Loneliness in Leadership: A Study Applied to the Portuguese Banking Sector

Abstract

In this study, we analyzed the feeling of loneliness in leadership in the Portuguese banking sector, seeking to identify variables that may instigate this feeling, such as gender, age, academic qualifications, function/position, number of working hours per week, and years of work/seniority, and the consequences that it may have, namely in terms of the decision-making process and the motivation of leaders. For this study, a quantitative research tool was used in the form of a questionnaire, which was applied to a group of collaborators, with leadership responsibilities, of the financial institutions authorized to operate in Portugal. We concluded that while some variables influence the feeling of loneliness in leadership (years of work, position, and academic qualifications), others do not (gender, age, and hours of work per week). We also found a relationship between loneliness and demotivation and proved that the feeling of loneliness affects leadership but does not affect decision making. The results are relevant, especially for the banking sector, which has undergone major restructuring in the Portuguese economy and needs guidance to face the country’s financial challenges.

1. Introduction

We aimed to analyze loneliness in leadership in the Portuguese banking sector. Although studies on leadership are common in the literature, the association of leadership with loneliness, with a focus on the banking sector, has not been sufficiently explored. Thus, it is important that studies on this topic continue to be encouraged (Yukl 1988), especially when applied to more specific contexts, which can add value to the concept itself and help us understand it better.

In order to better understand this context, we endeavored to answer the following research question: Which variables most influence loneliness in leadership in the Portuguese banking sector?

The study of loneliness in leadership gains even more strength when faced with adverse circumstances, such as the current one created by the pandemic, which caused greater isolation and feelings of loneliness, which are particularly relevant in older leaders (Miller et al. 2020). In fact, the pandemic context (and the increase in remote working) promoted the feeling of loneliness in a general way (Andel et al. 2021; Charoensap-Kelly et al. 2021; Çolak and Çetin 2021; Holt-Lunstad 2021).

It is indisputable that leadership skills are important for the effective management of people and organizations (Cooke 2013). Leadership has proven to be increasingly relevant for organizations because it leaders are able to anticipate essential changes and create a strong commitment and a highly suitable atmosphere for workers and teams, so that they understand and adopt the necessary changes successfully. This action by the leaders is decisive, not only for the effectiveness of the organization, but also for its survival (Burke and Cooper 2006). In this sense, it is pertinent to analyze leadership in a context of loneliness and the impact that this context has on the level of leadership itself.

2. Literature Review and Hypotheses

What is leadership? The definitions of leadership are many (Alfarajat and Emeagwali 2021; Bass 1990; Bennis and Nanus 1985; Matos and Machado 2020; Rost 1991; Stogdill 1974). In order not to be exhaustive on a topic that already gathers a lot of information, we move forward with some definitions of this concept, which we consider important in terms of the state of the art. According to Vecchio (1987), leadership is the influence that a person has on others. Prentice (1961) relates leadership to the achievement of an objective. According to Lord and Hall (2005), a leader is someone with exceptional talent and who can develop his/her skills throughout his/her career. Ganta and Manukonda (2014) define leadership as a kind of power through which someone can influence or change the values, beliefs, behaviors, and attitudes of others. Katz and Kahn (1978) state that leadership is an influence that intends to go beyond compliance with the guidelines and the organization’s routine Leadership can also be defined as a relationship in which some people persuade others to adopt new values, attitudes, and goals and to exert efforts on their behalf (Hogg 2005). Banai and Reisel (2007) define leadership as the process of providing guidance and influencing others. Leadership also implies seeing the overall objective within the context of change and taking responsibility to motivate others to work towards it (Dickmann et al. 2010). By applying his/her knowledge and skills, the leader develops what Jago (1982) calls the Process of Leadership. When our traits influence our actions, we are facing trait leadership, which makes us believe that leaders are already born leaders and are not made leaders (Jago 1982). However, there are theories that indicate that it is possible for people to become leaders, either through their personality or through the context from which events conducive to leadership can emerge (Bass 1990). Also in this segment, it is important to mention authentic leadership, a key mediator between authentic leadership and desirable employee outcomes (Lux et al. 2019), and transformational leadership, which focuses on the effect of the leader on his/her followers and on their attitudes to achieve this effect (Aryee et al. 2012; Yukl 2002) and on teams (Ayoko and Chua 2014). Transformational leadership is also positively related to performance (Charbonnier-Voirin et al. 2010; Wang et al. 2011). This leadership style is based on a model oriented to the motivation of individuals, leading them to exceed their expectations and to improve their performance with greater ease (Bass and Avolio 2004). Regarding transformational leadership and bridging the gap with the other topic of this study—loneliness—some authors suggest that loneliness is more emergent when leaders are either new to the leadership role or enact more “transformational”, “transactional”, or “authentic” leadership behaviors (Silard and Wright 2020). Thus, transformational leadership may have a role in explaining loneliness at work (Dussault et al. 2017).

Another kind of leadership that is important to mention is servant leadership, which may promote employee performance (Kaltiainen and Hakanen 2020). Studies show that organizations face better periods of turmoil and change when faced with a servant leader (Kaltiainen and Hakanen 2020; Keith 2008).

This is particularly relevant in the current pandemic context. In fact, this context has also been associated with increased loneliness, which is why this study becomes even more pertinent (Wickens et al. 2021).

Studies on loneliness at work are also recurrent (Anand and Mishra 2021; Firoz et al. 2020; Sîrbu and Dumbravă 2019; Wright et al. 2006; Silard and Wright 2020; Zhou 2018). That is why it is important to define what loneliness is. Long and Averil (2003) define loneliness as a state of relative “detachment”, generally characterized by a decrease in social expectations. According to Hollenhorst and Jones (2001), loneliness is a psychological detachment from society, with the aim of cultivating our interior. Loneliness can also be seen as the general lack of a satisfactory personal, social, or community relationship. It is a lasting condition of emotional suffering that arises when a person feels distant, misunderstood, or rejected by others or lacks adequate social partners for activities that provide social integration (Anderson 1998). Furthermore, it is also important to note that social isolation and loneliness affect well-being (Rubin 2019) and that workplace loneliness inhibits creativity (Peng et al. 2017).

