When timing is key: How autocratic and democratic leadership relate to follower trust in emergency contexts

Rosing, Florian; Boer, Diana; Buengeler, Claudia

doi:10.3389/fpsyg.2022.904605

ORIGINAL RESEARCH article

Front. Psychol., 03 August 2022

Sec. Organizational Psychology

Volume 13 - 2022 | https://doi.org/10.3389/fpsyg.2022.904605

When timing is key: How autocratic and democratic leadership relate to follower trust in emergency contexts

$\r\nFlorian Rosing*$ Florian Rosing^1*

Diana Boer¹

Claudia Buengeler²

¹Department of Social and Organizational Psychology, Institute of Psychology, University of Koblenz-Landau, Koblenz, Germany
²Department of Human Resource Management and Organization, Institute of Business, Kiel University, Kiel, Germany

In emergency contexts, leaders’ ability to develop others’ trust in them is critical to leadership effectiveness. By integrating functional leadership and team process theories, we argue that democratic and autocratic leadership can create trust in the leader depending on the performance phase of the action team. We further argue that action and transition phases produce different task demands for leadership behavior to enhance trust in the leader, and different leader characteristics (i.e., leader benevolence and leader ability) mediate these effects. The results of a scenario experiment (N = 125) and field survey (N = 165) among firefighters revealed that autocratic rather than democratic leadership elevates trust in the leader during the action phase by increasing leader ability. In contrast, democratic rather than autocratic leadership enhances trust in the leader during the transition phase by elevating leader benevolence. These findings highlight the importance of leader characteristics in emergencies, demonstrating the value of mixing autocratic and democratic leadership behaviors across different team performance phases to build trust in the leader.

Introduction

Although follower trust in the leader is a key factor for effective leadership across diverse situations (Dirks and Ferrin, 2002; Burke et al., 2007; Hasel and Grover, 2017), arguably the most critical context for exploring trust is in emergency settings. In this context, followers must trust their leaders and work well together because errors may lead to danger and death (Myers, 2005; Fisher et al., 2010; Kolditz, 2010). If followers do not trust their leader, they may not show strong cohesion and commitment (Weick, 1993; Hamby, 2002).

The previous literature has focused on identifying the behavioral antecedents that increase trust in leaders (Sweeney et al., 2009; Liu et al., 2010; Sweeney, 2010; Kelloway et al., 2012). These studies have implicitly assumed that such antecedents have similar effects on trust in leaders across various tasks and contexts (Lapidot et al., 2007; Colquitt et al., 2011). However, an emerging research strand has suggested that the behavioral antecedents of trust vary in emergency contexts (Lapidot et al., 2007; Sweeney, 2010). Although the current research argues that leadership is critical for mission success and survival (Hällgren et al., 2018), the role of leadership behaviors in this variability remains largely unclear.

We address these research gaps by investigating which leadership behavior functions best in emergency contexts. Many emergency teams perform their work in two recurrent performance phases: action and transition (Kozlowski et al., 1996; Marks et al., 2001; Desmond, 2006, 2008; DeChurch et al., 2011; Farh and Chen, 2018). In action phases (e.g., firefighting), the team acts toward goal accomplishment, whereas in transition phases (e.g., debriefing, reflection, and reexamination), the team focuses on mission analysis, planning, and goal setting. Several studies have suggested that both phases create different team demands, requiring various (functional) leadership behaviors (Marks et al., 2001; Morgeson et al., 2010). In the action phase, proper emergency management is key to ensuring safety, high speed, and efficiency (Yun et al., 2005; Klein et al., 2006; Maynard et al., 2017), as emergencies are unpredictable and demand quick responses (Kozlowski et al., 2009). In the transition phase, demands associated with contemplation and debriefing become salient (Karrasch et al., 2011).

Two “classical” leadership behaviors have been highlighted as significant for effective leadership: autocratic leadership, where solely the leader holds decision-making duties and power, and democratic leadership, where decision-making duties and power are shared with followers (Vroom and Yetton, 1973; Bass, 1990; Van Vugt et al., 2004). While some studies have suggested that autocratic leadership behavior is preferred at the workplace (Isenberg, 1981; Gladstein and Reilly, 1985; Mulder et al., 1986), other studies have favored democratic leadership behavior (Yun et al., 2003, 2005; Sims et al., 2009). Prior work has typically considered non-emergency contexts (Hannah et al., 2009, 2010) and yielded mixed results for the effective of these two leadership behaviors (Berkowitz, 1953; Bass, 1990; Gastil, 1994; Yukl, 2006; Schoel et al., 2011; De Hoogh et al., 2015). As action phases are more risky, urgent, and dangerous than transition phases, we assume that the impact of these two leadership behaviors on follower trust may differ. Thus, the current study investigates how such leadership behaviors are antecedent to follower trust in the leader during both phases.

