Human Resource Management Review 12 (2002) 25 42

www.HRmanagementreview.com

Understanding pay plan acceptance: The role of distributive justice theory

Alison E. Barbera,*, Marcia J. Simmeringb,1
a

Department of Management, Eli Broad School of Business, Michigan State University, East Lansing, MI 48824-1122, USA b Rucks Department of Management, E. J. Ourso College of Business Administration, Louisiana State University, 3171 B CEBA, Baton Rouge, LA 70803-6312, USA

Abstract As U.S. businesses shift from individual rewards toward more aggregated pay systems, they must address the issue of reward allocation within groups. Specifically, should aggregate rewards be allocated equally to all group members, or should individual contributions be recognized? In this paper, the multiprinciple view of distributive justice is used as a starting point to predict employee reactions to different allocation methods. Propositions for future research that could facilitate the implementation of alternative pay plans are offered. D 2001 Elsevier Science Inc. All rights reserved.
Keywords: Compensation; Distributive justice; Pay plan acceptance

1. Introduction The decade of the 1990s has been one of change in American compensation practices. Restructuring the nature of work from an individual, job-based focus to more flexible teambased management systems (Bridges, 1994) has led many companies to look for compensation systems more compatible with a team-oriented environment (Lawler, 1992; Seaman, 1997; Sisco, 1992), and outcome-oriented, group-level systems such as profit sharing and gainsharing have grown in popularity (Gomez-Mejia & Balkin, 1992; Hansen, 1998; Lawler, 1990).
* Corresponding author. Tel.: +1-517-432-3522; fax: +1-517-432-1111. E-mail address: aebarber@.msu.edu (A.E. Barber). 1 Tel.: + 1-225-578-5187; fax: + 1-225-578-6140. E-mail address: msimmel@lsu.edu (M.J. Simmering). 1053-4822/02/$ see front matter D 2001 Elsevier Science Inc. All rights reserved. PII: S 1 0 5 3 - 4 8 2 2 ( 0 1 ) 0 0 0 3 9 - 0

26

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

Additionally, many companies are changing their compensation plans through broadbanding, which involves collapsing many salary grades into a few wide bands (Abosch & Hand, 1994). Broadbanding is used to flatten organizational structures, so pay is less tied to hierarchical level, which often leads to difficulty in determining rewards (LeBlanc & Ellis, 1995). One type of compensation that fits the needs of these changing organizations is aggregate pay plans. These plans, which are . . . any arrangement for a group of employees to receive a variable award based on increased performance against a target (DeMatteo, Eby, & Sundstrom, 1998), can apply to small teams or entire organizations. Some of the most popular types of aggregated pay plans are profit sharing, gainsharing, and team-based performance bonuses. Profit sharing is a group- or organization-based pay plan in which payments to employees are based on the profit of the group or organization (Gerhart & Milkovich, 1992). Gainsharing is also a group- or organization-based pay plan, but payments are based on productivity measures, rather than profits, and tend to be distributed more often than profit-sharing bonuses (Gerhart & Milkovich, 1992). Finally, team-based performance bonuses are based only on team performance, and payments may be distributed evenly among members or can vary based on performance levels of team members. A challenge associated with all of these aggregate plans involves the allocation of grouplevel reward to individual group members. For instance, if the bonus for a team of five members is US$1000, how should that money be allocated to individuals on that team? Should it be divided equally so that each member received US$200? Should the US$1000 be allocated based on the existing salaries of the employees, with those earning more receiving a larger bonus? Alternately, should the reward be distributed based on evaluations of individual performance within the group? Because there is no one right answer to this question, effective allocation of aggregate pay is a difficult decision. Many practitioners have lamented the difficulty of determining allocation of rewards to individuals in groups (e.g., Hayes, 1997; LeBlanc & Mulvey, 1998; Masternak, 1997; Weinberger, 1998). In this paper, we use an existing theoretical framework of distributive justice to address the issue of allocation of group rewards. We begin by discussing the importance of employee reactions, specifically fairness, to aggregate pay plans. We then address several contingency factors that may influence the effectiveness of different pay plan decisions. After addressing the contingency factors individually, we examine several combinations of these factors and their effects on pay plan acceptance. Finally, we identify priorities for future compensation research in this area.

2. Why employee reactions matter Employee reactions to changes in compensation plans are important, as they can influence the success of such innovations. Johns (1993) argued that many human resource innovations fail because their advocates do not pay sufficient attention to the context in which the innovations will be adopted. As Kossek (1989) argued, employee reactions are an important part of that context, and employee acceptance of human resource innovations is a necessary (though not sufficient) condition for their effective implementation. Nevertheless, although stories of innovative pay plans abandoned because of employee resistance are not hard to find

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

27

(e.g., Brown & Huber, 1992; Gandossy & Gubman, 1988; Petty, Singleton, & Connell, 1992), we know little about the circumstances under which new pay plans survive or fail (Gerhart, Trevor, & Graham, 1996). Gerhart et al. (1996) identified pay program survival as an important new direction for compensation research. As they noted, the early demise of new pay plans can lead to both tangible costs (e.g., staff time, consultants fees) and intangible costs (e.g., employee morale, credibility of human resource management staff). In addition, there may be opportunity costs associated with the termination of plans that could have been effective if implementation had been handled differently. The acceptance of aggregated pay plans is of particular importance. There is evidence that aggregated pay plans are unpopular with employees when compared with individual level plans (Cable & Judge, 1994; LeBlanc & Mulvey, 1998). Furthermore, negative attitudes regarding pay plans can lead to detrimental employee actions. In particular, employees may leave organizations in which they have negative reactions to compensation plans (Pfeffer & Davis-Blake, 1992). Therefore, the acceptance of group pay allocation decisions is critical to the success of these plans.

