Abstract
Love (in the agape form) forms the foundation of most leadership concepts and has been ignored in research. We respond to the debate on universal applicability of leadership forms by bringing followers into the spotlight through our examination of the interactive influence of loving (agape-based) and non-loving (non-agape-based) leadership styles and followers’ attachment dimensions (self-model and other-model) on follower outcomes. Two hundred and eighty-two business management students worked in teams on a task under the direction of leaders who demonstrated agape-based behaviours and leaders who demonstrated non-agape-based behaviours in a laboratory experiment. Agape-based leadership was positively related with follower satisfaction with the leader, team commitment and perception of leaders’ effectiveness. Further, followers’ attachment dimensions (self- and other-model) moderated the relationship between agape-based leadership and follower work attitudes, such that the relationship was positive for followers with a negative self-model and for followers with a positive other-model, and the relationship was negative for followers with a negative other-model. We provide a practical set of tools for demonstrating agape leadership behaviours which are useful for educators and organizations. We suggest that leaders must alter their leadership style depending on their followers’ attachment dimensions.
Leadership theories have been criticized for lacking an explicit moral dimension (Hoch et al., 2016). Although love (in agape form) has been described as the foundation of many leadership concepts such as servant leadership (Patterson, 2003), it has remained elusive. Leading with love (agape) or the term ‘agape-based leadership’ has been conceptualized as an intentional response to promote well-being when confronted by that which generates ill-being through humility, forgiveness, self-sacrifice, impartiality, emotional control and empathy (Carvalho & Mulla, 2021).
There have been numerous examples of agape-based leadership in theology (Jesus Christ who forgave sinners who condemned Him), history (Mahatma Gandhi who through the practice of non-violence peacefully endured abuse, insults and repeated prison terms), politics (Abraham Lincoln who chose Stanton for the post of Secretary of War in spite of Stanton’s deep hatred for and humiliation of Lincoln) and organizational management (Isaac Tigrett of Hard Rock Café who recruited social outcasts such as street people and bikers). Surprisingly, there has been very little scholarly attention devoted to agape in leadership studies due to its variance and ambiguity (Oord, 2005) . Moreover, the positive psychology movement has brought increasing attention to character strengths and virtues such as love, forgiveness and humility (Peterson & Seligman, 2004). We respond to the need for including love within corporate and scholarly leadership development research (Van Dierendonck & Patterson, 2014) by examining if loving (agape-based) leadership has a positive effect on follower outcomes in a laboratory conflict resolution experiment.
Additionally, leadership research has focused almost exclusively on the unidirectional impact of leaders’ behaviours on followers’ attitudes (Yukl, 1989). What remains to be seen is whether all followers are equally susceptible to its influence. Since the leadership relationship is jointly produced by leaders and followers, it becomes crucial to focus on followers (Howell & Shamir, 2005). According to Meindl (1995, p. 331), ‘Leadership is very much in the eyes of the beholder: followers, not the leader—and not the researchers—define it’. Klein and House (1995) suggest that leadership is a social process requiring a spark (leader), flammable material (susceptible followers) and oxygen (conducive context). Yukl (1999) asserts that insecure, alienated and fearful followers may be more susceptible to certain leadership styles. Leadership effectiveness will be biased if researchers fail to incorporate the influence of follower profiles such as their attachment styles (Shalit et al., 2010). We respond to the debate on ‘universal applicability’ (Yukl, 1999, p. 301) of leadership styles by examining if follower attachment dimensions such as their self-model associated with the degree of anxiety and dependency on others and their other-model associated with the degree of avoidance in relationships influence the effectiveness of agape-based leadership on follower outcomes.
