In Greek, the word charisma means gift of grace. Charismatic authority is derived from faith in the leader's exemplary character (Conger & Kanungo, 1987). Furthermore, "the charismatic is set apart from ordinary men and treated as endowed with supernatural, superhuman, or at least … exceptional powers and qualities … [which] are not accessible to the ordinary person but are regarded as of divine origin or as exemplary, and on the basis of them the individual concerned is treated as a leader" (Weber, 1968:63). Charismatic leadership attempts to radically alter and shape current realities. Ultimately, the success of charismatic relationship depends on followers as well as leaders (Bass, 1985). Indeed, to a large extent, the degree to which followers display admiration, affection and trust depends on the leader's personality and their perceptions of it.

House (1977) proposed a set of testable hypotheses about leaders' personal characteristics, leaders behaviors, and their effects on followers. To House, the personal characteristics of the charismatic leader include a high degree of self-confidence, strong moral convictions, and a tendency to influence others as well as engaging in impression management behaviors to boost trust and confidence in the leader. Furthermore, the articulation of a mission, setting challenging goals, and arousing motives are also important.

To understand transformational leadership, we must differentiate it from transactional leadership. Transactional leadership is based on the exchange process where the leader administers rewards and sanctions. One way or another, the leader and follower agree, explicitly or implicitly, that desired follower behaviors will be rewarded, while undesirable behaviors will draw out punishment. Potential rewards include an increase in salary, promotions, and more benefits. Conversely, penalties may include pay cuts, demotions, and terminations. It can be seen that this type of leadership is not satisfactory for most situations. Indeed, one could say that transactional leadership behaviors do not even qualify for a "true" leadership label (Bryman, 1992). Since it is based on exchange, transactional leadership does not seek to motivate followers beyond the level that is required to avoid punishment or gain extrinsic rewards. In sum, complete dependence on this leadership style may cause performance and satisfaction to suffer (Bass, 1985; Bryman, 1992; Burns; 1978; Peters & Austin, 1985).

In light of the above realization, Bass (1985) proposed a new theory of transformational leadership and outlined its components. Transformational leadership is thought to achieve remarkable levels of performance from followers. It engages followers by appealing to their upper level needs (e.g. self-actualization) and ideals that yield higher levels of follower satisfaction, performance, and organizational commitment in individuals (Bass, 2000; Bryman; 1992; Shin & Zhou; 2003) and teams (Bass, Avolio, Jung, & Berson, 2003; Pillai & Williams, 2004). For example, Fernandes and Awamleh (2004) found that transformational leadership had a strong effect on job satisfaction. Although Burns and Bass agree on the definitions of transactional and transformational leadership, they hold contrasting views of the relationship between these two constructs. Burns (1978), on one hand, viewed them as opposite ends of a continuum whilst Bass (1985) sees them as being closely related. An effective leader will exhibit aspects of both transactional and transformational leadership. According to Bass, transformational leadership is more concerned with developing followers to their fullest potential (Bass & Avolio, 1990), whereas the focus of transactional leadership is on satisfying basic follower needs.

Empirical evidence offers support for Bass's view that to maximize their effectiveness, leaders should exhibit transformational and transactional behaviors (Avolio, Waldman, & Einstein, 1988; Waldman, Bass, & Yammarino, 1989). In other words, transformational leaders should be capable of engaging in transactional behavior. Hence, transformational leadership does not serve as a substitute for transactional leadership; rather, it builds upon and augments transactional leadership in achieving desired goals (Bass & Avolio, 1990).

As modeled by Bass, transactional leadership is comprised of two fundamental dimensions: contingent rewards and management-by-exception, while transformational leadership is comprised of four central components: charisma, inspiration, individualized consideration, and intellectual stimulation. Contingent rewards take place when the leader and follower have a mutual understanding of the rewards or sanctions for performance or non-performance. The emphasis is on completing tasks that have been agreed upon based on previous expectations. In Management-by-Exception, however, the leader takes action only when major deviations from plans are evident.