3. Role of justice in group pay allocation Gerhart et al. (1996) argued that pay plan survival may be largely dependent on employee perceptions of pay plan fairness. Drawing from the procedural justice literature, they proposed that process issues such as communication and input or voice might be significant predictors of employee acceptance of new pay plans. Their arguments are consistent with the notion that the drive to achieve justice is a central element of the human character (Lerner, 1977): policies or procedures that violate ones sense of justice are likely to be resisted. There are two main types of justice that are detailed in the literature: distributive justice and procedural justice. Distributive justice refers to the perceived justice of allocation outcomes in the context of compensation, this is the perceived justice of the distribution of monetary rewards (Greenberg, 1987). The fairness of the procedure or process by which outcomes are determined is captured in procedural justice. There has been a great deal of attention to procedural justice issues recently (e.g., Cole & Latham, 1997; Skarlicki, Ellard, & Kelln, 1998; Taylor, Masterson, Renard, & Tracy, 1998); however, the literature on distributive justice has been neglected. Therefore, we focus on distributive justice as a starting point for understanding pay plan acceptance. This paper extends the argument that perceptions of justice may be critical to pay plan survival by focusing on how examination of new pay programs from the perspective of distributive justice might enrich our understanding of their acceptance and survival. Reward allocation can be based on one of two principles: equity or equality. An equitable distribution of rewards, in which employees receive an outcome based on some individual contribution, is generally determined by individual performance (Milkovich & Newman, 1999). In contrast, egalitarian rewards are independent of individual performance measures and are intended to foster group cohesion and team work (Milkovich & Newman, 1999). Traditional U.S. pay systems have generally allocated rewards at the individual level and are based on an equity principle. Therefore, those individuals who make the greatest contribution

28

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

receive the greatest individual reward. For example, when merit pay raises are given to employees, they are based on the performance of each individual. Those employees who perform better receive greater rewards. Equitable pay systems are expected to maximize individual performance. In contrast, aggregated pay plans by design de-emphasize individual contributions, and focus instead on the attainment of group outcomes (Milkovich & Newman, 1999). One intent of aggregate pay is to reduce individual competition and promote cooperation. Therefore, although it is possible to recognize individual contributions to group outcomes within these group-based plans, both scholarly and practitioner writings suggest that attempts to create such hybrid pay systems may have negative results (e.g., Masternak, 1997; Wageman, 1995). However, pay plans that do not allow for differentiation among individuals in effect shift allocation norms away from individual equity and toward within-group egalitarianism. Egalitarian pay may take many forms. In the strictest sense, egalitarianism may mean the dollar amount of rewards is identical for each group member. However, egalitarian may also apply to rewards that are based on a set percentage of salary (e.g., everyone gets 5%). In this scenario, even though the actual dollar amount paid varies based on different salary levels, people are treated equally in the sense that, by our metric, the decision rule is applied equally to everyone, and does not vary on the basis of individual contributions. This suggests that some forms of egalitarian pay are more equal than others. The current research does not address the differences in what is labeled egalitarian, or the degree to which group members perceive the rewards to be equal. So, for the purposes of this paper, we consider egalitarian rewards to be a broad definition. Given that aggregated pay plans may change the rules under which compensation is allocated, employee perceptions of the distributive justice of these alternative compensation programs may be critical determinants of their acceptance and ultimately of their survival. Distributive justice theory provides a rich theoretical framework for addressing the question of when more egalitarian allocation rules are likely to be accepted. According to multiprinciple or contingency theories of distributive justice (e.g., Cullen, 1992; Deutsch, 1985; Lerner, 1977; Leventhal, 1976; Markovsky, 1991; Sampson, 1986) justice norms are socially derived (and therefore variable) rather than natural or inherent (and therefore universal). While in some situations equity norms prevail, in other situations egalitarian norms prevail. A substantial body of literature, largely ignored in compensation research, examines the circumstances under which various norms are more or less likely to be viewed as just. These circumstances encompass two primary issues: individual or situational contingencies that influence acceptance of different allocation norms, and the impact of custom or experience on preferred allocation norms. The purpose of this paper is to introduce these issues to compensation scholars and to illustrate how this existing knowledge might be used to guide research on employee reactions to new pay plans.

4. Contingency factors in distributive justice preferences The distributive justice literature examines a wide variety of contingency factors influencing individual beliefs about the fairness of alternative distributive justice norms such as

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

29

equity or equality. Unfortunately, this stream of research can be characterized as a proliferation of a long list of more or less unrelated factors (Tornblom, 1992). There have been no efforts to create parsimonious categories of factors or to identify important relationships among factors. Additionally, the majority of the research related to the fairness of group-level pay has been done in the laboratory under artificial conditions (DeMatteo et al., 1998). The nature of the research often precludes field studies and quasi-experiments; fairness issues are often delicate subjects in organizations and are unlikely to be manipulated. However, the findings in this literature provide a foundation for conclusions regarding contingency factors. In order to provide some structure to the discussion, we focus on factors that seem to have particular implications for compensation and organize them into three categories: betweengroup differences, the nature of the reward being allocated, and the nature of the group to which allocation is made. We address these categories individually, and then discuss possible interactions among them. Following a discussion of existing distributive justice research on specific factors, we identify priorities for future compensation research. 4.1. Group differences Our focus in this paper is group acceptance of aggregated pay plans, and therefore, we address differences among people by discussing their membership in groups. Certain sets of people, because of their experiences, form values that are likely to be consistent within that group. Although not all people within groups will be identical, research has supported the notion that group membership is related to acceptance of certain forms of compensation. Group differences in distributive justice beliefs and behaviors present significant challenges for compensation professionals, in that they suggest the inherent difficulty of designing pay programs that will be viewed as just by all affected parties. There is substantial rationale for and evidence of such differences in perceptions of justice. We discuss three sets of factors, which are grouped together because of similarities in their underlying logic: gender and nationality, hierarchical position and performance, and unionization. 4.1.1. Gender and nationality/national heritage A large proportion of the research on alternative allocation norms examines differences on the basis of gender or nationality/national origin. Numerous extensive reviews of this material have already been published (e.g., Bierhoff, Buck, & Klein, 1986; Dornstein, 1991; James, 1993; Major & Deaux, 1982; Mikula, 1980). However, ties to compensation practice have rarely been made. With regard to gender, many authors have concluded that women are more inclined toward egalitarian distributions, and men are more inclined toward equitable distributions (Bierhoff et al., 1986; Dornstein, 1991; James, 1993; Mikula, 1980). One common explanation for this finding is that women attach more importance to interpersonal relationships than do men, leading women to prefer equality, a distribution norm believed to lead to greater group harmony (Dornstein, 1991; Major & Deaux, 1982). Closer examination reveals a number of contingency factors that moderate this finding, including sex typing of task (Bierhoff et al., 1986), demographic composition of the group (Major & Deaux, 1982), and interpersonal orientation of the allocators (Major & Adams, 1983). In fact, this finding is most consistently