Agape in Leadership
Agape has been conceptualized as purposeful love independent of emotions, based on good motives, done to promote the good of the other when confronted with hostility, abuse and injury (Haggai, 2009). Agape means ‘love of benevolence’ (Melé, 2012) that involves a willingness to sacrifice and to forgive (King, 2010). It is distinct from other forms of love like ‘eros (romantic, passionate love), ludus (game-playing love), storge (friendship love), mania (possessive, dependent love), pragma (logical, “shopping list” love)’ (Hendrick & Hendrick, 1986, p. 392). Agape-based leadership can be differentiated from existing related independent constructs of humility, self-sacrifice and existing leadership theories such as servant, ethical, authentic and spiritual leadership due to its unique features, namely that it is deliberate and independent of emotions, requires good motives, that is, congruent with deontological philosophy, focuses especially on individuals who are undeserving of love and includes only those acts which are performed in response to ill-being (Carvalho & Mulla, 2021). Agape-based leadership and servant leadership, though closely related to each other, are different in terms of their ethical approaches. Servant leadership is congruent with the moral consequentialist theory which evaluates the morality of an act by its outcomes (Lemoine et al., 2019). Agape-based leadership, on the other hand, emphasizes good motives and is congruent with the deontological philosophy of treating individuals as ends in themselves rather than as means to an end. It also involves acting from duty which lies in the will, independent of inclination. While servant leadership may involve all employees and acts that promote their well-being, agape-based leadership focuses especially on undeserving followers and acts that promote well-being in response to ill-being.
Model of Agape-Based Leadership
We developed a model of agape-based leadership drawing from the prior conceptualization of agape in the work context which is intentional behaviours (independent of emotion) exhibited by leaders to promote the well-being of individuals when confronted with ill-being through six manifestations, namely self-sacrifice, empathy, humility, emotional control, forgiveness and impartiality (Carvalho & Mulla, 2021).
Agape-based leaders showed openness to contrasting views and opinions of others, and accepted critical feedback without getting defensive (humility); let go of anger and resentment over their team members’ mistakes (forgiveness); showed sensitivity to their team members’ personal needs and feelings (empathy); managed disruptive emotions when confronted with arguments and stress (emotional control); treated every person with equal regard irrespective of their worthy attributes (impartiality); and made others’ needs a priority before their own (self-sacrifice).
Model of Non-agape-based Leadership
Drawing from models of pseudo-transformational leadership (Barling et al., 2008; Bass & Steidlmeier, 1999; Christie et al., 2011), we developed a model of non-agape-based leadership based on behaviours that contrasted with the six manifestations of agape described above.
Non-agape-based leaders dissuaded contrasting views and opinions of others, and got defensive when challenged (arrogance); harboured resentment over their team members’ mistakes (retribution); focused more on results rather than relationships (detachment); showed anger during arguments and stress (impatience); treated individuals differently depending on their worthy attributes (favouritism); and made their own needs a priority (self-interestedness).
Impact of Leadership Style on Follower Outcomes
Research has shown positive relationships between agape-based leadership behaviours such as humility, forgiveness, empathy and emotional intelligence and follower outcomes such as employee task satisfaction, psychological empowerment, trust and loyalty in the leader, perception of leader effectiveness, task satisfaction, Organizational Citizenship Behaviour and risk-taking (Caldwell & Dixon, 2010; Conger et al., 2000; De Cremer & Van Knippenberg, 2004; Jeung & Yoon, 2016; Kock et al., 2019; Owens et al., 2013; Van Knippenberg & Van Knippenberg, 2005; Wong & Law, 2002). Social exchange theory suggests that the quality of social interactions induces unspoken obligations to return favours to those who have acted in one’s interest (Blau, 1964). Social modelling theory emphasizes role modelling by the leader as an important process through which social influence occurs in the workplace (Bandura, 1971). Therefore, we hypothesize:
Hypothesis 1: Agape-based leadership will have a positive effect on follower outcomes such as (a) followers’ satisfaction with the leader, (b) team commitment, (c) task satisfaction and (d) perception of leaders’ effectiveness.
Interacting Effect of Leadership Style and Followers’ Characteristics
Leadership research has been criticized for being too leader-centric (Meindl, 1995). Lord et al. (1999, p. 167) state that ‘the follower remains an unexplored source of variance in understanding leadership processes’. The perception and preference for leadership styles are influenced by follower characteristics such as their cultural orientation (Jung & Avolio, 1999); values and needs (Ehrhart & Klein, 2001); need for leadership (De Vries et al., 2002); self-concepts (Howell & Shamir, 2005); personality (Felfe & Schyns, 2006; Schyns & Sanders, 2007); core self-evaluations (Nübold et al., 2013); need for autonomy and growth need strength (Wofford et al., 2016); and personality and work values profiles (Thoroughgood & Sawyer, 2018). Leaders who are motivating to one follower may not be motivating to other followers (Ehrhart & Klein, 2001; Shalit et al., 2010). People-oriented leaders may be more effective among followers motivated by an empathetic approach as compared to followers characterized by an avoidant attachment style (Shalit et al., 2010). Therefore, we examined the influence of followers’ characteristics on leadership effectiveness.