Charisma is the key component of transformational leadership, it generates profound emotional connection between the leader and follower and it creates excitement about the mission (Bass, 1985). Charisma is operationalized through vision where the charismatic leader earns the respect and trust of followers, which leads to the acceptance of challenging goals (Bass & Avolio, 1990). The second transformational component is inspiration where leaders communicate their vision with optimism and enthusiasm. They also use symbols to heighten awareness of desired goals (Bass & Avolio, 1990). Another major component of transformational leadership is individualized consideration where the leader gives personal attention to followers by treating them "differently but equitably" (Bass & Avolio, 1990). In exhibiting individualized consideration, the leader first identifies the individual needs and abilities of followers and then mentors and coaches them, and also uses delegation. Finally through intellectual stimulation, the leader helps followers to think on their own and analyze problems from their personal perspectives, encourages creativity, innovation, and challenges conventional wisdom (Bass & Avolio, 1990).

Bass (1985) developed the Multifactor Leadership Questionnaire (MLQ) to measure the components of transformational and transactional leadership. Since its development, the MLQ has received extensive evidence of its reliability and validity, and is commonly used in leadership research (Bryman, 1992).

Wells and Marwell (1976) define self-esteem as being a set of attitudes and beliefs that a person brings with him or herself when facing the world. Self-esteem is commonly addressed in management research. It has been used to explore such areas as conformity, responses to threats, social participation, competitive behavior, and causal attributions. Moreover, it has been studied under a variety of labels. Some of the related terms include: self-love, self-confidence, self-respect, and self-worth. High self esteem is associated with risk taking, job satisfaction, and low inclination to please others (Brockner, 1988).

This construct refers to the generalized beliefs that individuals have regarding the significance of leadership to organizations which may influence how they see their leaders (Meindl, Ehrlich, and Dukerich; 1985). Consequently, these beliefs enhance followers' perceptions of charismatic/transformational qualities. To investigate this proposed individual difference, Meindl and Ehrlich (1988) developed the Romance of Leadership Scale (RLS). However, up to this point, only inconclusive empirical evidence is available regarding this concept (e.g., Al-Dmour & Awamleh, 2002; Awamleh & Gardner, 1999; Meindl, 1988).

The UAE banking system is unique from other banking systems within the Arab and non-Arab world. First, it is highly congested. For a relatively small population of approximately three million, the UAE tends to have a large number of banks as well as bank branches[1]. Currently the UAE has 49 national and foreign banks. In terms of bank branches, domestic banks have a total of 263 branches, or an average of 12 per bank. Foreign banks on the other hand are limited to a maximum of eight branches per bank[2]. Currently, there are 106 branches of foreign banks or an average of 4 branches per bank. In addition to this there are 45 representative offices of foreign banks. The large number of banks in the UAE has led to a branch density of 12.8 branches for every 100,000 people. If one compares the number of banks and bank branches with UAE nationals then it has the highest levels of bank congestion in the world[3].

The banking sector in many countries has undergone considerable development and is now a mature industry with established styles of management and leadership. This is not the case with banking in the UAE, which is relatively young compared to other countries. Nonetheless, the banking sector is well capitalized and profitable. The financial sector in the UAE in general is underlined by a solid legal framework and judicial system.

The government actively encouraged the growth in the banking sector after the discovery of oil. Prior to the discovery of oil, in 1960, there was only one foreign and no local bank in the UAE. Despite attempts by the Abu Dhabi government to encourage other foreign banks, none were willing to enter the market (Mahate et al., 2004). However, the discovery of oil changed this and the main shift towards the growth of banking took place after the first OPEC price rise in 1967. The large oil revenues allowed the government of Abu Dhabi and some of the other emirates to establish local banks in order to recycle surplus funds. Therefore, it is no surprise that except for two, all local banks have direct or indirect government shareholdings. The governmental shareholding of local banks has not hindered the entrance of foreign banks into the market, even though the latter are limited to only eight branches. The main factor leading to the growth of foreign banks has been the need to service the high number of expatriates as well as the ability to earn high profits. More recently, emirates such as Dubai have been actively encouraging foreign banks to establish branches within the emirate through the creation of financial free zones such as the Dubai International Financial Center.

A total of 865 questionnaires were distributed to banks operating in the United Arab Emirates. The number of questionnaires delivered to each bank was determined by the size of its workforce. The target population was full-time bank employees, working in non-managerial positions. The questionnaire included a total of 85 items and a detailed discussion of the questionnaire components is given in the following section. From the questionnaire distributed, two hundred and forty seven questionnaires were collected, of which 13 were excluded due to incomplete data. The final sample consisted of 194 respondents.

