Procedural fairness and endorsement of prototypical leaders: Leader benevolence or follower control?

Abstract

This research explored why strongly identifying followers endorse prototypical leaders by addressing the role of procedural fairness in this process. We introduced the distinction between procedural fairness rules relating to leader benevolence (i.e., whether the leader supports the group’s interests) and follower control (i.e., whether followers can influence the leader’s decisions). We predicted that strongly identifying group members endorse prototypical leaders because they perceive such leaders as acting in line with benevolence related fairness rules rather than because such leaders are perceived as giving followers control. An organizational field study and a laboratory experiment revealed support for these ideas. Our results thus provide insights into why prototypical leaders are endorsed among strongly identifying followers. They also have implications for the procedural fairness literature in showing that frequently studied procedural fairness rules (e.g., voice) do not explain endorsement of leaders believed to support the group’s interests.

Introduction

An important precondition for effective leadership is that followers support their leaders and voluntarily comply with their decisions (Yukl, 2006). Research consistently shows that followers are more willing to accept decisions and to support their leaders when they perceive that these leaders enact decision making procedures fairly (see Colquitt, Greenberg, & Zapata-Phelan, 2005, for an overview). Findings like these have increased recognition of the relevance of procedural fairness for leadership research (De Cremer & Tyler, in press). The present study explores the role of procedural fairness in the social identity analysis of leadership. This analysis notes that leaders who represent important and salient group characteristics – prototypical leaders – are endorsed more, particularly among followers displaying strong group identification (see Hogg and van Knippenberg, 2003, van Knippenberg and Hogg, 2003, for overviews).

We aim to clarify why prototypical leaders are endorsed by addressing the mediating role of procedural fairness in this process. We argue that strongly identifying followers perceive prototypical leaders as acting in line with a number of procedural fairness rules. More specifically, we distinguish procedural fairness rules (cf. Leventhal, 1980) referring to follower control (i.e., whether followers can influence their leader’s decisions) from rules referring to leader benevolence (i.e., whether the leader takes care of the group’s interests). We will argue that endorsement of prototypical leaders (as it occurs among high identifiers) is based on perceptions of leader benevolence, rather than on perceptions of follower control (see Fig. 1, for a graphic depiction of our predictions). This distinction arguably has important consequences for our understanding of the social identity analysis of leadership and also for our understanding of procedural fairness.

In contrast to virtually all other approaches to studying leadership, the social identity analysis of leadership starts from the idea that leaders are members of the groups they lead, making them subject to but also able to benefit from the same social influence processes as other group members (see Hogg and van Knippenberg, 2003, van Knippenberg and Hogg, 2003; for detailed discussions and reviews of empirical evidence). This idea builds on the assumption derived from self-categorization theory (Turner, 2005, Turner et al., 1987) and social identity theory (Tajfel & Turner, 1986) that a subjectively sensed group membership, in reference to specific outgroups, makes group members view themselves and their fellow group members more in terms of a specific salient group membership and less as unique individuals (e.g., “we”, at the social psychology department view ourselves in specific terms when comparing ourselves with “them” at the business school). This cognitive representation of the group – the group prototype – describes in abstract terms what group membership involves and consequently also how group members should think, feel, and behave. Importantly, the likelihood that people find a prototypical description of themselves and their fellow group members personally relevant should be enhanced by group identification, that is, the chronic importance of a social category to the self (Haslam & Ellemers, 2005).