It is also important to distinguish loneliness from solitude, as both concepts tend to be confused when, in fact, they are different. According to McPherson et al. (2006), loneliness is an emotional state in which the individual experiences a feeling of emptiness, of having been rejected, disconnected, or alienated from other people. At the same time, Carter (2000) argues that solitude is the state of being alone and separated from other people, and often, it is related to a conscious and desired choice to be alone—that is, solitude is something we cultivate. In this way, loneliness (seen as solitude) also has positive aspects, as there are studies that show that people who experience periods of loneliness can gain a new understanding of themselves and their priorities (Long 2000). Loneliness can also offer an opportunity to engage in activities selected by us and free from social expectations (Burger 1995; Larson 1990). It is also important to mention a study about the loneliness of migrant workers in China, which shows that lonely workers are the most satisfied with their work (Chan and Qiu 2011).

Loneliness has also been linked to the concept of mindfulness, which is relatively recent in organization theory (Weick and Sutcliffe 2006) and which defines the state of awareness of what is happening around us (Brown and Ryan 2003). Mindfulness can decrease when individuals behave compulsively or automatically, without real awareness of the behavior of others (Brown and Ryan 2003; Deci and Ryan 1980). It is possible that loneliness potentiates mindfulness, because when the levels of stimulation drop significantly below the ideal—as occurs in a state of loneliness—the person can begin to generate (or better perceive) certain internal stimuli, such as sensations, thoughts, and emotions; that is, in a state of loneliness, the mind may be more “conscious” (Suedfeld 1982). In this segment—and because this study is based on leadership and loneliness—the concept of mindful leadership can be mentioned, which is described by Dickmann and Stanford-Blair (2009) as a complex process that can have several interpretations. The authors argue that leadership is associated with the nature and development of intelligence (namely emotional intelligence), which is related to conscious mental development. In this way, mindful leaders achieve greater effectiveness in leading and guiding their followers in reaching organizational goals, with a higher level of awareness and intelligence and sustainably. In this sense, we can say that holistic leadership will be essential in the future (Dickmann et al. 2008).

Linking leadership with loneliness, Gumpert and Boyd (1984) report that 52% of CEOs often feel alone. Although not all studies suggest that leaders are necessarily lonelier than their employees, many clearly indicate that leaders are at the top of the pyramid when talking about loneliness (Bell 1985; Bell et al. 1990; Hojat 1982). However, we cannot assume that all leaders feel lonely, because loneliness depends on the work environment in general and not only on the position in the organizational hierarchy (Wright 2012). However, if it is true that a leader must respond to the needs of his or her employees, there is not always someone who responds to the needs of leaders, which can generate loneliness (Kets de Vries 1989). Another factor that contributes to leaders’ loneliness is their greater access to resources and power. Consequently, employees can get closer to having access to those same resources. When a leader perceives this, he/she avoids his/her employees, which perpetuates his/her feeling of loneliness (Mao 2006). Lam and Lau (2012) also mention that in the workplace, due to technology and the growing existence of virtual teams, face-to-face communication is becoming more limited, which enhances loneliness. There are also studies that report that there is a tendency for people to try less when they work in groups than if they work alone (Liden et al. 2004), which can stimulate a leader to look for a more solitary job. At the same time, when loneliness is perceived by peers, the surrounding feelings can increase even more (Hareli and Rafaeli 2008). In addition, the quality of the employees’ interpersonal relationships (namely between them and the leaders) has a significant impact on the way they perceive and feel in their organizations (Carmeli 2009). In this sense, (Ryan and Deci 2000; Deci and Ryan 2014) argue that people have a need for interpersonal relationships and suffer when those needs are not met, even within organizations or cultures that are not very collective. However, it is curious to associate loneliness with leadership when this concept is closely linked with others, such as a group or a team (Kaiser et al. 2008; Van Vugt et al. 2008). There are studies that show that leader workplace loneliness is positively related to team turnover intentions via the effects of team cognitive trust, rather than affective trust in leaders (Chen et al. 2021).

It is also important to note that loneliness in leadership may be associated with other variables, such as gender, age, academic qualifications, function/position, number of hours of work per week, and years of work/seniority, which are precisely the variables that we intend to analyze in this study (the sociodemographic variables). Regarding gender, in general, its influence on loneliness in the workplace among leaders has several results (Stoltzfus et al. 2011). However, there is a tendency to consider that loneliness is influenced by gender, being typically considered a female problem and even more undesirable for men in many cultures (Lau and Gruen 1992). Several studies indicate that women are more likely than men to admit that they are lonely (Borys and Perlman 1985; Kleinke et al. 1982; Liu et al. 2019). This situation has been attributed mainly to social expectations and to psychosocial variables (i.e., self-efficacy, body image, and self-esteem) (Allgood-Merten et al. 1990). In some studies, it was found that women reported higher rates of depression (Weissman and Klerman 1977), but other studies also indicate that depression and loneliness rates have increased among men, especially among those in leadership positions (Rokach 2014). Several studies use measures of loneliness in which the participants call themselves lonely (that is, they state: “I am a lonely person”) and, in these cases, women usually report being lonelier than men. However, when the questions do not involve the word lonely, or the need for the person to have to label him/herself, there is little or no difference between genders (Borys and Perlman 1985; Radloff 1991). Finally, linking this topic to leadership, we can mention that there are studies showing that a leadership role is associated with greater loneliness for women, but not for men. In this case, Ong (2022) states that the greater loneliness for women is mediated by authenticity, since women experience less authenticity when they occupy leadership roles, but men do not. Ibarra et al. (2013) suggest that female leaders experience more loneliness in leadership positions due to subtle gender bias. Although many women work to eliminate gender from the equation and desire to be recognized only for their skills and talents as a leader, gender bias still exists. Robinson (2016) argues that many women with leading roles feel the need to function as “superwomen” to protect their image of being as competent as their male colleagues, which can promote loneliness (Rokach 2018). Another study conducted by Robinson and Shakeshaft (2015) suggests that female leaders have a greater sense of executive loneliness, related to a loss of friends and close work relationships.