In addition, to better understand the effectiveness of autocratic and democratic leadership behaviors in emergency contexts, we further examine the mechanisms that underlie the relationship between autocratic or democratic leadership and follower trust during the two phases. Based on functional leadership theory and the integrative model of trust in leadership (Burke et al., 2007), this article investigates the mediating role of leader ability—a key determinant of trust in the leader during action phases (Lapidot et al., 2007)—in the relationship between autocratic leadership and trust in the leader in action phases. Furthermore, we examine the mediating role of leader benevolence—a key determinant of trust in the leader during transition phases (Lapidot et al., 2007)—in the relationship between democratic leadership and trust in the leader in transition phases.

This study extends the literature on leadership in multiple ways. First, we integrate the literature on leadership and team performance phases (Marks et al., 2001) by investigating the impact of autocratic and democratic leadership behavior on follower trust in emergency contexts. Thus, we contribute to the ongoing debate regarding what leadership behaviors are most effective in emergencies (Hannah et al., 2009). This work also offers a nuanced description of the costs and benefits of autocratic and democratic leadership by arguing that their impact on follower trust varies according to the team performance phase. Second, in recognizing key mediators between autocratic and democratic leadership and follower trust, we offer new insights into why these leadership behaviors create trust in emergency contexts. We test our assumptions in the context of firefighting, which possesses many characteristics of organizations operating in emergency contexts (Jankovic et al., 1991; Colquitt et al., 2011; Burtscher et al., 2018). We use a multimethod approach to integrate causal relationships from a scenario experiment (Study 1) with the broader generalizability of an online survey field study (Study 2).

Background and hypotheses development

Leadership behaviors and trust in the leader for action teams

Action teams are “teams where members with specialized skills must improvise and coordinate their actions in intense, unpredictable situations” (Edmondson, 2003, p. 1421). Many action teams in emergency contexts accomplish their tasks through dual temporal-phase cycles (McGrath, 1991; Colquitt et al., 2011): action and transition phases. Morgeson et al. (2010) proposed that both team performance phases create different demands, requiring different leadership behaviors. This finding aligns with functional leadership theory (McGrath, 1962), which posits that leadership behaviors are effective when they meet certain functions critical to the team’s needs (Zaccaro et al., 2001). For example, in the action phase, leadership functions include providing resources, monitoring the team, managing team boundaries, and challenging the team (Morgeson et al., 2010). In the transition phase, leaders meet critical team needs through training and developing, providing feedback, engaging in sense-making and sense-giving, defining team missions, and establishing expectations and roles (Morgeson et al., 2010).

This paper focuses on two leadership behaviors, autocratic and democratic, widely used and recommended for risk management in emergency contexts (e.g., Vroom and Yetton, 1973; Gladstein and Reilly, 1985; Yun et al., 2005; Farh and Chen, 2018). We examine the impact of both leadership behaviors on trust in the leader (Hannah et al., 2010), defined as followers’ perception that their leaders are fair and reliable (Rousseau et al., 1998; Dirks, 2000; Schoorman et al., 2007). We suggest that the effect of autocratic and democratic leadership on follower trust varies between the two team performance phases. Specifically, we propose that autocratic leadership enhances trust in the leader during action phases. Vroom and Yetton (1973) emphasized that when the timing is critical, a leader should make decisions alone rather than delegate to team members. Hence, autocratic leaders can make quick, unilateral choices that accelerate decision-making (Eisenhardt, 1989). In particular, novel and unstructured situations require action-oriented leadership (Fiedler, 1964, 1967). Autocratic leadership consolidates administrative control and properly manages resources (Hannah et al., 2009). Thus, we argue that a leader needs to personally take charge of the operation accomplishment to build and maintain follower trust. This initiative accelerates decision-making, reduces uncertainties, and enhances members’ task knowledge (Morgeson, 2005; Kozlowski et al., 2009). We also argue that a democratic leader does not increase follower trust during action phases because these phases enhance the need for direct leader intervention to increase coordination clarity and recover shared understanding (Kozlowski et al., 2009). Therefore, we hypothesize the following:

H1a. During the action phase, follower trust in the leader is higher under autocratic than democratic leadership.