Attachment theory ‘is a way of conceptualizing the propensity of human beings to make strong affectional bonds to particular others’ (Bowlby, 1977, p. 201) and therefore becomes highly relevant for predicting followers’ susceptibility and reactions to certain leadership styles. Research in attachment theory has moved away from its focus on infants and their relationships with caregivers to its implications for romantic relationships (e.g., Hazan & Shaver, 1987), work relationships (Hazan & Shaver, 1990) and more recently leadership relationships (e.g., Hinojosa et al., 2014; Popper et al., 2000; Shalit et al., 2010). As with parent–child relationships, leader–follower relationships involve ‘guiding, directing, taking charge, and taking care of less powerful others whose fate is highly dependent on them’ (Popper & Mayseless, 2003, p. 42). While a lot of research has been done on the attachment styles of leaders, there is very little known about how followers’ attachment styles influence the leader–follower relationship (Popper et al., 2000). Followers may form emotional attachment relationships with agape-based leaders who fulfil their needs and aid their development.
Interacting Effect of Leadership Style and Followers’ Attachment Dimensions
Attachment theory has implications for the leadership relationship, as it ‘provides us with a conceptual framework for classifying followers on their potential and tendencies to form different emotional relationships with leaders’ (Shalit et al., 2010, p. 461). Influenced by their early relationships with their caregiver, individuals have three styles of attachment, namely secure, insecure–ambivalent, and insecure–avoidant, which guide their future interactions with others (Ainsworth et al., 1978).
Bartholomew and Horowitz (1991) distinguished between two avoidant attachment styles—fearful attachment and dismissing attachment—and expanded the categorization into four categories, namely secure, insecure–preoccupied/ambivalent, insecure–dismissing and insecure–fearful. These categories form a matrix based on two internal working models (Griffin & Bartholomew, 1994)—a self-model which is ‘the degree to which individuals have internalized a sense of their own self-worth and therefore expect others to respond to them positively’ (p. 431) and an other-model which is ‘the degree to which others are generally expected to be available and supportive’ (p. 431). The self-model is associated with the degree of anxiety and dependency, and the other-model is associated with the degree of avoidance in relationships (Bartholomew & Horowitz, 1991; Griffin & Bartholomew, 1994).
Secure individuals have a positive self-model and positive other-model (low anxiety and low avoidance), dismissing individuals have a positive self-model and negative other-model (low anxiety and high avoidance), fearful individuals have a negative self-model and negative other-model (high anxiety and high avoidance) and ambivalent individuals have a negative self-model and positive other-model (high anxiety and low avoidance; Bartholomew & Horowitz, 1991; Griffin & Bartholomew, 1994).
Securely attached infants tend to have mothers who are responsive to them (Ainsworth et al., 1971). They have a high sense of self-worth, do not fear intimacy in close relationships and find it easy to trust that others will be available when they turn to them (Griffin & Bartholomew, 1994). Avoidant infants tend to have mothers who are insensitive to them (Ainsworth et al., 1971). Insecure–dismissing (avoidant) individuals avoid close relationships with others, as they do not expect that others will be responsive when they turn to them. They hold a high sense of self-worth, dismiss their need for close relationships and become self-sufficient (Griffin & Bartholomew, 1994). On the other hand, insecure–fearful (avoidant) rely highly on others for self-validation but avoid intimacy in close relationships for fear of rejection (Griffin & Bartholomew, 1994). Lastly, insecure–preoccupied/ambivalent infants tend to have mothers who are inconsistent in their responsiveness to them (Ainsworth et al., 1971). They have low self-worth and consistently seek out close relationships with others for self-validation (Griffin & Bartholomew, 1994).
Researchers have argued that individuals fall somewhere along these anxious (self-model)/avoidant (other-model) dimensions rather than within separate category boundaries (Brennan et al., 1998). Therefore, we focus on attachment styles conceptualized along the two dimensions of self-model (anxiety) and other-model (avoidance).