Regarding function/position, Bell et al. (1990) found a small but negative correlation between the organizational level and loneliness, indicating that loneliness is more associated with those at the bottom of the hierarchy. This correlation is surprising, given that leaders with higher positions tend to work longer hours and spend less time with their families. As an explanation, the researchers argued that the social skills that drive these individuals up the hierarchical chain may also be responsible for their lower levels of loneliness. Page and Cole (1991) also suggest that functions with higher levels of management tend to experience less loneliness than those that imply lower levels. This conclusion goes against previous studies, already mentioned here, that report that top executives are more prone to being lonely due to the pressures of the role: increased social distance, lack of social support, and exhaustion related to the role (Zumaeta 2019). Another study, developed by Silard and Wright (2022), which involved 28 managers and 235 employees, concluded that although the results related to loneliness, in terms of the score obtained, do not vary much according to the hierarchical level, the loneliness predictors differ between both analyzed subjects (managers and employees), with emotional connection and mutuality predicting loneliness in employees but not in managers.

In the case of academic qualifications, Page and Cole’s research argues that economic status influences loneliness, as lower income and lower academic qualifications are influential factors in reporting loneliness. Professionals who typically have a higher income and more advanced educational levels are, according to Page and Cole, less likely to be lonely due to their economic and social well-being, which leads to great power. In fact, low power can increase loneliness and high power can decrease loneliness (Waytz et al. 2015). Moreover, we can say that the association between power and solitude is no longer new, since it is a relationship already explored by other authors, such as Foucault (Magalhães et al. 2015).

Concerning the number of hours of work per week, according to Bell et al. (1990), there is a strong positive relationship between hours worked and loneliness, but only for those who felt that their work group was not cohesive. The authors argue that if the work environment is hostile, working long hours can contribute more to loneliness. Therefore, the number of hours worked is irrelevant to loneliness if the work environment is cohesive and the person feels satisfaction at work (Bell et al. 1990). However, linking this with leadership, the long working hours that are typical of senior-level positions impose time constraints on leaders in social interactions (Kark and Eagly 2009). As a result, leaders’ attention must be split among many employees, limiting the time they can spend with each one of them, and thus the extent to which the leader can feel close to most of them (Magee and Smith 2013). In this way, the high number of working hours of a leader can lead to a greater feeling of loneliness.

Regarding the age factor, some studies have already tried to analyze the relationship between this variable and loneliness in leadership or in the decision-making process (Wright 2012). There is some evidence to suggest that younger leaders find it more difficult to adjust to the leadership role and can often experience so-called “command isolation”. This may be due to the social impact of making difficult decisions (Longnecker et al. 2006), which implies some maturity and experience.

Finally, regarding the years of work/seniority, the literature indicates that the behaviors necessary to reach seniority in an organization usually include having a high level of social confidence, emotional maturity, self-awareness, and social aptitude (Goleman 2004; Yukl 1988). At the same time, the literature on loneliness suggests that people who are most prone to loneliness are those who report shyness, low social competence, emotional instability, and low self-esteem (Ernst and Cacioppo 1998). In this way, effective leaders are usually well interconnected in their social relationships and those relationships are precisely what make them powerful and respected (Lee and Tiedens 2001). Thus, if we relate behavior to seniority and leadership, as more sociable people tend to stay longer in companies and to be more effective leaders, we can conclude that seniority in leadership tends to reduce the feeling of loneliness. For these reasons, these studies suggest that if an effective leader feels lonely, the reasons for this lie in personal, social, and contextual factors, rather than in seniority itself (Lee and Tiedens 2001).

In all, the relationship between the sociodemographic variables studied here (such as gender, age, academic qualifications, function/position, number of hours of work per week, and years of work/seniority) and the feeling of leadership has already been the subject of analysis by several authors (Allgood-Merten et al. 1990; Bell et al. 1990; Borys and Perlman 1985; Goleman 2004; Ibarra et al. 2013; Kark and Eagly 2009; Kleinke et al. 1982; Lau and Gruen 1992; Lee and Tiedens 2001; Liu et al. 2019; Longnecker et al. 2006; Magee and Smith 2013; Ong 2022; Page and Cole 1991; Radloff 1991; Rokach 2014, 2018; Robinson 2016; Robinson and Shakeshaft 2015; Stoltzfus et al. 2011; Yukl 1988; Waytz et al. 2015; Weissman and Klerman 1977; Wright 2012; Zumaeta 2019).

Hypothesis 1 (H1).

Sociodemographic variables influence the feeling of loneliness in leadership.

Another finding is that lonely workers reduce their levels of in-role performance (Lam and Lau 2012). We also know that the social context of work can greatly influence people’s behavior (Grant 2007). In addition, loneliness not only influences how people feel in organizations, but also their level of performance in the workplace; that is, the lack of security that this feeling can generate easily leads to lower performance (Ayazlar and Güzel 2014; Barsade and Gibson 2007; Ozcelik and Barsade 2011; Stoica and Brate 2013). There are also studies that demonstrate that loneliness can generate stress (Bakker et al. 2004; Maslach et al. 2001). Jung et al. (2021) argue that workplace loneliness decreases employees’ engagement with their jobs and the relationship between engagement and motivation is a reality, as less engagement leads to less motivation (Rosli and Hassim 2017). Studies also show that leaders’ loneliness has a significant negative impact on social self-efficacy and voice-taking behavior (Zhang et al. 2019), and that loneliness at work inhibits motivation (Barsade and Ozcelik 2018).