We also propose that as the team engages in transition phases, autocratic behaviors decrease in importance, and democratic behaviors become a relevant antecedent for trust in the leader. During transition phases, developmental concerns emerge as demands associated with reflection and reexamination become salient (Kozlowski et al., 2009; Karrasch et al., 2011). Vroom and Yetton (1973) suggested that in situations (1) when time is not essential, (2) the leader alone does not possess the information required to solve a problem, or (3) the interests of leaders and subordinates diverge, democratic leadership is most appropriate for achieving follower trust and consensus-building.

Existing research has demonstrated that followers tend to feel more vulnerable and more willing to scrutinize their leader’s actions and organizational processes after facing autocratic structures during emergencies (Bartunek, 1988; Hurst, 1995; Hannah et al., 2009). Leaders can use democratic behaviors to help followers voice their opinions (Farh and Chen, 2018) and develop their skills and competencies (Sims et al., 2009), which may enhance follower trust. Thus, we argue that a leader must employ democratic leadership to build follower trust during transition phases, enhancing learning opportunities while reducing conflicts among followers and leaders. Accordingly, we hypothesize the following:

H1b. During the transition phase, follower trust in the leader is higher under democratic than autocratic leadership.

The mediating role of trustworthiness in the leader’s ability and benevolence

One mechanism by which autocratic and democratic leadership may exert different effects on trust in the leader is the perception of the leader’s characteristics comprising trustworthiness (Mayer et al., 1995; Cunningham and MacGregor, 2000). Mayer et al.’s (1995) theory distinguished trust from trustworthiness, with three characteristics contributing to the prediction of trust: ability, integrity, and benevolence (Mayer and Davis, 1999; Davis et al., 2000; Gill et al., 2005; Schoorman et al., 2007; Colquitt et al., 2011). This study focuses on leader ability and benevolence, major predictors of follower trust in emergency contexts (Yukl, 2006; Lapidot et al., 2007; Hannah et al., 2009; Sweeney et al., 2009; Sweeney, 2010). We omitted leader integrity from our hypotheses because tests of functional leadership and team process models have produced more meaningful results with regard to the ability and benevolence of leaders than on their integrity (Hannah et al., 2009, 2010; Farh and Chen, 2018). Several studies have found that in situations of heightened vulnerability (i.e., firefighting operations), followers were more attentive to leader behaviors reflecting their ability (compared to leader benevolence; Freitas et al., 2002; Lapidot et al., 2007; Hannah et al., 2009). In such situations, basic security needs are aroused and followers prefer to be with a leader who is determined (Geier, 2016), and competent (Yun et al., 2005; Eberly et al., 2017), partly because in vulnerable situations, followers usually seek to establish a sense of control and mastery (Lazarus, 1966). Conversely, in situations of lowered vulnerability (i.e., debriefing), followers were more sensitive to leader behaviors reflecting benevolence (compared to leader ability; Yun et al., 2005; Lapidot et al., 2007). In such situations, relational needs are aroused and followers prefer to be with a leader who is focused on relationship development (Geier, 2016), and nurturance (Yun et al., 2005; Eberly et al., 2017). Thus, trustworthiness in the leader’s ability was more salient in action phases, and trustworthiness in the leader’s benevolence was more salient in transition phases.

The mediating role of leader ability in action phases

We assume that follower trustworthiness in the leader’s ability explains greater trust improvement associated with autocratic leadership compared to democratic leadership during the action phase. Leader ability is “that group of skills, competencies, and characteristics that enable a party to have influence with some specific domain” (Mayer et al., 1995, p. 717). Leader ability means the competence and expertise required to do a specific job with the interpersonal competencies and general wisdom required to succeed in the workplace (Gabarro, 1978; Davis et al., 2000; Burke et al., 2007; Colquitt et al., 2011; Ritzenhöfer et al., 2017).