30

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

obtained where subjects are allocating rewards to same-sex coworkers (in reality, typically fellow laboratory subjects) whose performance is inferior to that of the allocator (Dornstein, 1991). In such a scenario, women are more generous (and therefore in this context more egalitarian) than are men, allocating a larger share of the reward to their partner and retaining a smaller share for themselves. This has led to a third explanation for gender differences in allocation: that women devalue their own work and have lower expectations of reward, leading to lower pay allocations to themselves (Dornstein, 1991; Jackson & Grabski, 1988). The practical value of these findings is in understanding that there is more than gender affecting pay plan acceptance. The literature suggests that it is factors such as socialization and group composition that relate to differences between males and females. Beyond this, there is little that can be done in practice with these results. Recommendations to base pay plans on gender flies in the face of U.S. employment laws and the emphasis on increased workforce diversity. However, there should be increased research attention to gender issues in pay plan acceptance; in particular, more field research in this area is warranted. Research should be aimed at understanding why males and females differ in their acceptance of pay plans. Gender differences in any aspect of compensation are interesting and important, given general concerns about effective diversity management and more specific concerns about persistent malefemale wage differentials. However, the practical value of the findings reviewed above is somewhat constrained. Results suggest that egalitarian norms are most likely to prevail when groups are homogeneously female. But recommendations to implement egalitarian pay innovations in homogeneous female groups would fly in the face of both U.S. employment law and actual workforce demographic patterns (i.e., increasing workforce diversity). However, understanding more about the differences in pay plan acceptance between males and females may lead to more effective methods of implementing aggregated pay plans. In contrast, global competition and the proliferation of multinational organizations suggest that there may be substantial practical value in studying nationality-based differences in justice norms. The logic behind nationality differences in justice beliefs is generally based on a functionalist perspective; norms that protect the basic social unit are held in high esteem (Kashima, Siegal, Tanaka, & Isake, 1988). Since primary social units differ across national cultures, so too should distributive norm preferences. In particular, with respect to distributive justice, it is often anticipated that individuals from individualistic cultures (which emphasize the individual as the primary social entity) will favor equity-based allocations, and individuals from collectivist cultures (which emphasize the group as the basic social unit) will favor equality-based allocations. More specifically, some have argued that the equity orientation is a primarily Western phenomenon (Major & Deaux, 1982). There is some evidence to support this position. For example, Bond, Leung, and Wan (1982) and Leung and Bond (1984) found greater preference for equity among Westerner cultures that were primarily individualistic (e.g., United States) as compared to Asians that were primarily collectivistic (e.g., China). But, as was the case with gender, further study has revealed a number of limitations on this finding. First, it seems to hold only with respect to distributions to members of ones own group. Collectivists, while tending to make egalitarian distributions to members of their own collective, have also been found to make strongly

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

31

equity-based allocations to individuals who are not members of their in-group (James, 1993; Leung & Bond, 1984). Second, the differences that are found can often be characterized as differences in degree (James, 1993; Kashima et al., 1988). For example, Kashima et al. (1988) found that Japanese allocators were only marginally more equality oriented than Australians. Furthermore, contrary to the generally anticipated pattern, Chen (1995), in a role-playing study, found that Chinese business people were consistent in supporting differential allocation of rewards, whereas Americans were more likely to prefer equality in certain situations. While these results are less than conclusive, they are intriguing, and improved understanding of these issues as applied to compensation plans would be of great value to compensation practitioners working in international organizations. It is essential that compensation practitioners understand the norms and values of different societies, and that they adopt practices that are not overly disruptive to or incompatible with prevailing philosophies (Levine, 1998; Miles & Greenberg, 1993). Group-based pay plans may well be met with greater acceptance in collectivist cultures, particularly if workgroups are relatively homogeneous. 4.1.2. Hierarchical level and performance A second set of group factors with strong implications for distributive justice includes hierarchical level and performance level. The importance of hierarchical level as a moderator of reactions in organizational contexts is widely recognized (Fagenson, 1984). With respect to distributive justice beliefs, several studies have demonstrated that equity-based allocations receive greater support at higher levels of the organization and that equality receives greater support at lower levels. Lansberg (1984), in a study of employees of a single business organization, found managers to be equity oriented, while clerical employees were indifferent between equity and equality norms. Dornstein (1985, described in Dornstein 1991), in openended field interviews, found employees with higher incomes more likely to respond to the question, what is a fair wage? by describing contributions, suggesting a stronger orientation toward equity among this group. Two explanations for this pattern of preferences have been offered. First, norms could be shaped by self-interest (Cook & Hegtvedt, 1983; Lansberg, 1984). Because they are already in higher paying jobs, managerial level employees may simply be justifying their current position based on contributions that differ by hierarchical level, such as education required or amount of responsibility involved in the work. For example, some high-earning United States CEOs justify large rewards by defending their unique contribution to the organization. Such equity-based justifications permit upper level employees to maintain their positions without threatening their belief in a just world (James, 1993). Conversely, lower level and therefore lower paid employees, who would gain financially from greater interclass equality, may prefer an equality norm. A second explanation involves attributions: higher level employees, who typically experience greater autonomy on the job, may feel more in control of work-related outcomes. This sense of control may validate the concept of basing distribution on contributions. Therefore, upper-level employees who are empowered to make strategic decisions are more likely to feel that they have made a contribution than employees who do not have authority to make large-scale decisions. On the other hand, those who feel little control over work-related