Hypothesis 2 (H2).

The feeling of loneliness affects the motivation for work.

In addition, several studies have shown that loneliness affects leadership. For example, loneliness affects performance and ends up generating occupational stress, which is important because there is a strong relationship between stress and organizational performance (Taris 2006; Vîrgă et al. 2019). It should also be noted that there are studies that associate loneliness with increased turnover (Erdil and Ertosun 2011). Thus, there are authors who focus on what steps can be taken to minimize loneliness in leadership, namely in terms of the negative impact it can have (Avolio and Gardner 2005; Boyd and Taylor 1998; Diddams and Chang 2012; Dussault and Barnett 1996; Goleman 1998; Mao 2006; Lincoln and Miller 1979), with one of the most suggested strategies being peer support from other leaders. However, although a leader must act with his or her peers and collaborators with a participative and open spirit, in practice, this is not always the case, because of the abyss between the high echelon and the other people of the hierarchical pyramid. Often, sharing problems with colleagues can lead to conflicts. As a result, even though leaders are constantly surrounded by people, they do not always feel comfortable sharing their decisions and concerns, and they become lonelier, which affects their leadership (Akande 1992). In this way, loneliness has a great influence on leadership (Rokach 2014).

Hypothesis 3 (H3).

The feeling of loneliness affects leadership.

Despite emphasizing the relational nature of leadership, Hollander (1964, 1978) also emphasizes the role of the leader’s internal processes, in terms of his/her self-awareness. According to Avolio et al. (2004) self-awareness is seen, in part, as being linked to self-reflection and introspection, which makes authentic leaders gain clarity and agreement in relation to their interests, values, emotions, and goals. Through self-awareness, the leader has better access to the information necessary to make fundamental choices and commitments and to align his/her actions with his/her values, vision, and goals (Bruch and Ghoshal 2004). This process of self-awareness is part of authentic leadership, because only through it can a leader find his/her place of reflection and renewal, where he/she listens to him/herself (Heifetz and Linsky 2002). Koestenbaum (2002) argues that loneliness can help to develop a leader’s vision, as he/she gains greater control over him/herself and over the space and time that surround him/her. The author also adds that the courageous leader is one who is prepared for loneliness. Beyond this, leader self-efficacy enhances the extent to which problem-solving pondering occurs when leaders feel lonely (Gabriel et al. 2020). Moreover, several studies indicate that there are many influences that lead to loneliness at work. These influences may be due to factors related to the individual or contextual factors. For example, the loneliness experienced in organizations can be related to responsibility and isolation in the decision-making process (Wright 2012). Additionally, studies demonstrate that high levels of loneliness in leadership predict less perceived self-control, which leads to a tendency to avoid risks—that is, loneliness tends to make leaders more conservative and less open to risk in the decision-making process. However, this effect is only significant in gain scenarios rather than loss scenarios (Wang and Zhu 2017).

Hypothesis 4 (H4).

The feeling of loneliness affects the decision-making process.

Thus, the intensity and frequency of loneliness experienced by leaders at work can vary and depend, to a large extent, on the qualitative aspects of the person and their work environment, instead of being determined by an objective variable (Wright 2012), which does not always find concordance in the literature. Both loneliness and organizations are complex phenomena, especially when it comes to the topic of leadership. Therefore, the association of these themes makes it necessary to carry out further studies on this matter.

However, if topics such as leadership and loneliness and their association among themselves are of great importance, it is pertinent to analyze them in specific contexts, as in this study, which focuses on the Portuguese banking sector. Banking leadership has also been the subject of studies, but without a focus on loneliness. Indeed, leadership is an essential foundation for banking organizations to operate today. In a study conducted at the banking institution Crédito Agrícola (Portugal), Matos and Machado (2014) concluded that leaders prefer to make decisions with the participation of other team members and not alone, a result that is particularly important in a study that aims to analyze loneliness in leadership with a focus on the banking sector.

It is also important to mention that the progress made by the Portuguese banking sector, in the period that followed the sovereign debt crisis, was significant. At the end of 2019, according to Associação Portuguesa de Bancos (2020), the sector showed considerable improvement in terms of efficiency, liquidity, asset quality, profitability, and solvency. The progress made is particularly relevant as, on top of the significant challenges already affecting the performance of banking activity, such as the environment of low interest rates, a new business model framework, growing competition by new operators, and the weight of legislation/regulations, the COVID-19 pandemic crisis generated an unprecedented economic shock and high levels of uncertainty. The coordinated implementation of monetary, fiscal, and regulatory policy measures was crucial in providing support to the economy and mitigating the impacts on economic agents.

The COVID-19 crisis brings new challenges and opportunities for the banking industry. Carletti et al. (2020) suggest that COVID-19 will accelerate some existing trends in the banking sector, will temporarily reverse others, and will influence the private and public players in the sector. Most importantly, it will accelerate the digitalization and restructuring of the sector. The potential effects of the COVID-19 pandemic and the measures needed to control it pose critical challenges for the banking sector in Portugal, which already faced significant pressure before the pandemic. Among these, the following challenges stand out: the deep economic recession, the maintenance of a low interest rate environment, the high level of uncertainty regarding the recovery, the high indebtedness level of the non-financial sector, and the potential increase in defaults.

For all these reasons, studying leadership in the banking sector is of great relevance, particularly involving the factors that affect it, such as the feeling of loneliness.

3. Research Methodology

3.1. Procedures

For this study, a quantitative research tool was used in the form of a questionnaire, which was applied to a group of collaborators, with leadership responsibilities, of the financial institutions authorized to operate in Portugal. Quantitative research is an approach for testing objective theories by examining the relationships among variables (Creswell 2014), and it is often applied to descriptive research that seeks to discover and classify relations between variables in addition to ascertaining the causal link that might exist between phenomena (Richardson et al. 1999).