We argue that followers with a positive self-model already have high self-worth and are less sensitive to rejection from leaders. They are not dependent on leaders for validation of their self-worth and, therefore, would be less susceptible to leaders’ demonstration of agape-based leadership behaviours. On the other hand, followers with a negative self-model tend to continually strive for self-acceptance by gaining the approval of their leaders. They are more sensitive to rejection and vulnerable to distress when their intimacy needs are not fulfilled. They have a high dependency on leaders to maintain positive self-regard and, therefore, would be more susceptible to leaders’ demonstration of agape-based leadership behaviours to fulfil their need for constant reassurance and validation. Therefore, we hypothesize:
Hypothesis 2: The relationship between agape-based leadership and follower outcomes will be moderated by followers’ self-model. Agape-based leadership will be more effective with regard to (a) satisfaction with the leader, (b) team commitment, (c) task satisfaction and (d) perception of leaders’ effectiveness when followers have a negative self-model as compared to a positive self-model.
We argue that followers with a negative other-model do not expect leaders to be available and responsive to them when called upon. They shun intimacy with their leader to avoid hurt and rejection. Therefore, they are less susceptible to leaders’ demonstration of agape-based leadership behaviours as they tend to maintain a high sense of independence. On the other hand, we argue that followers with a positive other-model seek out close relationships with their leaders as they expect that their leaders will be responsive to their needs. Therefore, they are more susceptible as they tend to value leaders more highly, expect their leaders to fulfil their needs and are comfortable with the closeness that is promoted by agape-based leadership. Therefore, we hypothesize:
Hypothesis 3: The relationship between agape-based leadership and follower outcomes will be moderated by followers’ other-model. Agape-based leadership will be more effective with regard to (a) satisfaction with the leader, (b) team commitment, (c) task satisfaction and (d) perception of leaders’ effectiveness when followers have a positive other-model as compared to a negative other-model.
Method
The laboratory experiment was performed in a quiet classroom with no outside distractions. Students were informed that they would be engaged in an 80-minute team activity on conflict resolution for the purpose of understanding what factors influenced team performance. Each session was conducted among 30–60 students each. We used management students to maximize internal validity as students have relative homogeneity because of their similar backgrounds in terms of age, education, experience, entrance test scores and cognitive abilities (Eva et al., 2013; Howell, 1986). To enhance the generalizability of the study, we selected those students who had recently finished internships with firms, had prior work experience and were all soon to be employed in organizations.
Participants
Participants were drawn from a student population who were enrolled in full-time programmes across seven leading management institutes in India. Previous leadership laboratory experiments have had sample sizes ranging from 144 to 194 (Grant & Hofmann, 2011; Howell & Frost, 1989; Jung & Avolio, 2000). We invited 360 students to participate in the experiment. Of these, 282 students acted as followers and 78 students acted as team leaders. The ages of the participants ranged from 18 to 30 years, and the median age was 22 years. Of the 270 follower participants that indicated their gender, 119 were female and 151 were male.
Task
The conflict resolution task was a ‘Lost at Sea’ consensus seeking team task (Nemiroff & Pasmore, 1975). This task consisted of a scenario where team members had to imagine that the ship that they had been travelling on was destroyed in a fire and that they were stranded in the middle of the ocean. They had to collectively arrive at a consensus on the ranking of 15 items (e.g., shaving mirror, mosquito net and water container) which were most important for their survival under the direction of their team leader. This task was selected as it provided opportunities for the team leaders to demonstrate agape-based/non-agape-based leadership behaviours during the discussions with their team members. Moreover, this task limited the extent to which participants used prior knowledge or experience to perform the task.
Procedure
Students were randomly divided into teams of five–six students each. Within each team, one of the students was assigned the role of the team leader and the other students were assigned the role of followers through the picking of lots. Thereafter, each of the students within the team were provided with kits containing an information sheet, consent form, and task and role descriptions. They were instructed to complete the attachment style measure and read their role and task descriptions. Students who were assigned the role of a team leader were provided with a one-page role description, which was created from the models of agape-based leadership and non-agape-based leadership. These descriptions were discussed with six experts consisting of two leadership experts, two methodological experts and two experts on Biblical studies. Team leaders were randomly assigned into either the agape-based leadership condition or the non-agape-based leadership condition and were additionally briefed separately by the researcher on their team leader roles.