32

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

outcomes may be less likely to embrace any kind of contingent rewards and may view an egalitarian system as more fair. From a compensation standpoint, either rationale implies that group pay plans that are egalitarian across hierarchical levels are more likely to be viewed as just by lower level employees than by upper level employees. In addition, the second rationale implies that group pay plans that are egalitarian within hierarchical levels also are more likely to be viewed as just by lower level employees. To our knowledge, however, these hypotheses have not been tested. Arguments regarding possible relationships between performance levels and allocation norm preferences parallel those offered with respect to hierarchical level. Self-interest would suggest that better performers prefer equity-based allocations (under which they would do relatively well) to equal allocations (under which they would do relatively poorly), and that the opposite would be true for poor performers (Deutsch, 1991; Messick & Sentis, 1983). For instance, employees who shirk may prefer egalitarian rewards because they receive rewards that are greater than their contributions. Arguments based on attributions reach the same results, using different logic: lower performers may perceive critical performance factors as out of their control, and therefore feel that an equality norm is more fair, while better performers may attribute their outcomes to their own talents and efforts and may therefore believe an equitable allocation is fair. Studies of performance level provide some evidence that pay plan preferences vary as a function of contribution. Deutsch and colleagues, in a series of laboratory experiments reported in Deutsch (1991), found that lower performers preferred equal or needs-based distributions, while better performers preferred equity or winner takes all distributions. Similarly, Austin (1980) found that people tend to prefer equity when their own performance is high, and equality when their performance is low. These findings suggest that egalitarian group-based plans may have detrimental effects on employee retention. If high performers perceive aggregate pay allocations to be unjust, they may be motivated to search for alternative employment, ultimately resulting in dysfunctional turnover. As yet, this linkage has not been documented, but it merits investigation. One study that provides a good starting point for further investigation is a large-scale survey conducted by Pfeffer and Davis-Blake (1992). The authors found that high earners in organizations with large salary dispersions were less likely to turnover as compared with low earners. This may indicate that high those individuals at the top of the salary distribution may be less accepting of egalitarian pay (small salary dispersion). 4.1.3. Unionization One of the major goals of unions is to secure what are believed to be fair pay for their members (Fossum, 1995). Research on unions suggests that they may be more in favor of egalitarian pay rather than equitable pay (Milkovich & Newman, 1999). The primary reason for this is the traditional U.S. unions focus on egalitarianism in general. The structure of U.S. unions is aimed at egalitarianism, and each dues-paying member expects the same amount of representation from the union. In general, unions have historically bargained for across-theboard pay increases of equal magnitude for all bargaining unit jobs (Fossum, 1995). This goal is likely to be similar when aggregated pay plans are being negotiated. Although there is little

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

33

research in this area, we can infer that unionized employees will be more in favor of egalitarian reward distributions than of equitable ones. 4.2. Nature of reward A number of arguments suggest that individuals prefer different allocation rules for different types of rewards. While empirical work on this topic is limited, conceptual efforts have identified the following dimensions of rewards as possible influences on preferred allocation norms: how the reward is generated, and the magnitude and availability of the reward. 4.2.1. Method of generation The primary issue relevant to reward generation is how much control potential recipients have over the amount of reward to be allocated. Rewards can be differentiated into those that are fortuitously acquired, (i.e., the reward is acquired independent of the actions of the recipient and is essentially a fixed sum in the eyes of the potential recipients), and those that are collectively created or dependent, (i.e., the amount of reward is a function of the behavior of the recipients and therefore is seen by them as variable) (Bell & Schokkaert, 1992). For example, if employees have stock options that increase in value due to general trends in the stock market, this would create a reward in which there is little instrumentality, that is, a weak link between performance and reward. Conversely, in cases where the total amount of reward available appears to depend on performance of group members, equity is likely to be the preferred allocation mode. For example, in a small team, members may be very aware of the varying contributions of individual members to the team outcome, and it may be difficult to ignore the role of individuals in creating the pool of reward. However, where the total amount of reward available is independent of group efforts, equal division is generally seen as more appropriate. In such a case, no single individual has any fundamental claim to a higher proportion of resources, and thus there is no compelling logic for an equity-based distribution (Bell & Schokkaert, 1992; Cook & Yamagishi, 1983; Messick & Sentis, 1983). These issues may be related to what compensation experts refer to as line of sight (Lawler, 1990). Line of sight refers to the degree to which employees see a connection between their own actions and behaviors and the outcomes that are to be rewarded. In general, line of sight weakens as the level of aggregation of the compensation system increases; it is strongest when the rewards are based on the individuals own personal outcomes and weakest when rewards are based on the performance of the entire organization. To the extent that weak line of sight corresponds to beliefs that rewards are not directly related to effort or performance, egalitarian distributions will be considered more just when reward systems function at higher levels of aggregation. For example, in many cases, profit sharing has a weak line of sight because of the wide range of factors beyond individual employee contributions (e.g., economic factors) that lead to increased profits. This suggests that egalitarian distributions based on profit sharing plans may be considered more just than egalitarian distributions based on division or team performance. Again, however, we know of no research directly addressing this issue.

34

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

4.2.2. Magnitude of reward The influence of the amount of reward available on distributive justice preferences has been addressed in diverse ways by various scholars, and drawing a unified conclusion is difficult (Dornstein, 1991). However, a number of authors have argued that equality will be preferred when amount to be allocated is small (Austin & Hatfield, 1980; Tornblom, 1992), and that equity is preferred when the amount to be allocated is large (Lane & Messe, 1972). Austin and Hatfield (1980) suggested that reward magnitude may have consequences for the feasibility of different distribution techniques. Allocations based on equality are simple: they require no information about recipients other than how many of them there are, and they involve only very simple calculations. They involve little time, little negotiation, and little justification. In contrast, specifying an equity distribution requires one to lay out fairly complicated rules regarding how inputs will be assessed and valued before an actual allocation can be made. This can be time consuming and difficult, and may not be worthwhile if rewards are small. As a result, Austin and Hatfield argued that equality will be the preferred rule when the amount to be distributed is small, essentially as the result of a costbenefit analysis. There is little empirical evidence relevant to this question, and none at all coming from actual compensation contexts (DeMatteo et al., 1998). This is unfortunate, as reward magnitude may be particularly salient for compensation plans in these times of intense global competition and low inflation. Bonus and salary increase pools have been fairly low as a percentage of payroll in recent years, particularly as compared to times of high inflation, causing some to question whether the effort that goes into individual performance-based distribution of such small sums of money is worthwhile (Lawler, 1990; Schuster & Zingheim, 1992). With smaller rewards, a shift to egalitarian allocation may be more acceptable to both managers and employees. The distributive justice literature does not provide insight into the important applied question of what is a small amount of reward, and what is a large amount. However, the compensation literature has developed the notion of the just-noticeable difference in pay. This amount varies across individuals depending on factors such as pay expectations, recent pay raises, and consumption needs (Krefting & Mahoney, 1977). This might suggest that egalitarian distributions are preferred for rewards that fall under the just noticeable difference threshold, and that equity-based distributions are preferred for rewards that exceed that threshold. This speculation merits further investigation. 4.3. Nature of group In this section, three characteristics of groups that bear on distributive justice norm preferences will be discussed: task structure and group interdependence, social relationships within the group, and group size. 4.3.1. Task structure The degree of interdependence that exists within a work group is a key determinant of the feasibility, and fairness, of equity as an allocation procedure. As argued above, the equality norm is universally easy to apply. In teams that have pooled interdepen-