Questionnaires have frequently been used for assessing and monitoring economic and social phenomena, as well as to compare several elements of a whole or dimensions within an element. The questionnaire was built from the analyzed literature and a focus-group, which determined the main topics to be addressed. Subsequently, the questionnaire was tested with professionals in leadership positions from other institutions. After collecting the suggestions that were given, it was possible to arrive at the data collection instrument used in the study. The research comprised a total of 20 questions which, apart from one, were closed-ended with options, and questions where the evaluation of statements was required. The survey by questionnaire involved managers as staff who coordinate the activities of other staff and whose function is to conduct business in the sense that the organizational objectives are achieved. For this purpose, the levels of management from the vertical perspective were considered: top, middle, and first-line managers. The questionnaire directed to those financial institution heads was a tool whose intent was to allow the researchers to access, evaluate, and monitor their opinions on various aspects associated with their perception of loneliness in leadership. The survey by questionnaire was implemented via an online platform, with a direct link sent by email straight to the potential respondents. This allowed the survey to be filled out promptly and be more convenient for the sample subjects, providing a greater guarantee of confidentiality, while respondents could feel free to express their views. Those filling out the questionnaires were guaranteed anonymity, the answers being treated on an aggregate basis, to preserve the identity of those who agreed to participate.

Concerning the research procedures,

Table 1. Summary of the study variables.

	Variables	Categories
Sociodemographic variables	Age	≤35 years; 36–45 years; 46–55 years; 56–65 years
	Gender	Male; Female
	Academic qualifications	12th year or equivalent; Degree or equivalent; Graduate or equivalent; Master; PhD
	Position	Administration; Intermediate Management; Line Management
	Working hours (per week)	≤40 h; 45 h; >45 h
	Years of work	≤10 years; 11–15 years; 16–20 years; >20 years
Variables related to the feeling of loneliness	Have you ever felt or feel loneliness when making decisions?	1—Strongly disagree 2—Disagree 3—No opinion 4—Agree 5—Strongly agree
	Do you consider that a leader’s loneliness is important and necessary?
	Do you admit that loneliness is a demotivating factor?
	Do you believe that loneliness in leadership affects only organizations whose decisions are very centralized?

Source: Own calculations based on own survey data.

presents a summary of the variables considered in this study. It includes both sociodemographic variables and variables related to the feeling of loneliness. Employees answered the questions related to the feeling of loneliness on a 5-point Likert scale from 1—“ strongly disagree” to 5—“strongly agree” (Likert 1932). It is important to highlight that the interest variable for our study is “Have you ever felt loneliness when making decisions?”. To answer H1, this variable was dichotomized considering: 1—“agree” (includes 4—agree and 5—strongly agree) and 0—“do not agree” (includes 1—strongly disagree, 2—disagree, and 3—no opinion).

3.2. Participants

According to Banco de Portugal (2021), the Portuguese Central Bank, institutions authorized to operate in the country are divided into 45 types, not including itself. Based on this information, the units of analysis targeted in our questionnaire were all bank headquarters and branches located in Portugal (domestic activity), including Banco de Portugal. Following Associação Portuguesa de Bancos (2020), on 31 December 2019, this association included 29 financial institutions that jointly accounted for 94.6% of the total value of the Portuguese consolidated banking assets. From these 29 financial institutions, only 25 provided information to the 2019 Annual Activity Report. At the end of 2019, the domestic activity total workforce of the financial institutions’ members was broken down as follows in

Table 2. APB HR by size and by leadership position.

By Size					By Leadership Position
Total	Men	%	Women	%	Total	Men	%	Women	%
28,301	14,203	50.2	14,098	49.8	7974	4904	61.5	3070	38.5

Source: Data obtained from the APB 2019 Activity Report (Associação Portuguesa de Bancos 2020).

.

In the same line, in what concerns Banco de Portugal (2020, p. 88), “As of 31 December 2019, the number of staff members employed by Banco de Portugal stood at 1778, of whom 1700 were active […] approximately 20% of the Bank’s staff members were managerial staff.” (

Table 3. Banco de Portugal HR by size and by leadership position.

By Size					By Leadership Position
Total	Men	%	Women	%	Total	Men	%	Women	%
1778	892	50.2	886	49.8	477	300	62.9	177	37.10

Source: Data obtained from the BP 2019 Annual Report—Activities and Financial Statements (Banco de Portugal 2020).

).

It is worth mentioning that BP presented the breakdown by quota of the three categories of leaders, which is not possible to extract from APB. From APB’s human resource statistics, it is not possible to isolate the four entities that did not provide information for the Activity Report—2019, so the total number of individuals with leadership functions was considered. This situation causes an excess estimate in the statistical data presented in

Table 4. HR by leadership position.

Total	Men	Women
8451	5204	3247

Source: Data obtained from Associação Portuguesa de Bancos (2020) and Banco de Portugal (2020).

, which will be used to calculate the sample.

The size of the sample was calculated bearing in mind the universe in question (8451 < 100,000), which is finite in nature, in which the confidence level established was 95% and the allowed estimation error was 5%. Hence, the following formula was used (Israel 1992):

$$n=\frac{N{z}^{2}p\left(1-p\right)}{\left(N-1\right)e^2+z^{2}p\left(1-p\right)}$$

where:

n = sample size;

N = size of the universe;

z = confidence level chosen in terms of deviation: 95% → z = 1.96;

p = proportion of characteristics researched in the universe. We assume p = 20%, as the proportion of active managerial staff of the bank’s staff members (Banco de Portugal 2020, p. 88);

q = proportion of the universe not having the characteristic researched (100% − p); and

e = allowed estimation error.

The formula gave the necessary sample size of 239 elements for the universe of 8451 individuals, considering an estimation error of 5%. The information we collected was only from 165 (69%) despite some insistence, which means that we have an estimation error of approximately 6% (6.04%).

