This study aimed to investigate the relationship between AL and trust in the leader (TL) within a precarious business context in South Africa. Moreover, we wanted to determine whether the perceptions of precariousness (in the form of job insecurity) would possibly moderate the relationship between AL and TL.

Although the relationship between AL and TL has been investigated and confirmed by various studies over the years (Agote, Aramburu, & Lines, 2016; Baker, 2020), limited studies are available that describe the relationship between AL, TL and the influence of job insecurity within a precarious business environment in South Africa. Although previous research has found a relationship between AL and trust, the direct moderating role of precariousness has not, to the best of our knowledge, been investigated.

The theoretical contribution made by this study deepens knowledge and understanding of the nature of the relationship between AL, TL, and perceived job insecurity as an element of work-related precariousness. Firstly, this study makes a new contribution to the existing leadership literature by depicting the relationships among constructs within a model, which, to the best of our knowledge, has not been researched before. Secondly, the study contributes to the limited research on AL within extreme uncertain and volatile business settings. Thirdly, this study contributes to the existing literature by testing the constructs in a developing country context within a real-time situation. The contribution is brought about by illustrating that AL may strengthen TL and adaptability to challenging employment conditions.

Luthans and Avolio (2003) explained that AL developed as an important research area in positive psychology. Authentic leaders are defined as individuals who are:

Authentic leadership in organisations is described by Luthans and Avolio (2003):

Authentic leadership consists of four dimensions: the leader’s relational transparency, internalised moral perspective, balanced processing and self-awareness (Gardner, Avolio, Luthans, May, & Walumbwa, 2005). Knowing one’s strengths and weaknesses and the social influence thereof on other individuals apply to self-awareness (Kernis, 2003). Balanced processing relates to a leader’s ability to be unbiased when considering all available information before reaching a final decision (Leroy, Palanski, & Simons, 2012). An internalised moral perspective would influence a leader to act with self-regulation and self-determination rather than to react to situational pressures (Avolio, Walumbwa, & Weber, 2009). Relational transparency points to the leader’s behaviour that demonstrates their authenticity and genuine feelings towards team members (Wei, Li, Zhang, & Liu, 2018). Exhibiting relational transparency encourages mutual trust in organisations (Wei et al., 2018). In support of this view, several studies found a positive relationship between AL and TLs (Clapp-Smith, Vogelgesang, & Avey, 2009; Maximo, Stander, & Coxen, 2019; Wong et al., 2010).

Trust can be defined as an expectation that one individual can rely on another individual’s deeds and promises and that the individual intends to be true to their word (Bligh, 2017). The actions and character of a leader impact the willingness of an employee to trust the leader (Heyns & Rothmann, 2015). In the context of social exchange theory, trust represents an acceptance of risk and uncertainty that the exchange partner may or may not reciprocate the expected behaviour (Whitener, Brodt, Korsgaard, & Werner, 1998). When team members observe open communication, support and cooperation, willingness to sacrifice, confidence, predictability, and fair treatment in their leader’s behaviour, it will enhance trust within the leader-follower relationship (Clapp-Smith et al., 2009). Trust in the leader is defined as the willingness of employees to be vulnerable to their leader’s conduct, which is beyond the employees’ control (Sendjaya & Pekerti, 2010). Although an employee demonstrates the willingness to be vulnerable to the actions and decisions of a leader, there are no guarantees that the outcomes of the trust relationship will favour the employee.

Immediate supervisors who treat their team members fairly and take their needs into consideration when making decisions may enhance employee trust in them (James, 2011). Trust in a leader plays an essential role in inspiring employees to exert greater effort and perform beyond expectations (Schaubroeck, Peng, & Hannah, 2013). Conversely, when leaders are perceived to be untrustworthy by their followers, these followers might spend more time protecting themselves, which can negatively influence their performance (Mayer & Gavin, 2005).

Trust in the leader plays an integral part in the successful implementation of change initiatives within an organisation as it is deemed necessary for getting individuals to work together towards a common goal (Søresen & Hasle, 2009), especially under high levels of perceived uncertainty. The way in which leaders deal with challenging events (e.g. layoffs or downsizing), may directly influence the level of trust that employees have in their leaders (Tourish, Paulsen, Hobman, & Bordia, 2004).

Instability and flexibility of the modern labour market have led to increased feelings of uncertainty, insecurity, and precariousness in the work environment (Urbini, Lo Presti, Chirumbolo, & Callea, 2020). The concept of precarity has its origin in French sociology and economics. Precarity initially referred to a social condition linked to poverty and only later referred to a different form of employment than the familiar permanent employment relationship (Barbier, 2002). The concept of precarity emerged from the early 2000s onwards and is characterised by increasingly insecure employment and generalised uncertainty and insecurity (Lazar & Sanchez, 2019). The consequences of precarious work and related job insecurity were exacerbated by the global economic crisis of 2008. The precariousness in the work environment has increased as countries attempted to respond to weakening financial situations and the increasingly fragile economy (Kalleberg, 2018).