Agape-Based Leadership Manipulation
Team leaders assigned the agape-based leadership condition were instructed to adopt a personal interaction style through a relaxed posture, friendly facial expressions (i.e., smiling, head nods) and direct eye contact. They were provided with a role description of the six agape-based behaviours that they needed to demonstrate during the task (i.e., self-sacrifice, empathy, humility, emotional control, forgiveness and impartiality) and also sample phases they could use during the task. Sample phrases included (a) ‘Although I am the leader, I may not know the best way forward. We must not be afraid of disagreement as differences of opinion will help us arrive at the best possible decision’ (reflecting humility); (b) ‘Ashish, I can see how upset you are’ (reflecting empathy); (c) ‘We all make mistakes…. Let us forgive one other, learn and move forward’ (reflecting forgiveness); (d) ‘It would be nice to hear from anyone that’s not yet had the chance to speak’ (reflecting impartiality); (e) ‘Ajay, you have got a lot of good points, but it is important to let Gita finish’ (reflecting patience); and (f) ‘Winning this game is not as important as arriving at a decision after considering every single person’s perspective’ (reflecting self-sacrifice).
Non-agape-based Leadership Manipulation
Leaders who were assigned the non-agape-based leadership condition were instructed to adopt an impersonal interaction style through an anxious posture, neutral facial expressions (i.e., absence of smiling and positive head nods) and intermittent eye contact. They were provided with a role description of the six non-agape-based behaviours that they needed to demonstrate during the task (i.e., self-interestedness, detachment, arrogance, impatience, unwillingness to forgive and bias) and also sample phases they could use during the task. Sample phrases included (a) ‘I know the best way forward’ (reflecting arrogance); (b) ‘I want results’ (reflecting detachment); (c) ‘I will not tolerate any excuses or mistakes!’ (reflecting unwillingness to forgive); (d) ‘Do not share your view if you have no knowledge and experience in this subject’ (reflecting bias); (e) ‘Do not waste my time’ (reflecting impatience); and (f) ‘We must win this game at any cost’ (reflecting self-interestedness).
Task Execution and Post-Task Evaluation
The team leaders led their team members on the task. After the task, we measured followers’ attitudes towards their respective team leader. During the post-experimental briefing, participants were asked about their perceptions regarding the activity purpose and how they thought they were supposed to respond. None of the participants indicated suspicion of the activity’s purpose or of the leaders’ roles.
Measures
Followers’ attachment styles. This was assessed using the self-report, continuous RQ measure of attachment style (Bartholomew & Horowitz, 1991). Respondents had to rate on a five-point scale ranging from 1 = ‘Not at all characteristic of me’ to 5 = ‘Highly characteristic of me’, the degree to which they resembled each of the four descriptions. Sample description for secure attachment style included ‘It is easy for me to become emotionally close to others. I am comfortable depending on others and having others depend on me. I don’t worry about being alone or having others not accept me’. Sample description of dismissing attachment style included ‘I am comfortable without close emotional relationships. It is very important to me to feel independent and self-sufficient, and I prefer not to depend on others or have others depend on me’.
Followers’ satisfaction with the leader. This was assessed using three items of the measure (Conger et al., 2000). Sample items included ‘I felt good to be around my team leader’ and ‘I was satisfied that my team leader’s style of leadership was the right one for getting our team’s task done’, with a five-point rating scale ranging from 1 = ‘Strongly Disagree’ to 5 = ‘Strongly Agree’. The Cronbach’s alpha was 0.80.
Followers’ team commitment. This was assessed using four items of the measure (Allen & Meyer, 1990). Sample items included ‘I felt as if my team’s problems are my own’ and ‘I felt a strong sense of belonging to my team’, with a five-point rating scale ranging from 1 = ‘Strongly Disagree’ to 5 = ‘Strongly Agree’. The Cronbach’s alpha was 0.72.
Followers’ task satisfaction. This was assessed using four items of the measure (Moyes & Redd, 2008). Sample items included ‘I enjoyed my task’ and ‘I liked the things I did during my task’, with a five-point rating scale ranging from 1 = ‘Strongly Disagree’ to 5 = ‘Strongly Agree’. The Cronbach’s alpha was 0.76.