The sampling techniques used in this study were of a non-probabilistic nature (quota sampling and purposive sampling). From the above result, 165 professionals from the Portuguese banking sector voluntarily participated in the study (aged 46.56 ± 9.86; 44.8% women and 55.2% men), belonging to 11 different banking institutions.

3.3. Statistical Analyses

Statistical analyses were performed using Statistical Package for Social Sciences (SPSS), V27.0^®, for Windows (IBM Corp 2020), and statistical significance was set at p ≤ 0.05. A descriptive statistical analysis (absolute and relative frequencies) was considered to describe the study variables.

The relationships between the feeling of loneliness (dichotomized) and the classification variables were checked using the Chi-square test. Cramer’s V was also obtained to measure the association between the variables. Its interpretation was made based on the following criteria (Rea and Parker 1992): <0.1, negligible association; 0.1–0.2, weak association; 0.2–0.4, moderate association; 0.4–0.6, relatively strong association; 0.6–0.8, strong association; 0.8–1.0, very strong association.

Then, a binomial logistic regression model was considered to assess which classification variables have a significant influence on the agreement of the feeling of loneliness (dichotomized). Only the classification variables whose univariate test was significant (p < 0.05) or showed a tendency for the existence of a significant relationship (p < 0.10) were selected for the model. The forward stepwise (likelihood ratio) selection method was considered, and the results were reported by odds ratio (OR) estimates and their 95% confidence intervals (CI). To evaluate the quality of the adjustment, Nagelkerke’s R2 was used. The model’s goodness of fit was assessed through the Hosmer–Lemeshow test and the area under the ROC curve (AUC) was used to evaluate the discriminative capacity of the model. Collinearity and multicollinearity situations were not detected, and the assumption of no extreme values was also verified.

The correlation between motivation for work, leadership, and decision making with feelings of loneliness was assessed by considering Spearman’s correlation coefficient, and its interpretation was based on the following criteria (Davis 1971): <0.09, negligible association; 0.10–0.29, low association; 0.30–0.49, moderate association; 0.50–0.69, substantial association; ≥0.70, very strong association.

4. Results

4.1. Relationship between the Sociodemographic Variables and the Feeling of Loneliness

The analysis of the results presented in

Table 5. Relationship between the sociodemographic variables and the feeling of loneliness (agree and do not agree).

			Have You Ever Felt or Feel Loneliness When Making Decisions?
		Total	Do Not Agree	Agree	p-Value #1	Cramer’s V
Gender	Male	74 (44.8%)	30 (40.5%)	44 (59.5%)	1.000	0.010
	Female	91 (55.2%)	36 (39.6%)	55 (60.4%)
Age	≤35 years	25 (15.2%)	14 (56.0%)	11 (44.0%)	0.124	0.187
	36–45 years	50 (30.3%)	23 (46.0%)	27 (54.0%)
	46–55 years	55 (33.3%)	17 (30.9%)	38 (69.1%)
	56–65 years	35 (21.2%)	12 (34.3%)	23 (65.7%)
Academic qualifications	12th year or equivalent	9 (5.5%)	5 (55.6%)	4 (44.4%)	0.068 **	0.208
	Degree or equivalent	59 (35.8%)	28 (47.5%)	31 (52.5%)
	Graduate or equivalent	40 (24.2%)	18 (45.0%)	22 (55.0%)
	Master and PhD	57 (34.5%)	15 (26.3%)	42 (73.7%)
Position	Administration	4 (2.4%)	0 (0.0%)	4 (100.0%)	0.066 **	0.182
	Intermediate Management	48 (29.1%)	15 (31.3%)	33 (68.8%)
	Line Management	113 (68.5%)	51 (45.1%)	62 (54.9%)
Working hours (per week)	≤40 h	26 (15.8%)	12 (46.2%)	14 (53.8%)	0.650	0.075
	45 h	62 (37.6%)	26 (41.9%)	36 (58.1%)
	>5 h	77 (46.7%)	28 (36.4%)	49 (63.6%)
Years of work	≤10 years	32 (19.4%)	18 (56.3%)	14 (43.8%)	0.004 *	0.283
	11–15 years	30 (18.2%)	17 (56.7%)	13 (43.3%)
	16–20 years	19 (11.5%)	3 (15.8%)	16 (84.2%)
	>20 years	84 (50.9%)	28 (33.3%)	56 (66.7%)

^#1 Chi-square test; * p < 0.05; ** p < 0.1. Source: Own calculations based on own survey data.

reveals a significant relationship between the feeling of loneliness and the years of work (p = 0.004), with a moderate association (V = 0.283). After 16 years of work, most agree with the feeling of loneliness when making decisions (16–20 years, 84.2% and <20 years 66.7%). We also conclude that there exists a tendency for the existence of a significant relationship between the feeling of loneliness and position (p = 0.066 < 0.1, weak association) and academic qualifications (p = 0.068 < 0.1, moderate association). From the total sample, it is important to highlight that in what concerns academic qualifications, only 5.5% have the secondary level (12th year or equivalent), and only 2.4% have a position in administration.

The remaining sociodemographic variables (gender, age, and hours of work per week) have no significant relationship with the feeling of loneliness (p > 0.05).

Table 6. Estimated regression coefficients from binary logistic regression analysis, only for significant variables (agree and do not agree).

	Coefficients	p-Value (Wald)	OR	IC95% OR
				Lower	Upper
Academic qualifications (a)		0.008
Degree or equivalent	0.734	0.321	2.083	0.489	8.862
Graduate or equivalent	0.913	0.236	2.493	0.550	11.305
Master and PhD	2.156	0.008	8.633	1.766	42.208
Years of work (b)		0.001
11–15 years	−0.156	0.774	0.856	0.295	2.483
16–20 years	1.983	0.008	7.266	1.665	31.704
>20 years	1.452	0.002	4.273	1.669	10.937
Constant	−1.675	0.042	0.187
R2Nagelkerke = 0.205; pHosmer and Lemeshow = 0.504; AUC = 0.722 (IC95% = [0.644, 0.800])

OR, odds ratio; ^(a) reference category = 12th year or equivalent; ^(b) reference category = ≤10 years; AUC, area under the ROC curve. Source: Own calculations based on own survey data.

reveals that only the independent variables of academic qualifications and years of work were included in the logistic regression model (p < 0.05). The overall model presents a well-fitting value (pHosmer-Lemeshow > 0.05), and its discriminant capacity is also suitable (AUC = 0.722).