Job insecurity refers to the unintended concern employees have concerning the future continuity of their jobs (Vander Elst, De Witte, & De Cuyper, 2014). The decision to measure precariousness because of job insecurity was informed by the statement of Fullerton, McCollum, Dixon and Anderson (2020) that job insecurity is a crucial element of precarious work and precariousness. Change and optimisation in the workplace is synonymous with uncertainty about the future content of an employee’s job (qualitative job insecurity) as well as the risk of job loss (quantitative job insecurity) (Urbini et al., 2020). Greenhalgh and Rosenblatt (2010) maintained that perceived job insecurity can have a destructive impact on employees and organisations. Additionally, job insecurity is linked to reduced levels of job satisfaction, organisational commitment, job performance, poor employee well-being (Lee, Huang, & Ashford, 2018), and decreased optimism about the future (Li, Li, Fay, & Frese, 2019).

Brower, Lester, Korsgaard and Dineen (2009) pointed out that trust is a crucial element in the supervisor-follower relationship. When authentic leaders exhibit behaviour in accordance with their personal values, they are able to build credibility, respect and trust with their team members through the encouragement of different perspectives and building synergetic relationships with them (Avolio et al., 2004). Kernis (2003) posited that the authentic leader’s relational authenticity involves striving for achieving openness and truthfulness in their relationship with their followers and colleagues, which may result in elevated levels of trust. Several studies found that AL significantly impacts trust in leaders (Levesque-Côté, Fernet, Austin, & Morin, 2018; Maximo et al., 2019).

Clapp-Smith et al. (2009) maintained that trust can be described as the vulnerability of one person to another person’s actions while being convinced that the actions of the other person will be carried out with good intentions. The willing exchange of actions between individuals also forms part of a trusting relationship. When the trusting employee believes that exploitation is unlikely, there will be a willingness to risk vulnerability, and a display of trusting behaviour (Eggers, 2011). Agote et al. (2016) stated that TL will influence followers’ attitudes and behaviour. Trust in the leader was positively related to job performance, positive attitudes towards jobs and organisations and organisational citizenship behaviour (Colquitt, Scott, & LePine, 2007). In contrast, employees who do not trust their leader are less likely to do more than what is expected of them or remain in the relationship because of the lack of a sense of obligation and low positive expectations (Brower et al., 2009). Additionally, because of the behaviour of untrustworthy leaders, the performance of employees may be compromised as a result of their concern of being treated unfairly (Mayer & Gavin, 2005).

As a result of the dynamic business environment in which employees are functioning presently, they frequently experience job insecurity since they feel that the future of their employment might be at risk (Richter & Näswall, 2019). Job insecurity has consequently become a familiar stressor in working life (Lee et al., 2018). Jiang and Lavaysse (2018) maintained that living in constant uncertainty and fear that their working life may be subjected to substantial changes at any time may adversely affect employees as well as organisations. Prior job insecurity-related research suggests that feeling uncertain about the future of one’s employment can be associated with several short (Cheng & Chan, 2008) and long-term negative consequences (De Witte et al., 2016). According to Arnold and Staffelbach (2012), trust in the organisation has been shown to negatively relate to job insecurity and perceived uncertainty during organisational restructuring and optimisation. Moreover, Huie, Cassaberry and Rivera (2020) argued that job insecurity may result in the loss of trust, jealousy and destructive competitiveness in a work context. This can potentially be viewed as the deterioration of the employee’s relationship with the employer and may result in reduced job satisfaction and employee wellbeing (Richter & Näswall, 2019). Mistrust forms the base of unsuccessful business relationships as it directly influences knowledge sharing that harms job performance and economic progress (Huie et al., 2020). Additionally, the future retention of employees with low levels of trust is likely to be negatively affected (Hopkins & Weathington, 2006).

Bases on the above, the research questions of this study were formulated as follows:

The researchers utilised a quantitative approach, together with a cross-sectional survey design during this study.

The study population comprised of employees from three managerial levels consisting of senior, middle, and junior managers working at the different plant sites of a South African manufacturing organisation. The questionnaires were distributed to 570 employees, and 314 completed questionnaires were received back, representing a response rate of 55%. Data collection was conducted via stratified random sampling.

Table 1 depicts the participants’ characteristics: 41.7% of the respondents indicated that they were in the 51–60 age bracket, 22.2% were employed at manager level, 30.5% of the participants had 31–40 years of service, 87.5% had up to 20 years of experience in their current position. More than half of the participants stated that they are employed at Vanderbijlpark works in Gauteng (51.9%), and 17.8% at the Newcastle production facility in KwaZulu-Natal.