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

35

dence, in which group members have equivalent roles and little interaction, equitable rewards are most likely to be acceptable and effective (DeMatteo et al., 1998; Montemayor, 1994). Application of the equity rule, however, requires additional information. It requires knowledge of the relative contributions of group members, as well as knowledge of the relationships between inputs and outcomes (Messick & Cook, 1983). Therefore, in such scenarios it is likely that egalitarian distributions would be preferred (Bierhoff et al., 1986). Applying the equity rule becomes more difficult as interdependence of team members increases. In sequential teams, members perform tasks in a prescribed order, and performance of the group is dependent on the weakest member. Therefore, equitable pay may be more acceptable, as individual results are less easy to identify and as increased cooperation in important (Montemayor, 1994). In reciprocal teams, where interdependence is the highest, egalitarian pay is most appropriate because of the need to reward collaborative behavior (Montemayor, 1994). In situations where individual contributions can easily and accurately be identified, equitybased allocations may be preferred. In fact, there is substantial experimental evidence to suggest that the lack of task interdependencies is associated with a preference for equitybased allocations (see Dornstein, 1991; Greenberg & Cohen, 1982 for reviews). This suggests that when task interdependence is high, egalitarian compensation systems will be perceived as just, and that when task interdependence is low, equity-oriented systems will be perceived as just. 4.3.2. Social relationships One of the most consistent findings in the allocation literature is that an equality norm is often preferred where the group is cohesive, relationships are close, and the group is perceived as a team (Bierhoff et al., 1986; Dornstein, 1991). For example, Bagarozzi (1982) found that experimental subjects assigned to a cohesive condition shared rewards equally following a 50-minute exercise, while subjects in a noncohesive condition did not. Similarly, Chiu (1990), in a study of students working on a class project, found that assigning equal grades to all group members was seen as more just by students working in cohesive groups. Merely labeling a work group as a team can lead to stronger preferences for equality based distributions, as discovered in experiments by Lerner (1974) and Schwinger (1980). This preference appears to hold whether actual inputs within the group are equal or not. The rationale typically offered for this finding is that it is a function of both affective tone (liking for other person) as well as shared identity and perceived similarity (Bagarozzi, 1982; Greenberg & Cohen, 1982). Both affective and similarity effects have been supported in empirical research (Dornstein, 1991). As a result, one would expect that members of highly cohesive groups will be more likely to view egalitarian distribution of rewards as just than will members of less cohesive groups. However, this conclusion is based on primarily laboratory research, as there are few field studies of this relationship. Despite our lack of knowledge about the effects of social relationships and pay plan acceptance in actual organizations, we suspect that the laboratory findings for the preference of egalitarian rewards is likely to be supported in other settings.

36

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

4.3.3. Group size The importance of group size as a potential influence on allocation norm preferences has been noted (Cohen & Greenberg, 1982), but to date little empirical research into this issue has been conducted. To some extent, existing research has been limited by the practical realities of laboratory research, which have often restricted group size to two or three members. Two countervailing logical arguments have been made regarding group size and allocation preferences. First, feasibility arguments suggests that egalitarian distributions may be preferred when groups are large. For instance, in the case of profit sharing, the group is the organization, which may be several thousand people. The information processing required to account for individual contributions in large groups may be daunting, leading to a preference for the simpler egalitarian division (Cohen, 1991). Alternatively, large groups are in general expected to be less cohesive, suggesting that egalitarian distributions will be less acceptable in large groups (Grzelak, 1991). Again, in a case of profit sharing, it is unlikely that an entire organization will be very cohesive. One way to assimilate these perspectives is to suggest that, all else equal, egalitarian distributions are more likely to be considered just when allocations are made to large groups in other words, that group cohesion must be controlled before the predicted relationship with size can emerge. 4.4. Contingency factors and compensation research Table 1 summarizes the findings reviewed above. These findings permit us to make predictions about conditions under which employees will perceive group-based pay plans to be just, and therefore may improve our understanding of when such pay plans are most likely to survive. However, these predictions are speculative, in that they are based on laboratory research where actors are brought together in small groups (often dyads or triads) for brief periods of time (Cohen, 1991), small (almost trivial) rewards are allocated (Deutsch, 1991), and contingency factors are studied one at a time (Tornblom, 1992). Research of this sort has been essential in establishing the basic principle that distributive justice norms are variable rather than universal. But there is certainly room to question whether or how well these findings generalize to complex field settings involving large ongoing groups, significant portions of compensation, and simultaneous operation of a variety of contingency factors. Research examining these issues in organizational settings is needed to increase the confidence with which we make predictions about compensation plan acceptance. Changes in the nature of groups and rewards studied may be inherent in the transition from laboratory to field. Perhaps the more difficult research design question involves identifying appropriate factors to be studied. Although existing literature focuses on individual contingency factors as separate topics, interrelationships among factors are not hard to identify and, indeed, are somewhat hard to ignore. We have addressed many contingency factors individually, but future research should begin to look at how these factors may interact to influence pay plan acceptance. For example, relationships between individual and group demography have already been noted, as have relationships between group size and group cohesion. Relationships that involve factors from more than one category are also possible. For instance, if employees at lower organization levels typically receive lower rewards than higher level employees,