Considering the obtained results, we may conclude that a professional who has a degree or equivalent is twice as likely to feel loneliness when making decisions compared to a professional who has a 12-year education or equivalent (OR = 2.083, IC95% = [0.489–8.862]). On the other hand, a professional who has a graduate degree or equivalent and a master’s degree or PhD is 2.5 and 8.6 times more likely, respectively, to feel loneliness when compared to a professional who has a 12-year education or equivalent (OR = 2.493, IC95% = [0.550; 11.305] and OR = 8.633, IC95% = [1.766; 42.208]).

Regarding the years of work, a professional who has worked 11–15 years in the banking sector has a 14.4% lower likelihood of feeling loneliness when making decisions when compared to a professional who has worked ≤10 years in this sector (OR = 0.856, IC95% = [0.295–2.483]). A professional who has worked 16–20 years or >20 years is 7.3 and 4.3 times more likely, respectively, to feel loneliness compared to a professional who has worked ≤10 years in the banking sector (OR = 7.266, IC95% = [1.665; 31.704] and OR = 4.273, IC95% = [1.669; 10.937]).

4.2. Relationship between the Feeling of Loneliness and Motivation to Work, Leadership, and Decision-Making Processes

From these results, we can observe that 60% of the respondents (agree and strongly agree) have felt or feel loneliness when making decisions, and agree and strongly agree that loneliness is a demotivating factor. Although these feelings are important, almost 62% (disagree and strongly disagree) have the opinion that a leader’s loneliness is not important or necessary.

Asked about whether loneliness in leadership affects only organizations whose decisions are very centralized, 47% do not share this opinion, while 26% have no opinion on this issue.

Table 7. Sample characterization regarding the loneliness in leadership.

Variables	1—Strongly Disagree	2—Disagree	3—No Opinion	4—Agree	5—Strongly Agree
Have you ever felt or feel loneliness when making decisions?	3 (1.8%)	51 (30.9%)	12 (7.3%)	69 (41.8%)	30 (18.2%)
Do you consider that a leader’s loneliness is important and necessary?	25 (15.2%)	77 (46.7%)	23 (13.9%)	38 (23.0%)	2 (1.2%)
Do you admit that loneliness is a demotivating factor?	11 (6.7%)	41 (24%)	14 (8.5%)	81 (49.1%)	18 (10.9%)
Do you believe that loneliness in leadership affects only organizations whose decisions are very centralized?	8 (4.8%)	70 (42.4%)	43 (26.1%)	35 (21.2%)	9 (5.5%)

Source: Own calculations based on own survey data.

presents the sample characterizations regarding loneliness in leadership.

The results of

Table 8. Relationship between motivation for work, leadership, and decision making with the feeling of loneliness.

		Do You Admit That Loneliness Is a Demotivating Factor?	Do You Consider That a Leader’s Loneliness Is Important and Necessary?	Do You Believe That Loneliness Affects Only Organizations Whose Decisions Are Very Centralized?
Have you ever felt or feel loneliness when making decisions?	Correlation Coefficient #2	−0.243	0.381	−0.024
	p-value	0.002	<0.001	0.761

^#2 Spearman’s correlation coefficient. Source: Own calculations based on own survey data.

reveal a significant correlation between the feeling of loneliness and the demotivation for work, being negative and low (p = 0.002, r = −0.243), which means that with the increase in the feeling of loneliness, the less it affects the motivation for work. The correlation between the feeling of loneliness and the importance of leadership is also significant, being positive and moderate (p < 0.001, r = 0.381). With the increase in the feeling of loneliness, greater is the agreement on the importance of a leader’s loneliness. On the other hand, the feeling of loneliness and the decision-making process are not significantly correlated (p = 0.761), which means that the feeling of loneliness does not affect the organizations in decision making.

5. Discussion

Regarding the first hypothesis—the sociodemographic variables influence the feeling of loneliness in leadership—we can conclude that there are some variables that affect the feeling of loneliness in leadership more than others. For example, the analysis of the results revealed a significant relationship between the feeling of loneliness and the years of work. There also seems to be a relationship between years of work and loneliness in the decision-making process, but not in a general way, and the range that is associated with less loneliness is that between 11 and 15 years of work. However, the literature has not determined a direct relationship between loneliness and the number of years of work, indicating that other variables may be the source of the feeling of loneliness, since people with more years of service are usually more sociable, which goes against the profile of people who tend to feel lonelier. In fact, the literature reveals that years of work, by themselves, are not sufficient to determine the feeling of loneliness (Ernst and Cacioppo 1998; Goleman 2004; Lee and Tiedens 2001; Yukl 1988). However, if we relate behavior to seniority and leadership, as more sociable people tend to stay longer in companies and tend to be more effective leaders, we could conclude that seniority in leadership tends to reduce the feeling of loneliness. However, the literature suggests that if an effective leader feels lonely, the reasons for this are rooted in personal, social, and contextual factors, and not so much in seniority itself (Lee and Tiedens 2001).

In the case of the position, we conclude that there exists a tendency for the existence of a significant relationship between the feeling of loneliness and this variable. Bell et al. (1990) argue that loneliness is more commonly associated with those at the bottom of the hierarchy. However, our conclusion fits with the studies of other authors who defend that people with leadership positions tend to feel lonelier than people with lower positions (Bell 1985; Bell et al. 1990; Hojat 1982). It also fits with the study of Zumaeta (2019), who argues that top executives are more prone to being lonely due to the pressures of the role, increased social distance, lack of social support, and exhaustion related to the role. Silard and Wright (2022) stated that the loneliness predictors differ between managers and employees, with emotional connection and mutuality predicting loneliness in employees but not in managers.