The majority of the respondants hold Bachelor degrees, were males aged between 25 to 40 years of age and on average had been employed by the Bank for 5 years. India, UAE and Pakistan were the dominant countries of origin.

The Multifactor Leadership Questionnaire (MLQ-5x/Short Form) (Bass & Avolio, 1995) was used to measure subjects' perceptions of transformational and transactional leadership styles. In this study, four subscales were used as a measure for transformational leadership: Charisma, for example, "the manager instills pride in being associated with him"; Inspiration, for example, "the manager talks enthusiastically about what needs to be accomplished"; Individualized Consideration, for example, "the managers helps me develop my strengths" ; and Intellectual Stimulation, for example, "the manager seeks differing perspectives when solving problems". Transactional leadership style was measured by two subscales, Contingent Rewards, for example, " the manager provides me with assistance in exchange for my efforts", and Management-by-Exception, for example, "the manager waits until things go wrong to take action."

Self-esteem was measured by using Rosenberg's (1979) 10-item scale. Example items of this scale include:"I feel I have a number of good qualities", "I am able to do things as well as most other people", and "At times, I think I am no good at all". The Romance of Leadership scale developed by Meindl and Ehrlich (1988) was adopted. The original Scale contained 32 items, however, several versions of RLS have appeared since its development. The current study used Form C (RLS-C), which has 11 items. Examples of items include: "when it comes right down to it, the quality of leadership is the single most important influence on the functioning of an organization", "the process by which leaders are selected is extremely important", and "a company is only as good as its leaders".

The authors also adopted two instruments to measure the dependent variables developed in a previous study (Al-Dmour, & Awamleh, 2002). Job satisfaction was assessed by a 14-item scale covering areas normally tapped in organizational behavior research. Examples of items include "In general, I am satisfied with work" and "I find that my opinions are respected at work". The self-assessed performance scale comprised of 5 items, such as, "I consider my performance better than the average person in my company" and "I always reach my work targets". All measures used a 5 point likert scale.

As a first step, scale reliability coefficients (Cronbach Alphas) for all measures adopted in this study were computed. Nunnally (1978) maintains that reliabilities which are less than 0.6 are considered poor, those in the 0.7 range are acceptable, while those above 0.8 are good. Results showed that the transformational leadership style scale reliability estimate is 0.89, and that of transactional is 0.71. Job satisfaction scale showed a reliability of 0.72, while performance scored 0.75. Reliabilities for self-esteem and RLS were 0.77 and 0.79 respectively.

Intercorrelations among all variables used in this study are summarized in Table 1. As expected the dependent variables are significantly correlated (r = 0.30, p < 0.001). It is also worth noting that transformational and transactional styles of leadership are highly correlated (r = 0.66, p < 0.001), which is not surprising given the fact that they act as paired and not as contradictory factors.

Two multiple regression models were run to test the first three hypothesis. Table 2 shows results of the multiple regression with satisfaction (m = 2.49; SD = 0.67) as the dependent variable and transformational leadership (m = 2.35; SD = .63), transactional leadership (m = 2.53; SD = 0.77), self-esteem (m = 2.24; SD = .63), and RLS (m = 2.23; SD = 0.55), as the independent variables. The overall model is significant at p < 0.001. The model revealed a significant impact of self-esteem (p <0.001) and transformational leadership (p <0.001). Conversely, transactional leadership and RLS failed to show any significant relationship with job performance.

Results of the second multiple regression are shown in Table 3. The dependent variable was self-perceived performance (m = 1.95; SD = 0.60) and self-esteem, RLS, transformational leadership, and transactional leadership were the independent variables.. Like the first test, the overall model is significant at p < 0.001. Both transformational leadership style, and self-esteem and RLS showed significance at the p < 0.001 level. Transactional leadership showed no relationship.

To test the remaining hypothesis, a partial correlation controlling for self-esteem and RLS (Table 4) was run. Compared to intercorrelations presented in Table 1, correlations between all main variables remained unchanged. The independent variables, RLS and Self-esteem, which showed significance earlier, do not seem to be moderating the main relationships.