Followers’ perception of leaders’ effectiveness. This was assessed using an item from the measures (Hassan et al., 2013) which included ‘Rate the overall effectiveness of your team leader’ with a five-point rating scale ranging from 1 = ‘Least Effective’ to 5 = ‘Most Effective’.
Manipulation Checks
To test the agape-based leadership manipulation, participants responded to the leaders’ agape measure (Carvalho & Mulla, 2021). Leaders’ agape was rated on a five-point Likert scale ranging from 1 = ‘Strongly Disagree’ to 5 = ‘Strongly Agree’. Sample items included ‘My team leader was open to the different views and opinions of the team’, ‘My team leader forgave mistakes’ and ‘My team leader freely admitted his/her mistakes or shortcomings’.
In the agape-based leadership condition (M = 4.05, SD = 0.41), participants perceived more agape leadership behaviours than in the non-agape-based leadership condition (M = 3.85, SD = 0.56). This difference was significant (t [244.70] = 3.36, p < .001, d = 0.40).
In order to ensure that the change in outcomes was due to the treatment of agape- and non-agape-based leadership and not due to group differences in leaders’ inspirational motivation behaviours, we measured leaders’ inspirational motivation behaviours post the treatment (Perdue & Summers, 1986; Podsakoff & Podsakoff, 2019). Participants responded to four items of the inspirational motivation measure (Fu et al., 2010) on a five-point Likert scale ranging from 1 = ‘Not Agree at All’ to 5 = ‘Strongly Agree’. Sample items included ‘The leader showed determination when accomplishing goals’ and ‘The leader communicated high performance expectations’. A t-test revealed no significant difference, (t [276] = 0.972, p > .05) in the inspirational motivation behaviours displayed for the agape-based leadership condition (M = 3.81, SD = 0.69) and the non-agape-based leadership condition (M = 3.73, SD = 0.76). This indicated that any change in the outcomes was due to the treatment of agape- and non-agape-based leadership and not due to group differences in leaders’ inspirational motivation behaviours.
Results
Correlations of the Measures
Calculation of the correlations showed that agape-based leadership was related to followers’ satisfaction with the leader (r = 0.55, p < .001), team commitment (r = 0.36, p < .001), task satisfaction (r = 0.26, p < .001) and perception of leaders’ effectiveness (r = 0.39, p < .001).
Main Effects
According to Hypothesis 1, agape-based leadership was expected to be positively related to follower work attitudes. To take into account the intercorrelations among the dependent variables, a MANOVA was conducted. The results are reported in Table 1.
ANOVA Results: Effect of Agape-Based Leadership on Followers’ Work Attitudes.
Wilk’s criterion yielded a significant overall effect for leadership style; Wilks’ Lambda = 0.84, F (4,271) = 12.30, p < .001, multivariate η2 = 0.15. Results revealed a significant impact of agape-based leadership on followers’ satisfaction with the leader, team commitment and perception of leaders’ effectiveness. Hypotheses 1a, 1band 1d were supported. However, results did not show a significant impact of agape-based leadership on followers’ task satisfaction. Hypothesis 1
Interaction Effects
Followers’ self-model attachment dimension was obtained by adding the ratings of their two attachment patterns with positive self-models (secure and dismissing) and subtracting the ratings of their two attachment patterns with negative self-models (preoccupied/ambivalent and fearful; Griffin & Bartholomew, 1994). Followers’ other-model attachment dimension was obtained by adding the ratings of their two attachment patterns with positive other-models (secure and preoccupied/ambivalent) and subtracting the ratings of their two attachment patterns with negative other-models (dismissing and fearful; Griffin & Bartholomew, 1994).
To test Hypotheses 2 and 3, we conducted hierarchical regression analyses. Non-agape-based leadership condition was coded as 0, and agape-based leadership condition was coded as 1. Leadership style along with followers’ attachment dimension (self- or other-model) were entered in the first block, followed by their multiplicative interaction term in the second block.