We suggest that autocratic leadership increases follower perceptions of leader ability by drawing on functional leadership (McGrath, 1962) and team process theories (Marks et al., 2001; Kozlowski et al., 2009). Research on trust suggests that the degree to which the leader takes charge, ensuring a clear direction and an enabling structure, is the degree to which the leader will be perceived as trustworthy in terms of leader ability (Belenky et al., 1985; Hackman, 1992, 2002; Bartone, 2006; Burke et al., 2007). Research on leadership in action teams emphasizes that leaders who provide clear step-by-step directions for followers while enabling adequate resources and structures signal their ability to constructively respond to emergency contexts (Isenberg, 1981; Gladstein and Reilly, 1985; Mulder et al., 1986; Bass, 2008; Martínez-Córcoles, 2018). Therefore, if leaders behave autocratically during action phases, followers are likelier to perceive their leaders as competent than if they behave democratically. In the latter case, followers are less likely to perceive leaders as competent, seeing them as ineffectual (Yun et al., 2005) and indecisive (Gladstein and Reilly, 1985; Mulder et al., 1986; Bass, 1990). Thus, we propose the following hypothesis:

H2a: During the action phase, follower perceptions of leader ability are higher under autocratic than democratic leadership.

Trustworthiness in the leader’s ability boosts trust in the leader (Mayer et al., 1995; Schoorman et al., 2016; Shao, 2019). Research on trust shows that leader ability enables leaders to maintain their influence within a specific situation (Burke et al., 2007). Given that followers may have little experience, knowledge, and training in some situations, leader ability is a factor that promotes trust (Zand, 1972). Lapidot et al. (2007) argued that when security needs are threatened in emergencies, followers should be more attentive to the leader’s ability to manage the operation and lead the team. Consequently, we predict that autocratic rather than democratic leadership can increase trust in the leader in an action phase through leader ability. Specifically, we expect the leader’s ability to mediate the relationship between autocratic leadership and trust in the leader during action phases (see Figure 1). Hence, we propose the following hypothesis:

FIGURE 1

Figure 1. Visual depiction of the proposed research model.

H2b: In the action phase, follower perceptions of leader ability mediate the positive relationship between autocratic leadership (compared to democratic leadership) and trust in the leader.

The mediating role of leader benevolence in transition phases

We assume that follower trustworthiness in the leader’s benevolence explains the greater trust improvement associated with democratic instead of autocratic leadership during the transition phase. Benevolence is a leadership attribute that captures consideration and sensitivity to follower needs and interests (Mayer et al., 1995). Benevolence reflects that leaders are willing to care and want to do good for their followers above what is formally required, even if they do not profit from it (Ritzenhöfer et al., 2017). Thus, we propose that democratic leadership increases follower perceptions of leader benevolence based on functional leadership and team process theories. Research on trust has suggested that the degree to which the leader engages in shared decision-making and provides opportunities to voice opinions increases the perception of leader trustworthiness in terms of benevolence (Hackman, 2002; Burke et al., 2007). Moreover, research on leadership in action teams has suggested that leaders who encourage a high degree of group members’ involvement and participation during decision-making signal that they care about follower welfare and interests during transition phases (Konovsky and Pugh, 1994; Burke et al., 2007; Caldwell and Hayes, 2007). Therefore, if leaders behave democratically in this phase, followers are likelier to perceive the leader as benevolent than if leaders behave autocratically. In the latter case, followers are less likely to perceive their leader as benevolent because they are seen as eroding their sense of control over group decisions (De Cremer, 2007). Hence, we predict the following hypothesis:

H3a: During the transition phase, follower perceptions of leader benevolence are higher under democratic than autocratic leadership.

Trustworthiness in the leader’s benevolence boosts trust in the leader (Mayer et al., 1995; Schoorman et al., 2016; Shao, 2019). Mayer et al. (1995) explained that benevolence influences trust because benevolent leaders build a positive attachment to followers and a low motivation to lie. These qualities play a prominent role in building and maintaining trust. Research on action teams shows that in less threatening but stressful situations, such as transition phases, benevolent leaders trigger followers’ needs for nurture and care. In these situations, followers should be more attentive to the leaders’ attributes that reflect their benevolence (Lapidot et al., 2007). Hence, we predict that democratic rather than autocratic leadership can increase follower trust during transition phases by enhancing perceptions of leader benevolence. Specifically, we expect leader benevolence to mediate the relationship between democratic leadership and trust in the leader. We propose the following hypothesis:

H3b: In the transition phase, follower perceptions of leader benevolence mediate the positive relationship between democratic leadership (compared to autocratic leadership) and trust in the leader.

Overview of the current work

Following recommendations for confirming our results’ robustness and generalizability (Chen et al., 2021), we used a multimethod approach to examine our hypotheses. We conducted two studies among professional firefighters. In Study 1, we conducted a scenario experiment to test our hypotheses in a controlled research context (Aguinis and Bradley, 2014). We then conducted a cross-sectional field study to enhance external validity. In Study 2, we replicated and extended the findings from Study 1 using cross-sectional data from a sample of 165 firefighters to test our hypotheses in an emergency context.