After having filled in the biographical questionnaire, participants were requested to complete instruments for measuring AL, precarity, and TL. The Authentic Leadership Inventory (ALI) (Neider & Schriesheim, 2011) was used to measure the followers’ perceptions of their direct leader’s AL characteristics. The ALI consists of four dimensions (self-awareness, internal moral perspective, balanced processing, and relational transparency), measured by means of 14 items. Example items include: ‘My leader asks for ideas that challenge his or her core beliefs’ and ‘My leader objectively analyses relevant data before making a decision’. A five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree) was applied to score the items. Cronbach’s alpha coefficients between 0.74 and 0.90 are acceptable (Men & Stacks, 2014). The reliability of ALI was also tested in South Africa (α = 0.93; Stander, De Beer, & Stander, 2015).

One scale of the Workplace Trust Survey (WTS) (Ferres & Travaglione, 2003), consisting of 9 items, measured TL. Mentioned items were scored by applying a seven-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). Examples of the items include: ‘I feel that my supervisor listens to what I have to say’ and ‘I believe that my supervisor follows through words with action’ (Ferres & Travaglione, 2003). Previous studies in both a South African and Australian context have reported Cronbach alpha reliability coefficients of between 0.90 and 0.97 (Ferres & Travaglione, 2003).

Job insecurity was measured by applying the Job Insecurity Scale (JIS) – a scale consisting of four items initially developed by De Witte (2000). This scale is a global job insecurity measure, and it includes items that refer to the threat or possibility of losing a job, as well the anxiety associated with job loss. Examples of the items include: ‘I feel insecure about the future of my job’, and ‘I think I might lose my job in the near future’. Respondents were requested to rate these items on a five-point Likert type scale, ranging from 1 (strongly disagree) to 5 (strongly agree). The Cronbach alpha reliability coefficient was recorded as 0.77 (Vander Elst et al., 2014).

The Chief Executive Officer granted permission for the study to be performed at the participating manufacturing company. A higher education institution gave scientific and ethical clearance for this study. Before participating in the study, participants completed a consent form. An information brochure describing the purpose of the research and ethical considerations accompanied the questionnaire.

Ethical approval was granted by the Economic and Management Sciences Research Ethics Committee (EMS-REC) at North-West University, reference number: NWU- 00609-20-A4.

Mplus 8.4 was utilised to analyse the data related to this study (Muthén & Muthén, 1998–2019). Confirmatory factor analyses (CFA) were performed to determine the validity of the instrument and identify the best-fitting measurement model. Structural equation modelling (SEM) was applied to assess how the data fitted various structural models and to test the research model. The items of all questionnaires were dealt with as continuous. The mean-adjusted Maximum Likelihood (MLM) estimator, which is robust to the non-normality of data (Wang & Wang, 2020), was used because of all values being recorded.

The best-fit model was identified using the Chi-square values, which were calculated to evaluate absolute fit. The incremental fit was determined by applying the Comparative Fit Index (CFI) and the Tucker-Lewis Index (TLI). Wang and Wang (2020) deemed 0.90 an appropriate cut-off value for these two fit indices. The Root Mean Square Error of Approximation (RMSEA) and Standardised Root Mean Square Residual (SRMR) were also evaluated. The RMSEA values less than 0.08 represent an acceptable fit (Kline, 2016). Values lower than 0.05 are accepted as an indication of a good fit between the data and the model for the SRMR indicator (Wang & Wang, 2020). The SRMR values closer to 0.08 indicate an acceptable fit between the model and the data (Wang & Wang, 2020). The comparative fit indices, Akaike information criterion (AIC) and the Bayes information criterion (BIC) were used to compare the different measurement and structural models. Kline (2016) mentioned that the model with the lowest value is preferred. Composite reliability coefficients with a minimum value of 0.70 were employed to determine measuring instrument reliability (Wang & Wang, 2020). The internal consistency of constructs was established by utilising composite reliability (Raykov, 2009). A simple moderator analysis was performed during this study by using PROCESS (Hayes, 2018).

To evaluate the possible relationships between the latent variables, a three-factor measurement model (Model 1) was specified and tested for fit to the observed data. Additionally, one competing model was identified and tested to confirm the fit of the preferred model.

Model 1 was constructed in line with what theory proposes: AL (measured as second-order construct), TL (measured by nine directly observed variables) and job insecurity (measured utilising four directly observed variables).

Model 2 was specified similar to Model 1 with the exception that AL was specified as a first-order latent variable measured employing only 14 directly observed variables.