Regarding academic qualifications, we conclude that a professional who has a higher degree has more chances of feeling loneliness when making decisions when compared with a professional who has a lower degree. However, Page and Cole’s (1991) research suggests that lower academic qualifications are an influential factor in reporting loneliness, because economic status influences loneliness, and lower income and lower academic qualifications are influential factors in reporting loneliness. Professionals who typically have a higher income and more advanced educational levels are, according to Page and Cole, less likely to be lonely due to their economic and social well-being, which leads to great power. In fact, according to the literature and assuming the relationship between higher academic qualifications and greater power, low power can increase loneliness and high power can decrease loneliness (Waytz et al. 2015). In this sense, higher academic qualifications can lead to a lesser feeling of loneliness.

For the remaining sociodemographic variables (gender, age, and hours of work per week), we found no significant relationship with the feeling of loneliness, which contradicts the literature that anticipates this relationship, namely with regard to gender, suggesting that women in leadership positions tend to feel more loneliness (Ibarra et al. 2013; Ong 2022; Robinson and Shakeshaft 2015; Rokach 2018), age, since, there is some evidence to suggest that younger leaders find it more difficult to adjust to the leadership role and can often experience the so-called “command isolation” (Longnecker et al. 2006), as well as hours of work. In this specific case, the literature suggests that the long working hours that are typical of senior-level positions impose time constraints on leaders in social interactions (Kark and Eagly 2009). In this way, the high number of working hours of a leader can lead to a greater feeling of loneliness.

From the second hypothesis—the feeling of loneliness affects the motivation for work—we conclude that the feeling of loneliness promotes the demotivation for work, which agrees with the studies that indicate that loneliness at work inhibits motivation (Barsade and Ozcelik 2018). Additionally, Jung et al. (2021) argue that workplace loneliness decreases employees’ engagement with their jobs and the relationship between engagement and motivation is a reality, as less engagement leads to less motivation (Rosli and Hassim 2017).

Concerning the third hypothesis—the feeling of loneliness affects leadership—we conclude that the correlation between the feeling of loneliness and the importance of leadership is significant. This is in line with studies that indicate that loneliness has a significant influence on leadership (Rokach 2014). Furthermore, although leaders are always surrounded by people, they do not always feel confident in sharing their decisions and concerns and thus become lonelier, which affects their leadership (Akande 1992).

Regarding the fourth hypothesis—the feeling of loneliness affects decision making—according to the results of this study, there is no relationship between these two variables, which does not agree with Wright’s (2012) ideas nor Wang and Zhu’s (2017) studies, which demonstrate that high levels of loneliness in leadership predict less perceived self-control, which leads to a tendency to avoid risks; that is, loneliness tends to make leaders more conservative and less open to risk in the decision-making process.

In short, we believe that the differences found between the literature and the conclusions of our study are due to the specificity of the studied sector, which constitutes an interesting contribution to this field of research, allowing us to take into account the need to consider the activity sector which, due to its own characteristics, can cause its leaders to express different levels of loneliness in the development of their tasks and face different consequences arising from this loneliness.

6. Conclusions and Future Research

We carried out this study to analyze the loneliness in leadership with application to the Portuguese banking sector. Regarding the research question—which variables most influence loneliness in leadership in the Portuguese banking sector?—we can conclude that while some variables influence the feeling of loneliness in leadership (years of work, position, and academic qualifications), others do not (gender, age, and hours of work per week). We also found a relationship between loneliness and demotivation. It was also proved that the feeling of loneliness affects leadership but does not affect decision making. In short, as theoretical contributions, with this study it was possible to analyze a theme that has not yet been sufficiently explored in the literature and apply it to a specific sector—the banking sector—where this type of study is not common, which makes it even more relevant in terms of the literature. On the other hand, the combination of several factors/variables applied to the feeling of loneliness of the leader in a single study also makes it innovative in the context of the literature.

Regarding practical contributions, this study has far-reaching implications, as it identifies the variables that most influence the feeling of loneliness among leaders, including at the level of their decision-making process and their motivational levels, which will allow organizations to better manage this situation, insofar as they will be able to anticipate the variables that most promote loneliness among leaders and reduce its incidence at that level through measures that can reduce the feeling of loneliness and even the demotivation that it can cause in people who hold leadership positions.

For the banking sector, this study is of considerable relevance, as there is no knowledge of other studies that have been dedicated to this theme in this sector, particularly in the Portuguese context. Another practical contribution is to understand that it is necessary to analyze the relationship between loneliness and leadership by sector, as this study suggests that the results may be specific to the banking sector, as some of them contradict the literature regarding studies carried out in other sectors of the economy.

The limitations of the study and further research directions are as follows.

The first limitation that could be found is in the use of a sample for the study instead of applying it to the whole population of the Central Bank and bank leaders. The decision to use a sample was affected by the size of the universe and the impossibility of isolating the entities that did not provide information for the APB’s Activity Report—2019, and because of the time available for carrying out the study and the number of leaders involved.

The second limitation is associated with the type of sample. Being non-probabilistic, it would not be possible to draw conclusions for the leader population of the Central Bank and banks as a whole, so the results will have to be considered only within the scope of the sample analyzed.

The third limitation is related to the non-fulfillment of quotas for all staff in leadership positions. As stated beneath Table 3, only BP presented the breakdown by quota of the three levels of management from the vertical perspective, so the different categories represented in the sample do not have any points of comparison.

For future studies, we propose to analyze the positive and negative aspects that the feeling of loneliness can imply in leadership and other variables that were not analyzed here, namely the organizational culture and the leadership style, which we believe also have an influence on this level. Applying this theme to other specific sectors can also be important, as well as in contexts more conducive to loneliness, such as the current pandemic context in which we live.