The interactions of transactional leadership style and followers' performance and satisfaction require further studying. Bass' versus Burns' conceptualization of leadership styles is in need of further exploration, especially as they relate to a highly programmed work environment such as banking. Was the organizational variable, banks responsible for lack of significance in the transactional leadership effects? Or are the reasons more general or specific? As reflected in the above results this study did not capture the relationship or the situational factors were strong enough to override the relationship. It needs to be reiterated that performance was measured using a self-assessment instrument. One possible limitation of self assessed performance is overrated responses (example, Bretz, Mikovich, & Read; 1992). Further, the nature of contact that a bank employee has with his or her supervisor and the level of closeness were not assessed. Experience, training, personality attributes and success requirements (e.g., Micali, 1981) of employees were not assessed in relation to the other constructs.

Findings of this study confirm that a transformational leadership style of bank managers will boost employees' job satisfaction and performance (self-assessed). When managers operationalize charisma and utilize inspiration, individualized consideration and intellectual stimulate, they elicit positive reactions from employees. Such transformational qualities do stimulate higher level needs of followers and result in feelings of satisfaction. This finding is supported by other leadership researchers (e.g., Bass & Avolio, 1990). The attention that managers give to employees will be reflected in their general positive attitude toward work and work conditions, which in turn, increases job satisfaction and facilitates performance.

The findings in our study show that transactional leadership style does not positively relate to employee satisfaction or performance. They seem to respond more positively to a work system that defines their tasks and clearly spells out performance targets and expectations,thereby making patent performance-reward linkages. Moreover, a transactional leadership style may be redundant or irrelevant in the banking environment where the majority of tasks are highly standardized and routinized. As a result, employees look for space and flexibility in the process of performing tasks which is more balanced and complemented by transformational style rather than a transactional style.

In order to understand the results, it is imperative that we turn our attention to the whole model. Based on theory and earlier empirical studies, we predicted both independent variables to be significantly related to satisfaction and performance. We view both leadership styles to be complementary as shown in a number of studies (Bass (1985) Al-Dmour & Awamleh, 2002). However, results clearly show that only transformational and not transactional style is positively and significantly related to satisfaction and performance. This is a result that requires explanation. One interpretation of this result is that the functional operations of a bank are highly specialized and standardized at the processing level. The environment is controlled and tasks are structured, such that, there is little scope for transactional leadership. Individual performance, is for the most part, determined by the flow of work, (e.g., for bank tellers) more than the immediate demands and expectations of their managers.

On the contrary, transformational leadership style contributes considerably to both satisfaction and performance. While bank employees perceived their performance and satisfaction to be unaffected by the transactional skills of their manager, they attributed a big part of their satisfaction and performance to his/her transformational skills. In a routine and programmed environment, satisfaction and self assessed performance are a result of appealing to higher needs which helps overcome the routine. It is also a result of recognizing when to administer different types of rewards. What does this finding do to Bass's (1985) view that both leadership styles are closely related and are, in effect, complimentary? Clearly, not so in this case. In fact, the current findings give support to Burns's (1978) position that these two styles are at opposite ends of the continuum. Perhaps the findings represent only an exception to Bass's conceptualization and are situation specific.

Mixed support was given to the romance of leadership construct. Results showed it to be positively related to performance, but not related to satisfaction.

The more positive a disposition toward leadership that a bank employee has, the more likely he or she will have high perceived performance, but not satisfaction. In particular, subjects who believe in the importance of leadership seem to think that it can facilitate their performance. This finding supports the general premise of the construct. Satisfaction in this case seems to be affected by other and more diversified factors. Self-esteem was shown to be significantly related to both performance and satisfaction. Specifically, the higher the self-esteem an employee has, the higher their self-assessed performance and satisfaction. Contrary to expectations however, neither constructs played the role of a moderator.

Areas deserving attention include the relationship between leadership styles and independently measured performance. Also worthy of scholarly attention is the assessment of effects that experience, level of skills, career aspirations have on perceptions of leadership. In addition, the satisfaction relationship should be further explored. For example, how would task structure, position power, and group norms impact satisfaction. In this regard, the UAE is a fertile ground where a very large percentage of the labor force is expatriate. Such a study will become even more relevant in light of the nationalization policy that the government is implementing (Emaritaization). This environment presents promising opportunities for cultural leadership studies.