Hypothesis 2 postulated an interaction effect between leadership style and followers’ self-model attachment dimension on follower outcomes. As shown in Table 2, the interaction term between agape-based leadership and followers’ self-model on followers’ satisfaction with the leader was marginally significant (β = –0.143, p = .07). Moreover, simple slope analysis (Dawson, 2014) showed that when followers had a negative self-model (i.e., 1 SD above the mean), agape-based leadership had a positive relationship with followers’ satisfaction with the leader (β = 0.25, p < .05); when followers had a positive self-model (i.e., 1 SD below the mean), the relationship between agape-based leadership and followers’ satisfaction with the leader was not significant (β = –0.03, p = .75). The results supported Hypothesis 2a. Figure 1 illustrates the interaction effect. Hypotheses 2b, 2c and 2d were not supported.
Hierarchical Regression of Agape-Based Leadership and Followers’ Self-Model on Followers’ Satisfaction with the Leader.
a Coding: 0 = Non-agape-based leadership, 1 = Agape-based leadership.
+p < .10, *p < .05.

Hypothesis 3 postulated an interaction effect between leadership style and followers’ other-model attachment dimension on follower outcomes. As shown in Table 3, the interaction term between agape-based leadership and followers’ other-model on followers’ task satisfaction was significant (β = 0.179, p < .05). Moreover, simple slope analysis (Dawson, 2014) showed that when followers had a positive other-model (i.e., 1 SD above the mean), agape-based leadership had a positive relationship with followers’ task satisfaction (β = 0.17, p = .06); when followers had a negative other-model (i.e., 1 SD below the mean), agape-based leadership had a negative relationship with followers’ task satisfaction (β = –0.18, p = .05). The results supported Hypothesis 3
Hierarchical Regression of Agape-Based Leadership and Followers’ Other-Model on Followers’ Task Satisfaction.
a Coding: 0 = Non-agape-based leadership, 1 = Agape-based leadership.
+p < .10, *p < .05.

Discussion
The aim of this research was to manipulate agape-based leadership in an experimental approach, thereby contributing to the validation of the concept of agape in the leadership domain. Our research further enriches servant leadership theory which states that agape explains the core motive of servant leadership (Patterson, 2003). We showed that agape can be clearly identified and distinguished from other leadership styles and can be controlled under laboratory conditions and that individuals can be trained to exhibit agape-based behaviours. Our finding that agape-based leadership had a positive relationship with followers’ satisfaction with the leader, team commitment and perception of leaders’ effectiveness in conflict resolution is consistent with prior research (Carvalho & Mulla, 2021).
We did not find a significant main effect of agape-based leadership on followers’ task satisfaction. This was not surprising, as our analysis showed a significant moderating effect of followers’ attachment dimensions on this relationship, that is, when followers had a negative other-model, agape-based leadership had a negative relationship with followers’ task satisfaction, and when followers had a positive other-model, agape-based leadership had a positive relationship with followers’ task satisfaction.
When does agape-based leadership cause higher follower work attitudes? We provided deeper insight in the conditions under which agape-based leadership translated in leadership effectiveness. Our finding showed that agape-based leadership was more effective for followers with a negative self-model during conflict resolution. This was because such followers were able to fulfil their need for constant attention and appreciation in agape-based leaders. We did not find any impact for followers with a positive self-model. This may be due to the fact that such followers already had high self-worth and were, therefore, less susceptible to leaders’ demonstration of agape-based leadership behaviours. Our finding is consistent with research that has shown that followers’ attachment anxiety (negative self-model) is positively related to perceptions of transformational leadership (Hansbrough, 2012) and the argument that anxious followers may be more susceptible to transformational leadership (Yukl, 1999).
Our finding showed that agape-based leadership had a positive relationship with followers’ task satisfaction for followers with a positive other-model during conflict resolution. A possible explanation could be that such followers perceived agape-based leaders as capable of meeting their needs and were comfortable with the closeness and intimacy encouraged by them. This finding is consistent with research that shows that followers with a secure attachment style (positive other-model) would prefer to be led by socialized charismatic leaders who are focused on teamwork, collaboration and development of their people (Shalit et al., 2010).