Because the group prototype contains normative behavioral prescriptions, group members embodying the group prototype most (prototypical group members) should be relatively strongly endorsed as legitimate group leaders (Hogg and Reid, 2001, Turner, 2005). Legitimate leaders are viewed as “appropriate, proper, and just” (Tyler, 2006, p. 376). Followers have been argued to voluntarily comply with and support such leaders even when they take controversial decisions because they are believed to support the group’s interest (Chemers, 1987, Dansereau et al., 1975, Tyler and Dawes, 1993). Research indeed shows that prototypical group members are relatively likely to emerge as group leaders (Fielding and Hogg, 1997, van Knippenberg et al., 2000). Further, prototypical leaders are viewed as more charismatic (Platow, van Knippenberg, Haslam, van Knippenberg, & Spears, 2006) and effective (Hains et al., 1997, van Knippenberg and van Knippenberg, 2005), and they are supported more than non-prototypical leaders among high identifiers (Hains et al., 1997, Hogg et al., 1998, Platow and van Knippenberg, 2001), Finally, prototypical leaders receive more leeway for controversial decisions from highly identifying group members: they are endorsed regardless of the fairness of their ingroup allocations (Platow & van Knippenberg, 2001) and the fairness with which they treat ingroup members (Ullrich, Christ, & van Dick, 2009), and even after task failure (Giessner & van Knippenberg, 2008).

Procedural fairness refers to the perceived fairness of procedures used to make allocation decisions (Tyler, 1988). The concept was introduced in the seventies by Thibaut and Walker (1975) who showed that people find procedures fairer when they can voice their opinion (i.e., process control) and when they can influence decision outcomes (i.e., decision control). Leventhal (1980) proposed a number of additional fairness rules that also enhance fairness perceptions: consistency (decisions are applied consistently over time and over people), accuracy (decisions are based on accurate information), bias suppression (decision makers are neutral and set self-interest aside), ethicality (decisions uphold standards of ethics and morality), correctability (followers can appeal to correct bad outcomes), and representation (all affected parties are heard from; note that all these expectations or aspects of fair treatment are restricted to ingroup, not outgroup, members).

A wealth of research shows that people react positively to authorities perceived as procedurally fair, such as with increased support for leaders (e.g., Huo, Smith, Tyler, & Lind, 1996) and voluntary cooperation to reach the group’s goals (e.g., De Cremer & van Vugt, 2002; see Cohen-Charash and Spector, 2001, Colquitt et al., 2001, for meta-analyses). This effect has been explained in reference to the idea that people expect fairly enacted procedures to guarantee fair outcomes in the long term (Shapiro and Brett, 2005, Thibaut and Walker, 1975). Research has also supported the role of identity processes in explaining such procedural fairness effects. In fact, there is now abundant evidence that being treated fairly by the group leader communicates that one is a valued and respected group member, and that the authority is trustworthy and neutral, which stimulates leader endorsement and voluntary cooperation at least among group members caring strongly about the group (i.e., high identifiers; Lind and Tyler, 1988, Smith et al., 1998, Tyler, 1997, Tyler and Degoey, 1995).

Although a number of studies support the idea that prototypical leaders are considered legitimate among strongly identifying group members, research to date has virtually neglected to examine the process that makes prototypical leaders legitimate. One key element contributing to leaders’ legitimacy is procedural fairness (Tyler, 1997, Tyler, 2006). Interestingly, the relationship between procedural fairness and leader prototypicality has also been speculated about by the leading theorists in the social identity analysis of leadership. They argue that “the role of leader fairness in leadership effectiveness may thus fruitfully be integrated with, and extend, the social identity analysis of leadership effectiveness’’ (van Knippenberg & Hogg, 2003, p. 281). Based on these insights, we wish to develop the argument that strongly identifying group members endorse leaders who embody the group prototype because they are perceived as acting more in line with a number of specific procedural fairness rules than non-prototypical leaders.

An important qualifying condition that we introduce, however, is that not all procedural fairness rules contribute equally to explaining endorsement of prototypical leaders. Specifically, we argue that procedural fairness rules can be classified into rules referring to leader benevolence (i.e., whether the leader acts in ways that support the group’s interest) and rules referring to follower control (i.e., whether followers can influence the leader’s decisions). Benevolent leaders do not put their own interest front center (i.e., bias suppression), do not discriminate against group members (i.e., consistency), and consider all relevant information when taking decisions that affect group members (i.e., accuracy). Follower control, on the other hand, implies that followers can influence their leader’s decisions (i.e., decision control), can voice their opinion, which arguably gives followers the idea of indirect control over the leader’s decisions (process control; see Shapiro & Brett, 2005), and, finally, that followers can correct decisions that lead to wrong outcomes (i.e., correctability). Follower control thus ensures that followers’ interests are represented in the leaders’ decisions (cf. Leventhal, 1980, Tyler, 1988).