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542 Table 1 Circumstances under which egalitarian pay plans are more likely to be accepted I. Egalitarian pay plans are more likely to be accepted by: A. females (in gender homogeneous groups) B. collectivists (in culturally homogeneous groups) C. employees at lower levels of the organization D. poorer performers II. Egalitarian pay plans are more likely to be accepted when: A. pay plans level of aggregation is high B. amount of pay to be allocated is low C. group task interdependence is high D. group is cohesive E. group is large (provided that cohesion held constant)

37

hierarchical level and amount of reward might be mutually reinforcing influences on justice perceptions. Another issue linking factors from more than one category is the relationship between hierarchical level and gender. In the U.S., women have recently moved into jobs that are higher in the organizational hierarchy, but men still occupy the majority of these upper echelon positions (Milkovich & Newman, 1999). Additionally, a number of influences perpetuate the wage gap between the genders, with women earning less on average than men (Milkovich & Newman, 1999). In organizations in which many females are in lower paying jobs that are also lower in the organizational hierarchy, this may account for the preference for egalitarian pay. The problem in these investigations, then, becomes one of identifying subsets of factors to examine jointly. One approach to this problem would be to acknowledge the central role of group cohesion in influencing preferences for equitable versus egalitarian distributions. With this approach, one might identify factors that influence justice norm preferences through their relationship to cohesiveness, either as antecedents or as consequences, thus building a contingency model based entirely on intragroup relations. A second approach would be to classify factors based on the rationale underlying their relationship to norm preferences. For instance, at least three distinct rationales emerge from the material discussed above: social issues related to group cohesion, self-interest, and feasibility. One could either attempt to investigate a comprehensive set of factors tapping into a single rationale, or to investigate individual factors representing distinct rationales. Indeed, research examining when specific rationales are likely to dominate would add value both to our knowledge of compensation as well as to distributive justice theory.

5. Familiarity, justice, and pay plan acceptance Compensation innovation experts have argued that one way to increase the likely success of a new pay plan is to implement it initially in a portion of the organization where it is most likely to be accepted (Gandossy & Gubman, 1988; Gross & Bacher, 1993). By helping to identify situations where more egalitarian distributions are likely to be perceived as just,

38

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

distributive justice theory can help organizations pick early winners (Gandossy & Gubman, 1988) for initial implementation. But distributive justice theory also provides a justification for the importance of early successes: Over time, individuals who initially perceived a particular allocation as unjust may come to view that same allocation as just (Cohen & Greenberg, 1982; Deutsch, 1986). For example, Shepelak and Alwin (1986), using data from a large-scale survey, concluded that tradition (i.e., beliefs about the way things are) was related to perceptions of fairness. Such effects are likely driven by the need to believe in a just world: allocations that are common must be just, if the world is just. Another concern related to acceptance of a new pay plan is the impact of group cohesion. Here, two competing arguments can be made with respect to the appropriate timing of a conversion to more egalitarian pay. First, it should be noted that work group cohesiveness is a malleable characteristic that can be altered by management practices. To the extent that management is already using effective team-oriented practices, and has successfully increased group cohesiveness, acceptance of aggregated or egalitarian systems should be higher. This logic suggests that changes in management structure and style should precede changes in compensation. Alternatively, pay changes can be used to lead rather than lag changes in management systems (Lawler, 1990), consistent with the view of compensation policies as statements of organizational culture (Kerr & Slocum, 1987), and suggesting an order of implementation that is the reverse of that suggested above. At this point, there is not sufficient evidence to resolve these competing perspectives. But while the proper ordering cannot be specified, one can argue that team management techniques and team-based pay are mutually reinforcing, and that egalitarian pay systems will be seen as more just when they occur in conjunction with the implementation of basic team management techniques. The relationship between familiarity and perceived justice has several implications for compensation practice. First, new plans that are initially met with resistance may gain acceptance if they are not abandoned too quickly. At this time, however, we do not know how long plans need to remain in effect to take advantage of familiaritys impact on justice perceptions. Second, successful pilot programs may increase the perceived justice of new allocation arrangements not only in the eyes of participants but also among others who are aware of the pilot. In other words, new pay plans may become familiar either through direct experience or through vicarious exposure. While it seems likely that either form of familiarity might increase justice perceptions, this issue remains to be tested. Here again, application of distributive justice theory to compensation scenarios permits an opportunity to enhance both compensation practice and distributive justice theory. Because employees typically participate in organizations for longer periods of time than subjects participate in laboratory studies, field settings should provide better opportunities to examine dynamic relationships between reward allocations and justice perceptions, and therefore could contribute greatly to our understanding of changes in justice beliefs over time.

6. Conclusion This paper provides an introduction to the contingency view of distributive justice and discusses how that literature can be relevant to the study of pay plan survival. In applying this

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

39

material to compensation contexts, we may be able to improve upon existing distributive justice research by including multiple contingency factors and incorporating time as a factor. This contribution would supplement the very practical advantages associated with better understanding of the circumstances under which certain pay plans are more likely to succeed. An acknowledged limitation of the present paper is that it focuses on one particular aspect of group-based pay plans: their capacity to allocate rewards in an egalitarian fashion. Clearly, these plans have other features to which employees might react, including but not limited to the processes by which they are implemented. A complete understanding of reactions to innovative pay plans would require investigation of both distributive and procedural justice issues, and would also require attention to considerations unrelated to justice. Recent work in the area of procedural justice (e.g., Brockner & Weisenfeld, 1996) offers a rich framework for future research in this area. However, the need for research beyond distributive justice does not diminish the need for further investigations into this issue.

