor in creating sub-groups. Responsibility for disparities would have to be shared. And if that is the case, the leader has far less power to change things.

   LMX uses the leader as a symbol, a point of reference for what is really going on with everyone, like an embodiment of group dynamics. LMX concentrates on the leader’s role and responsibility. Is that concentration fair to the leader, when everyone cooperates to create the situation? That is the question.

   In systems theory, the elements in a social system can be counted and arranged conceptually in permutations to show all of the possible relations, so that each member of a group stands in some kind of relationship with every other member and with different combinations of members. Jack has a relationship with Doris that is different from his relationship with Henry, which is different from his relationship with both Doris and Henry together, which is different from the relationship Henry and Doris have with each other, and so on. This is mere quantification, however, as the sociologist Niklas Luhmann points out in his book Social Systems. The “elements” of a system (in this case, obviously, the individual group members) “acquire quality only insofar as they are viewed relationally, and thus refer to one another” (Luhmann, 1984/1995, p. 21). That is precisely what LMX is trying to do: show the quality of members based on their relationship with the leader, so that if the relationship changes, their quality vis-á-vis the firm will change as well. Only it overlooks the relationship of elements to the rest of the system. What does that mean? Luhmann can be cited for a couple of different ways of making the same claim.

   (a) The leader-member relationship is itself a system within an environment (the group) that collaborates in producing effects; elements by themselves rarely if ever “cause” effects (p. 19).

   (b) Each member of the group is in relationship with more than the leader, and to overlook that fact is to miss the point of systems thinking. Each member acquires quality based on the totality of relationships and not just on one (p. 21).

   (c) The leader-member relationship is “conditioned” by the other relationships reciprocally (p. 23).

5. Conceptually, LMX does tend to isolate the role of the leader, as though one can speak of two basic roles in any group: one person as the leader and all the rest as the followers. This schema overlooks what has become an important distinction. Leaders and leadership are different things. “Being a leader” is a role that a person plays within a group, and it is usually designated by a title such as supervisor, team leader, or boss. Leadership, by way of contrast, is a sociological phenomenon that occurs regardless of organizational structure. It is the name for something that happens among people, rather than a name for something a person is. There is (or ought to be) a correlation between the two, yet they must remain conceptually distinct, because it often happens that a person of an equal or lower rank ends up leading co-workers, even if only for a short time on a specific task.

   LMX concentrates on the role (what a person is) and assumes that the leader is the one who leads in actual practice. That is not always the case. In fact, a group often creates multiple leaders. Multiple leaders exist in hierarchies, for example, when a boss has a boss: that would be two leaders right there. Multiple leaders exist in the division of labor, as one member leads a sub-group and another member leads a different sub-group. Multiple leaders exist over time, as first one person and then another rotates in, depending on the norms of the group. For instance, the expert on topic A might lead as the group concerns itself with A, but then when it switches to topic B an expert on B steps forward.

   Finally, as LMX can actually be said to describe, multiple leaders exist if the group is being run by something on the order of an executive committee, a leadership clique, whether the organization ever officially sanctions it or not. Some organizations try to get away from designating anyone on a team as its leader. Who then in those situations does LMX describe? These are very real scenarios in today’s workplace, and they all raise doubts about the simplistic model of having only one leader and many followers.

6. Assuming the split exists (and it often does in fact), the next question we would ask is whether the split per se is what we are meant to condemn as unfair, or does it have more to do with the BASIS for the split? Not every split within groups is pernicious. Some are perfectly legitimate. A person might actually prefer a split based on seniority, let us say, in a two-tier wage structure or a split based on productivity. However, when the split is based on gender, for instance, or kinship, then we in our culture object. That strikes us as unfair. We do not oppose subdividing groups in the abstract, so long as we approve of the basis for that split. If that is the case, then LMX is misplaced, because it condemns a symptom, rather than the underlying cause. It makes a sweeping judgment of in-groups and out-groups, when in actuality sometimes the split is perfectly justifiable. LMX will not consider the possibility.

   What follows from this is that LMX prohibits subdivisions that might be ethically superior. In those cases, LMX would create the unfairness. This is what lies behind the famous quotation by Nietzsche: " 'Equality for equals, inequality for unequals' -- that would be the true voice of justice: and, what follows from it, 'Never make equal what is unequal' " (Nietzsche, 1889/1968, p. 102). Georg Simmel himself once commented, almost in passing, that “despotism is correlated with the leveling of the ruled” (Simmel, 1950, p. 374). In his mind, the egalitarian ideal serves the domination by the one leader. Seen in this light, the in-group operates as a kind of buffer of shared leadership. If everyone were made to be equal, then no one would be in a position to filter, soften, or dilute the leader’s power.