This leader benevolence – follower control distinction builds upon the distinction between control (i.e., process control and decision control) and neutrality (including rules relating to bias suppression and accuracy) that is often used in tests of the group value model of procedural fairness (Lind & Tyler, 1988). Control and neutrality related procedural fairness rules are empirically distinct and they also influence different types of outcome variables (Tyler, 1989). It also builds upon Brockner, Ackerman, and Fairchild’s (2001) argument that process control (one aspect of follower control) differs from Leventhal’s (1980) rules of accuracy, bias suppression, and consistency because the former refers to active input from followers whereas the latter refer to leader behavior only. We build upon these works and, importantly, include all of Leventhal’s procedural fairness rules (which prior research did not do), to introduce the distinction between follower control and leader benevolence.

This distinction has important consequences for the present research because it directly addresses the essence of what endorsement of leaders as legitimate means. A wealth of studies show that followers voluntarily support and comply with legitimate leaders even when they take controversial decisions (see Tyler, 2006, for an overview). However, legitimate leadership implies specific role prescriptions, not only for followers but also for leaders (Biggart and Hamilton, 1984, Dornbusch and Scott, 1975, Emerson, 1962, Magee and Galinsky, 2008). Specifically, followers have been argued to voluntarily comply with legitimate leaders because they expect such leaders to support the group’s interests (Chemers, 1987, Dansereau et al., 1975, Tyler and Dawes, 1993). In other words, being perceived as benevolent is arguably the core antecedent of being endorsed as a legitimate leader.

Further, because legitimacy implies that leaders receive leeway to take decisions without followers wanting to be involved (Hegtvedt & Johnson, 2000), follower control over the leader’s decisions should be less impactful as a predictor of leader legitimacy. Interestingly, research shows that the procedural fairness rules that we classified as referring to leader benevolence strongly correlate with perceptions of leader legitimacy. However, fairness rules that we classified as referring to follower control (process control and decision control) correlate less strongly with leader legitimacy (Tyler, 1997; see also Colquitt et al., 2001).

This focus on legitimacy as implying leeway for leaders because they are perceived as benevolent is directly relevant for our argument: given that leader benevolence perceptions are necessary precursors of legitimacy, and given that prototypical leaders receive leeway to take decisions, suggesting that they are believed to have the groups interest at heart, leader benevolence perceptions are likely to be an integral part of leader prototypicality (among strongly identifying group members). In fact such perceptions may well explain why prototypical leaders are endorsed as legitimate group leaders.

Building on previous work (see Hogg and van Knippenberg, 2003, van Knippenberg and Hogg, 2003, for overviews), we, first of all, predict that prototypical leaders are endorsed more than non-prototypical leaders among strongly identifying group members (Hypothesis 1). The present research aims to explain why this occurs by studying the role of procedural fairness in this process. We therefore introduce the distinction between procedural fairness rules relating to leader benevolence and follower control. We predict that endorsement of prototypical leaders (among strongly identifying group members) is based on such leaders being perceived as benevolent, rather than being perceived as giving followers control in their decisions (Hypothesis 2). Fig. 1 graphically summarizes our predictions.

We test our predictions in two studies in which we distinguish leader benevolence from follower control. Study 1 is a cross-sectional field study in the context of work organizations. We use leader charisma as an indicator of leader endorsement that is positively influenced by leader prototypicality (Platow et al., 2006) as well as by procedural fairness (van Dijke, De Cremer, & Mayer, submitted for publication). Study 2 is a laboratory experiment in which we manipulate leader prototypicality and collective identification orthogonally. Here, we include support for the leader as an indicator of endorsement of prototypical leaders (e.g., Platow, Reid, & Andrew, 1998) that is also influenced by procedural fairness (e.g., Huo et al., 1996).