References
Abosch, K. S., & Hand, J. S. (1994). Characteristics and practices of organizations with broadbanding: a study of alternative approaches. ACA Journal, 3, 6 17. Austin, W. (1980). Friendship and fairness: effects of type of relationship and task performance on choice of distribution rules. Personality and Social Psychology Bulletin, 6, 402 408. Austin, W., & Hatfield, E. (1980). Equity theory, power and social justice. In: G. Mikula (Ed.), Justice in social interaction ( pp. 25 62). New York: Springer-Verlag. Bagarozzi, D. A. (1982). The effects of cohesiveness on distributive justice. Journal of Psychology, 110, 267 273. Bell, J., & Schokkaert, E. (1992). Interdisciplinary theory and research on justice. In: K. Scherer (Ed.), Justice: interdisciplinary perspectives. Cambridge, UK: Cambridge Univ. Press. Bierhoff, H. W., Buck, E., & Klein, R. (1986). Social context and perceived justice. In: H. W. Bierhoff, R. L. Cohen, & J. Greenberg (Eds.), Justice in social relations ( pp. 165 186). New York: Plenum. Bond, M. H., Leung, K., & Wan, K. C. (1982). How does cultural collectivism operate? The impact of task and maintenance contributions on reward distribution. Journal of Cross-cultural Psychology, 13, 186 200. Bridges, W. (1994). The end of the job. Fortune, 130, 62 74 (September 19). Brakner, J., & Weisenfeld, B. M. (1996). An integrative framework for explaining reactions to decisions: interactive effects of outcomes and procedures. Psychological Bulletin, 120, 189 202. Brown, K. A., & Huber, V. L. (1992). Lowering floors and raising ceilings: a longitudinal assessment of the affects of an earnings-at-risk plan on pay satisfaction. Personnel Psychology, 45, 279 311. Cable, D. M., & Judge, T. A. (1994). Pay preference and job search decisions: a person organization fit perspective. Personnel Psychology, 47, 317 348. Chen, C. C. (1995). New trends in rewards allocation preferences: a Sino U.S. comparison. Academy of Management Journal, 38, 408 428. Chiu, C. (1990). Distributive justice among Hong Kong Chinese college students. Journal of Social Psychology, 130, 649 656. Cohen, R. L. (1991). Membership, intergroup relations, and justice. In: R. Vermunt, & H. Steensma (Eds.), Social justice in human relations Societal and psychological origins of justice (vol. 1, pp. 239 258). New York: Plenum. Cohen, R. L., & Greenberg, J. (1982). The justice concept in social psychology. In: J. Greenberg, & R. L. Cohen (Eds.), Equity and justice in social behavior (pp. 2 41). New York: Academic Press. Cole, N. D., & Latham, G. P. (1997). Effects of training in procedural justice on perceptions of disciplinary fairness by unionized employees and disciplinary subject matter experts. Journal of Applied Psychology, 82, 699 705.

40

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

Cook, K. S., & Hegtvedt, K. A. (1983). Distributive justice, equity, and equality. Annual Review of Sociology, 9, 217 241. Cook, K. S., & Yamagishi, T. (1983). Social determinants of equity judgments: the problem of multidimensional input. In: D. M. Messick, & K. S. Cook (Eds.), Equity theory: psychological and sociological perspectives ( pp. 95 126). New York: Praeger. Cullen, B. (1992). Philosophical theories of justice. In: K. Scherer (Ed.), Justice: interdisciplinary perspectives ( pp. 15 64). Cambridge, UK: Cambridge Univ. Press. DeMatteo, J. S., Eby, L. T., & Sundstrom, E. (1998). Team-based rewards: current empirical evidence and directions for future research. Research in Organizational Behavior, 20, 141 183. Deutsch, M. (1985). Distributive justice. New Haven, CT: Yale University Press. Deutsch, M. (1986). Cooperation, conflict, and justice. In: H. W. Bierhoff, R. L. Cohen, & J. Greenberg (Eds.), Justice in social relations ( pp. 3 18). New York: Plenum. Deutsch, M. (1991). Egalitarianism in the laboratory and at work. In: R. Vermunt, & H. Steensma (Eds.), Social justice in human relationsSocietal and psychological origins of justice (vol. 1, pp. 195 209). New York: Plenum. Dornstein, M. (1991). Conceptions of fair pay. New York: Praeger. Fagenson, E. A. (1984). Womens work orientations: something old, something new. Group and Organization Studies, 11, 75 100. Fossum, J. A. (1995). Labor relations: development, structure, process. Chicago, IL: Irwin. Gandossy, R. P., & Gubman, E. L. (1988). Change pay for performance without resistance. Personnel Journal, 67, 127 131. Gerhart, B., & Milkovich, G. T. (1992). Employee compensation: research and practice. In: M. D. Dunnetter, & L. M. Hough (Eds.), Handbook of industrial organizational psychology (vol. 3, pp. 481 570). Palo Alto, CA: Consulting Psychologists Press. Gerhart, B., Trevor, C. O., & Graham, M. E. (1996). New directions in compensation research: synergies, risk, and survival. In: K. Rowland, & G. Ferris (Eds.), Research in personnel and human resource management (vol. 14, pp. 143 203). Greenwich, CT: JAI Press. Gomez-Mejia, L. R., & Balkin, D. B. (1992). Compensation, organizational strategy, and firm performance. Cincinnati, OH: South-Western Publishing. Greenberg, J. (1987). A taxonomy of organizational justice theories. Academy of Management Review, 12, 9 22. Greenberg, J., & Cohen, R. L. (1982). Why justice? Normative and instrumental interpretations. In: J. Greenberg, & R. L. Cohen (Eds.), Equity and justice in social behavior ( pp. 437 469). New York: Academic Press. Gross, S. E., & Bacher, J. P. (1993). The new variable pay plans: how some succeed, why some dont. Compensation and Benefits Review, 25, 51 56. Grzelak, J. L. (1991). The model of homo economicus: should it be revisited? In: R. Vermunt, & H. Steensma (Eds.), Social justice in human relationsSocietal and psychological origins of justice (vol. 1, pp. 211 238). New York: Plenum. Hansen, F. (1998). Incentive plans are now commonplace in large firms. Compensation and Benefits Review, 30, 8. Hayes, C. (1997). Whats yours is mine. Black Enterprise, 27, 75. Jackson, L. A., & Grabski, S. V. (1988). Perceptions of fair pay and the gender wage gap. Journal of Applied Social Psychology, 18, 606 625. James, K. (1993). The social context of organizational justice: cultural, intergroup, and structural effects on justice behaviors and perceptions. In: R. Cropanzano (Ed.), Justice in the workplace: approaching fairness in human resource management ( pp. 21 50). Hillsdale, NJ: Erlbaum. Johns, G. (1993). Constraints on the adoption of psychology-based personnel practices: lessons from organizational innovation. Personnel Psychology, 46, 569 592. Kashima, Y., Siegal, M., Tanaka, K., & Isaka, H. (1988). Universalism in lay conceptions of distributive justice: a cross-cultural examination. International Journal of Psychology, 23, 51 64. Kerr, J., & Slocum, J. W. (1987). Managing corporate culture through reward systems. Academy of Management Executive, 1, 99 108. Kossek, E. E. (1989). The acceptance of human resource innovation by multiple constituencies. Personnel Psychology, 42, 263 281.