Study 1

Method

Participants

In this study, 125 firefighters participated, of which 89% were male, ages 16 to 62 years (M = 32.35, SD = 9.99), with an average of 16 years of work experience. Firefighters were recruited via professional firefighting conferences, department visits, email news briefs, blogs, and forums. All firefighters volunteered to complete the study.

Design and experimental procedure

This study used a 2 × 2 between-subject design, with the leader’s behavior (autocratic/democratic) as the first factor and the temporal phase (action/transition phase) as the second factor. Participants were randomly assigned to one of four vignettes [autocratic/action phase (n = 31), democratic/action phase (n = 32), autocratic/transition phase (n = 30), democratic/transition phase (n = 32)] to control for order effects.

The online experiment included a screening page, informed consent, instructions, the vignette (to manipulate the type of leadership behavior and temporal phase), and a questionnaire (Geier, 2016). The questionnaire included the dependent measures and controls. We informed participants that they were participating in a study that “examines critical success factors of effective teamwork” and that they were going to read a scenario and would respond to questions about that scenario. Specifically, they were asked to imagine that they were experiencing the written scenario and assume the role of a fire brigade member.

Vignette development

We began vignette development by constructing a set of draft scenarios based on nine in-depth interviews in which firefighters were asked about realistic examples of leadership behaviors during the action and transition phases (Ashill and Yavas, 2006; Levy, 2006; Aguinis and Bradley, 2014; Rosing et al., 2022; see Appendix 1). The most frequently mentioned examples of autocratic and democratic leadership within action and transition phases were selected to create the vignettes (Leicher and Mulder, 2016). The vignettes were reviewed by three researchers (who were not part of the research team) and two firefighters until consensus was achieved on their true-to-life correctness (Hughes and Huby, 2002; McAlearney, 2006; see Table 1).

TABLE 1

Table 1. Vignettes used in Study 1.

Measures

Unless otherwise stated, the measures in this study used a 7-point Likert scale (1 = strongly disagree to 7 = strongly agree).

Autocratic leadership behavior

We measured the effectiveness of manipulating autocratic leadership with six items from the autocratic leader behavior scale (De Hoogh et al., 2004). The Cronbach’s alpha score was 0.92. Examples included items such as “the leader makes decisions in an autocratic way” and “the leader often pushes his/her opinions.”

Democratic leadership behavior

We measured the effectiveness of manipulating democratic leadership using five items from the participative decision-making questionnaire (Arnold et al., 2000). Sample items included “the leader encourages group members to express ideas/suggestions” and “the leader gives all group members a chance to voice their opinions.” The Cronbach’s alpha score was 0.70.

Leader ability and benevolence

Mayer and Davis (1999) developed a measure of trustworthiness that included subscales of leader ability and benevolence. We used six items to measure leader ability (“the leader is very capable of performing his job,” α = 0.96) and five items to study leader benevolence (“the leader is very concerned about followers’ welfare,” α = 0.97).

Trust in the leader

We used a three-item scale developed by Giessner and van Knippenberg (2008) to assess overall trust in the leader. The items were as follows: “I trust the leader absolutely,” “I think this leader does the right things,” and “I think this leader is trustworthy” (α = 0.95).

Control variables

We controlled for firefighters’ age (continuously in years), gender (1 = male, 2 = female), and job tenure (continuously in years). The results remained unchanged with or without these controls in the model.

Data analysis

We analyzed our data in several steps. First, the variance inflation factor (VIF) was calculated for each variable to determine multicollinearity (Suen, 1990). Second, a one-way analysis of variance (ANOVA) tested the effects of leadership behaviors on follower trust (H1a-b) and perceptions of the leader’s abilities and benevolence (H2a and H3a). We also used planned comparison (Rosenthal and Rosnow, 1985) tests to compare the means between the conditions. Third, to analyze H2b and H3b, we used the PROCESS tool in SPSS (Hayes, 2012). We chose model 4 of this tool with 5,000 bootstraps and, as recommended (Hayes and Preacher, 2014), a confidence interval of 95% for estimating the respective effects.

Results

Multicollinearity

We assessed the absence of multicollinearity with VIF scores of less than 10 (VIF < 10 = no serious multicollinearity; Cohen et al., 2003; Neubert and Taggar, 2004). The results of our multicollinearity analysis showed that the VIF scores ranged within acceptable values from 1.09 to 2.87 (VIF = 2.45 for leader ability, 2.52 for leader benevolence, 2.82 for age, 1.09 for gender, and 2.87 for job tenure), indicating an extremely low level of multicollinearity in our study (Dingel and Wei, 2014).