7. Only now can we raise the most basic question of all. What is fair? By what standard is it determined? Making a conscious effort to be consistent with all subordinates sounds good, and according to an egalitarian ethic, it would be fair, but what if you have employees who perform better than others? Maybe they do deserve more perks or freedom. That is the kind of dilemma LMX seems not to consider. So, what standard of fairness should apply here? There are at least five to consider:
   • full equality of outcome (regardless of any other factors)
   • fairness according to need
   • fairness according to contribution
   • fairness according to power (might makes right) or
   • fairness according to right (right makes might)

   As we said at the beginning, people do not agree on what constitutes fairness, so obviously we cannot answer the question decisively here. The purpose in asking the question is to show that LMX has a burden to defend its egalitarian standard. Other standards are just as plausible. We mentioned earlier the famous Fairness Principle of John Rawls. It reads as follows: “a person is required to do his part as defined by the rules of an institution when two conditions are met: first, the institution is just (or fair),…and second, one has voluntarily accepted the benefits of the arrangement or taken advantage of the opportunities it offers to further one’s interests” (Rawls, 1971, p. 111f). Rawls asks whether the firm where leadership takes place is itself fair. LMX takes no thought of that. Then, he asks whether each member of the group – the leader, as well as the member of the in-group and the member of the out-group – has voluntarily accepted the benefits of the arrangement or taken advantage of the opportunities it offers to further one’s interests. If so, then they cannot rightly complain if someone else does even better. It was voluntary, and it did work to the advantage of each member. The member of the out-group gets something out of the arrangement, even if only a paycheck. Why else would he or she stay in it?

   But we could take this a step further. Even if LMX does lead to an unfair split, is it possible there are offsetting benefits? Is it not possible the out-group actually does better than it would being folded into the in-group? For example, does the arrangement in some way (say, because it fosters competition within the group) encourage overall productivity and reward initiative? If so, then the out-group benefits from the overall profitability of the group. They have jobs because the in-group keeps the whole group viable. In fact, the in-group could very easily come to perceive the members of the out-group as free riders, which is also an ethical judgment.

   Additionally, being in the out-group has its rewards, since it lowers expectations and reduces scrutiny of their work. They get to operate under the leader’s radar, as it were, and preserve their anonymity. Some group members may help to place themselves in the out-group for this very reason. After all, there are costs associated with being in the in-group. What do we mean by that? If we look at a split closely, then we might need to weigh the power or status which comes with being a part of the in-group against the cost or investment it requires to belong. Sometimes the in-group must do more, either to make it into the in-group in the first place or once they belong. The in-group has to produce. It has to justify its standing. It comes under closer scrutiny.

   This consideration of the burden of being a member of the in-group ties in to another old notion: noblesse oblige. Of those to whom much is given, much will be required in return. (Luke 12:48b) In addition, the multiplicity of sub-groups within a group serves individual development (Simmel, 1955, p. 162-163). Simmel even went so far as to write the following: “In short, freedom and obligation are more evenly distributed, if the process of socialization allows for the possibility that homogenous individuals come together from heterogeneous groups rather than compel the heterogeneous components of the individual to be [accommodated to the uniform demands] of one group” (Simmel, 1955, p. 167). In other words, an egalitarian model might be unfair. That is the core of our concerns about the fairness of LMX.

RECOMMENDATIONS FOR PRACTITIONERS

Perhaps it goes without saying that as a matter of ethics, it is a leader’s responsibility to treat each subordinate in a fair and equitable manner. That is little more than saying one must do the right thing because it is the right thing to do. And let us be explicit: We simply do not agree with Machiavelli that leader effectiveness requires being unethical. All other things being equal, the leader would also be advised to allow each follower to “be all that you can be,” as the Army slogan used to put it. In that way, the follower grows and the firm benefits. Again, from a biblical standpoint from the book of Deuteronomy 2:4-6, by behaving wisely and justly toward others, you may be able to establish or restore relationships, however “fair” they might be perceived by others.

SUMMARY

Leader-Member Exchange Theory attempts to describe two types of relationship between a leader and his followers. The distinction between the two types leads to a division of groups into two sub-groups, the in-group and the out-group. Distinctions of this kind alert us to look for unfairness between the two, which this paper has attempted. Furthermore, LMX then prescribes what leaders ought to do, for the sake of effectiveness, about the distinction, and this too the paper has attempted to examine through the lenses of fairness. There is plenty of work to be done, however: LMX raises legitimate issues about the intersection of leadership and ethics.

REFERENCES

Bennis, Warren. Edu-Leadership.com (n.d.) Retrieved January 31, 2002 from http://www.edu-leadership.com/lead.html

Fukuyama, F. (1995). Trust. New York: Free Press.

Gardner, J. (1990). On Leadership. New York: Free Press.

Greenberg, J. (1996). The Quest for Justice on the Job. Thousand Oaks: Sage.

Levine, D. (1995). Visions of the Sociological Tradition. Chicago: The University of Chicago Press

Luhmann, N. (1995). Social Systems (J. Bednarz, with D. Baeker, Trans.). Stanford: Stanford University Press. (Original work published 1984)

Machiavelli, N. (1991). The Prince (R. Price, Trans.). New York: Cambridge University Press. (Original work published 1531)

Nietzsche, F. (1968). Twilight of the Idols (R.J. Hollingdale, Trans.)(pp. 19-112). New York: Penguin. (Original work published 1889)

Northouse, P. (2001). Leadership (2nd ed.). Thousand Oaks: Sage.

Pareto, V. (1991). The Rise and Fall of Elites. New Brunswick: Transaction.

Rawls, J. (1971). A Theory of Justice. Cambridge: Belknap Press.

Simmel, G. (1955). Conflict & the Web of Group-Affiliations (K. Wolff, Trans. & R. Bendix, Ed.). New York: Free Press.

Simmel, G. (1950). The Sociology of Georg Simmel (K. Wolff, Ed. & Trans.). Glencoe: Free Press.

Young, P. (1991). “Fair Division.” In P. Young (ed.). Negotiation analysis (chapter two). Ann Arbor: University of Michigan Press.