Additionally, our finding showed that agape-based leadership had a negative relationship with follower task satisfaction for followers with a negative other-model during conflict resolution. A possible explanation could be that such individuals perceived the intimacy promoted by agape-based leaders as threatening in an attempt to maintain their independence. Such followers preferred emotionally distant relationships and, therefore, reported higher satisfaction under non-agape-based leaders. Our finding is consistent with the argument that followers with an avoidant attachment style (negative other-model) would prefer to be led by personalized charismatic leaders who are task-oriented, achievement-driven, and who display narcissism and self-aggrandizement (Shalit et al., 2010).
The results of our study may help to better understand the influence of follower characteristics as contingency elements of agape-based leadership and provide an understanding on how to adapt leadership style to match follower characteristics during conflict resolution.
Our study is more reliable than ‘paper–people’ vignette experiments, as it had a realistic manipulation where real people performed leadership behaviours in front of participants during a team task (Choi & Mai-Dalton, 1999). Our research adds knowledge to the literature on agape, leadership, attachment styles, conflict resolution strategies and follower work attitudes.
Limitations and Agenda for Future Research
Although our experiment provides causal evidence of the effectiveness of agape-based leaderships, our experimental design did not accurately replicate the complexities of the ‘real-life’ leader–follower relationship, as the leaders chosen were students. We selected students like in other prior leadership studies (e.g., Howell & Frost, 1989) to maintain the sampling homogeneity in terms of their ages, educational backgrounds, work experience, entrance test scores and so on. To enhance the generalizability, we selected business management students who would subsequently occupy managerial positions. The team task exercise was bounded; it lasted for only 80 minutes. In organizations, followers interact with their leaders over an extended time and, therefore, it is important that future research uses longitudinal designs to investigate the long-term effects of agape-based leadership on followers. They could also examine these hypotheses in the field where relationships between leaders and followers are more intense and prolonged.
We used an Indian sample which raises concern with the generalizability of the findings to different cultural contexts. It was interesting to note that while the Indian culture is more power-distance-oriented, followers reacted more positively to agape-based leadership. More research is needed to verify this aspect and validate the model in other cultural environments.
We encourage future researchers to control for other leadership styles such as ethical and authentic leadership and investigate other moderating variables (e.g., power distance orientation) that could influence the effectiveness of agape-based leadership.
Future studies would benefit from obtaining professionally trained actors to demonstrate agape-based leadership behaviours like in the study done by Howell and Frost (1989). Since the task we selected was less complex than in organizational work processes, it remains to be seen if this pattern generalizes to more difficult or creative tasks. We encourage future studies to use other attachment style measures to increase confidence in these findings.
Practical Implications
Our laboratory experiment provides an understanding of the processes that underlie leader behaviours during conflict resolution. We provide evidence that agape-based leadership can be studied under controlled laboratory conditions and that it is easy to train individuals to display agape-based leadership behaviours. This substantiates our claim that agape is trainable since it is an intentional behaviour arising out of will and not based on emotions or feelings (King, 2010; Lewis, 1952; Manby, 2012; Oord, 2005).
This study provides a practical set of tools for demonstrating agape-based leadership which proves useful for educators and consultants. Our study strengthens the argument that moral leaders can be developed by focusing on elements that can be controlled rather than just relying on stable personality attributes. Agape-based leadership may be used for conducting planned leadership training and development interventions. Our findings have important implications for leaders, as these emphasize the need for them to alter their leadership style depending on their follower’s attachment dimensions which can be assessed as part of their hiring, training and evaluation. Followers may be asked to respond to the attachment measure. Additionally, leaders can identify followers’ attachment styles by observing their behaviours and adapt their own behaviour accordingly. For example, followers with a positive other-model will turn to leaders for help when stressed, followers with a negative other-model will distance themselves from leaders and maintain their independence, and followers with a negative self-model will be clingy and frequently attempt to seek reassurance from leaders.
Conclusion
The model of agape-based leadership that has been developed in this article and its empirical support are important contributions towards our understanding of this leadership phenomenon in organizational settings. We hope that this research on agape-based leadership makes it an intriguing field of study and practice, especially at a time when we are grappling with the COVID-19 crisis. We hope this study encourages scholars and practitioners to bring ‘followers’ into the spotlight to accurately determine the benefits and costs of various leadership forms. Our research emphasizes the need to determine which follower profiles are suited to which leadership style to build a holistic understanding that explains both sides of the leader–follower equation.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.