Manipulation checks

Table 2 presents descriptive statistics and intercorrelations among all study variables. Table 3 shows the means and standard deviations for all experimental conditions. We analyzed the manipulation checks in two steps. First, we analyzed whether our leadership manipulation in the action phase was effective. The results of a univariate ANOVA indicated that participants assigned the autocratic leadership condition perceived the leader as being more autocratic than participants assigned the democratic condition [M = 5.10, SD = 1.18 vs. M = 2.39, SD = 1.52, with t(121) = 8.95, p < 0.01]. In contrast, participants assigned the democratic leadership condition perceived the leader as more democratic than participants assigned the autocratic condition [M = 2.76, SD = 0.90 vs. M = 4.97, SD = 0.94, with t(121) = −10.59, p < 0.01].

TABLE 2

Table 2. Descriptive statistics and intercorrelations among Study 1 variables.

TABLE 3

Table 3. Mean values, standard deviations, and significances of differences between experimental conditions (Study 1).

Second, we used a univariate ANOVA to test whether our manipulation of leadership in the transition phase was successful. The results again showed that participants assigned the autocratic leadership condition perceived the leader as being more autocratic than participants assigned the democratic condition [M = 6.30, SD = 0.54 vs. M = 3.56, SD = 1.32, with t(121) = 8.95, p < 0.01]. In contrast, participants assigned the democratic leadership condition perceived the leader as being more democratic than participants assigned the autocratic condition [M = 5.35, SD = 0.55 vs. M = 2.27, SD = 0.87, with t(121) = −14.66, p < 0.01]. In conclusion, all manipulations worked as expected.

Hypotheses testing

We supported H1a since performing autocratic leadership behavior during the action phase resulted in higher trust ratings (M = 4.22, SD = 1.67) than performing democratic leadership (M = 3.22, SD = 1.90; p < 0.05). We also supported H1b since performing democratic behavior during the transition phase resulted in higher trust in the leader (M = 5.58, SD = 1.14) than performing autocratic leadership (M = 2.41, SD = 1.31; p < 0.01; see Figure 2). Moreover, we supported H2a since autocratic leadership resulted in higher leader ability ratings (M = 4.21, SD = 1.41) during the action phase than democratic leadership (M = 3.29, SD = 1.76; p < 0.05, see Figure 3) and also supported H2b since the indirect effect of autocratic leadership (as compared with democratic leadership) on follower trust during the action phase through leader ability was significant (effect = –0.90, 95% CI[–1.63, –0.13]).

FIGURE 2

Figure 2. The relationship between autocratic and democratic leadership and trust in the leader for the action and transition phase (Study 1).

FIGURE 3

Figure 3. The relationship between autocratic and democratic leadership and leader ability for the action phase (Study 1).

Finally, we supported H3a since leader benevolence ratings were higher for democratic leadership behavior (M = 5.84, SD = 0.98) during the transition phase than for autocratic leadership behavior (M = 1.66, SD = 1.01; p < 0.01, see Figure 4). Furthermore, in the transition phase, the indirect effect of democratic (rather than autocratic) leadership behavior on follower trust through leader benevolence was significant (effect = 3.95, 95% CI[2.84, 5.57]), supporting H3b.

FIGURE 4

Figure 4. The relationship between autocratic and democratic leadership and leader benevolence for the transition phase (Study 1).

Discussion

Using a scenario-based experimental methodology, we designed Study 1 to provide an experimental test of the impact that autocratic and democratic leadership has on follower trust. Study 1 supported that the influence of autocratic and democratic leadership on follower trust varies depending on the team performance phase since the action and transition phases involve different task demands and thus leadership requirements. We found that autocratic rather than democratic leadership elevates follower trust by increasing leader ability in the action phase. In contrast, democratic rather than autocratic leadership enhances follower trust during transition phases by elevating leader benevolence. Our scenario-based experimental methodology had the advantage of isolating the effects of autocratic and democratic leadership (Antonakis et al., 2010). However, the drawback was that respondents could not directly experience the team performance phase to which they were responding (Belschak and Den Hartog, 2009). Therefore, we conducted a second study in the context of firefighting to replicate and extend the results from Study 1 using a different method involving a sample of firefighters who actually experience leader behaviors in extreme operations.