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

41

Krefting, L. A., & Mahoney, T. A. (1977). Determining the size of a meaningful pay increase. Industrial Relations, 16, 83 93. Lane, I. M., & Messe, L. A. (1972). Distribution of insufficient, sufficient and oversufficient rewards: a clarification of equity theory. Journal of Personality and Social Psychology, 21, 228 233. Lansberg, I. (1984). Hierarchy as a mediator of fairness: a contingency approach to distributive justice in organizations. Journal of Applied Social Psychology, 14, 124 135. Lawler, E. E. (1990). Strategic pay: aligning organizational strategies and pay systems. San Francisco: JosseyBass Publishers. Lawler, E. E. (1992). Designing pay systems for teams. Los Angeles, CA: University of Southern California. Center for Effective Organizations (CEO) Publication T 92-12. LeBlanc, P. V., & Ellis, C. M. (1995). The many faces of banding. ACA Journal, 4, 52 62. LeBlanc, P. V., & Mulvey, P. W. (1998). How American workers see the rewards of work. Compensation and Benefits Review, 30, 24 28. Lerner, M. J. (1974). The justice motive: equity and parity among children. Journal of Personality and Social Psychology, 29, 539 550. Lerner, M. J. (1977). The justice motive: some hypotheses as to its origins and forms. Journal of Personality, 45, 1 52. Leung, K., & Bond, M. (1984). The impact of cultural collectivism on reward allocation. Journal of Personality and Social Psychology, 47, 793 804. Leventhal, G. S. (1976). The distribution of rewards and resources in groups and organizations. In: L. Berkowitz, & E. Walster (Eds.), Advances in experimental social psychology (vol. 9, pp. 91 131). New York: Academic Press. Levine, H. Z. (1998). How pay should reflect culture. Compensation and Benefits Review, 30, 76 77. Major, B., & Adams, J. B. (1983). Role of gender, interpersonal orientation, and self presentation in distributive justice behavior. Journal of Personality and Social Psychology, 45, 598 608. Major, B., & Deaux, K. (1982). Individual differences in justice behavior. In: J. Greenberg, & R. L. Cohen (Eds.), Equity and justice in social behavior ( pp. 43 76). New York: Academic Press. Markovsky, B. (1991). Prospects for cognitive structural justice theory. In: R. Vermunt, & H. Steensma (Eds.), Social justice in human relations, vol. 1. New York: Plenum. Masternak, R. (1997). How to make gainsharing successful: the collective experience of 17 facilities. Compensation and Benefits Review, 5, 43 52. Messick, D. M., & Cook, K. S. (1983). Equity theory: psychological and sociological perspectives. New York: Praeger. Messick, D. M., & Sentis, K. (1983). Fairness, preferences, and fairness biases. In: D. M. Messick, & K. S. Cook (Eds.), Equity theory: psychological and sociological perspectives ( pp. 61 94). New York: Praeger. Mikula, G. (1980). Justice in social interaction. New York: Springer. Miles, J. A., & Greenberg, J. (1993). Cross-national differences in preferences for distributive justice norms: the challenge of establishing fair resource allocations in the European community. In: J. B. Shaw, P. S. Kirkbridge, K. M. Rowland, & G. R. Ferris (Eds.), Research in personnel and human resource management (Suppl. 3, pp. 133 156). Greenwich, CT: JAI Press. Milkovich, G. T., & Newman, J. M. (1999). Compensation (6th ed.). Boston, MA: Irwin McGraw-Hill. Montemayor, E. F. (1994). A model for aligning teamwork and pay. ACA Journal, 3, 2 9. Petty, M. M., Singleton, B., & Connell, D. W. (1992). An experimental evaluation of an organizational incentive plan in the electric utility industry. Journal of Applied Psychology, 77, 427 436. Pfeffer, J., & Davis-Blake, A. (1992). Salary dispersion, location in the salary distribution, and turnover among college administrators. Industrial and Labor Relations Review, 45, 753 763. Sampson, E. E. (1986). Justice ideology and social legitimation: a revised agenda for psychological inquiry. In: H. W. Bierhoff, R. L. Cohen, & J. Greenberg (Eds.), Justice in social relations ( pp. 87 102). New York: Plenum. Schuster, J. R., & Zingheim, P. K. (1992). The new pay: linking employee and organizational performance. New York: Lexington Books. Schwinger, T. (1980). Just allocations of goods: decisions among three principles. In: G. Mikula (Ed.), Justice in social interaction ( pp. 95 126). New York: Springer-Verlag.

42

A.E. Barber, M.J. Simmering / Human Resource Management Review 12 (2002) 2542

Seaman, R. (1997). Case study: rejuvenating an organization with team pay. Compensation and Benefits Review, 29, 25 30. Shepelak, N. J., & Alwin, D. F. (1986). Beliefs about inequality and perceptions of distributive justice. American Sociological Review, 51, 30 46. Sisco, R. (1992). Put your money where your teams are. Training, 29, 41 45 (July). Skarlicki, D. P., Ellard, J. H., & Kelln, B. R. C. (1998). Third-party reactions of a layoff: procedural, derogation, and retributive aspects of justice. Journal of Applied Psychology, 83, 119 127. Taylor, M. S., Masterson, S. S., Renard, M. K., & Tracy, K. B. (1998). Managers reactions to procedurally just performance management systems. Academy of Management Journal, 41, 568 579. Tornblom, K. (1992). The social psychology of distributive justice. In: K. Scherer (Ed.), Justice: interdisciplinary perspectives ( pp. 177 236). Cambridge, UK: Cambridge Univ. Press. Wageman, R. (1995). Interdependence and group effectiveness. Administrative Science Quarterly, 40, 145 180. Weinberger, T. E. (1998). A method for determining the equitable allocation of team-based pay: rewarding members of a cross-functional account team. Compensation and Benefits Management, 14, 18 26.