The results for model 1 reflected that the Chi-square test was significant (p < 0.001), with a Chi-square (χ²) value of 506.189 and a degrees of freedom (df) value of 317, suggesting that a perfect fit to the data was not achievable. As a result of the oversensitivity of the measure of fit, Hancock and Mueller (2010) recommended that researchers consider reporting multiple fit indices. The alternative fit indices indicated that an approximate fit to the data was attainable: Both CFI and TLI were above the cut-off value of 0.90, RMSEA indicated a good fit (< 0.05, p > 0.05). The SRMR value of less than 0.08 also indicated a good fit (Wang & Wang, 2020).

Table 2 presents the goodness-of-fit statistics for the competing measurement models described above. Table 2 indicates that Model 1 has the best statistical fit of the two models under review. The initial measurement model obtained an χ² value of 506.189 and a df value of 317. The fit indices for CFI and TLI were also acceptable (> 0.90), as was the model fit for the RMSEA (< 0.05). The SRMR value for Model 1 was acceptable (< 0.08) and was equal to the value of the alternative model. Lastly, the AIC and BIC fit indices were used to compare models to identify the best fit (lowest value). Model 1 had the lowest AIC value, and Model 2 had the lowest BIC value.

As part of the Model 2 results, the chi-square results were once again significant (χ² = 518.642, df = 321, p < 0.0001). The alternative fit indices represented by the TLI and CFI values being higher than 0.90 were acceptable. Lastly, both the RMSEA (p > 0.05) and SRMR (< 0.05) values were also acceptable.

Although acceptable comparative fit indices were indicated for both competing models, the CFI (0.959) and TLI (0.954) values and the RMSEA confidence intervals of Model 1 pointed to this model being the best fit to the data. Additionally, with the lowest AIC value of 19 783.448, it is suggested that Model 1 had the best data fit. However, the BIC value pointed to Model 2 as being the best-fitting model. When evaluating all the fit indices in conjunction, Model 1, which was also the model that most closely represented theory, was chosen as the preferred model.

The structural model was tested based on the chosen measurement model and resulted in similar fit statistics as was obtained from the measurement model. The results in Table 3 indicate that the Raykov’s rho coefficients were well above the minimum threshold with values above 0.9 and are thus considered highly reliable. The Raykov’s rho coefficients of all the measuring instruments ranged from 0.709 to 0.985 and were deemed acceptable. Furthermore, Table 3 provides the correlation coefficients of the study variables. The AL was found to be significantly related to TL (r = 0.820), and AL and job insecurity are inversely related (r = -0.137; p < 0.05).

Table 4 depicts the path coefficients of the structural model (Model 3) as estimated by Mplus 8.4 (Muthén & Muthén, 1998–2019). Table 4 also portrays the standard path coefficients for AL and job insecurity as independent variables and trust as a dependent variable.

As can be deducted from Table 4, AL is a significant predictor of TL (β = 0.814; p < 0.01). However, job insecurity was not significantly associated with TL (β = -0.046; p > 0.05). Authentic leadership explains 45.43% of the variance in TL.

To test whether the relationship between AL (independent variable) and TL (dependent variable) depends on the extent to which employees experience job insecurity (moderator), a simple moderator analysis was performed using PROCESS (Hayes, 2018). The interaction between AL and job insecurity was not statistically significant (β = -0.036, 95% CI, LCI = -0.242, UCI = 0.170). Therefore job insecurity cannot be considered a moderator of the relationship between AL and TL.

The objectives of this study were to determine whether the AL style can predict TL among the selected sample of employees from various functions in a manufacturing organisation in South Africa. More specifically, the study was conducted to gain knowledge and understanding of how AL can potentially enhance TL despite the turbulent economic and business conditions experienced in the target organisation. The potential moderating effect of job insecurity on the relationship between AL and TL was also tested:

The results of this study confirmed the first part of our research question by indicating that AL directly and positively influences TL – higher AL lead to increased TL. This result also confirms the outcome of previous studies on AL that found that AL is a positive predictor of TL (Maximo et al., 2019; Qiu, Alizadeh, Dooley, & Zhang, 2019).

Authentic leaders are able to establish engaging and constructive organisational conditions because of them being profoundly aware of their values and beliefs, and being genuine, dependable and trustworthy (Avolio and Gardner 2005). Gardner et al. (2005) put forth that authentic leaders can develop trust in team members through their ability to be conscious of their capabilities and shortcomings and disclosing their real self to others while being sensitive to the impact their actions may have on others. Additionally, high self-awareness levels have been associated with positive follower attitude, behaviour, and performance (Avolio & Gardner, 2005) and, in this study, trust. One of the prerequisites of trust is to possess knowledge of and insight into the other person. The leader who is willing to allow his employees to get to know them has a better chance to instil trust in their leadership. Relational transparency is a component of AL that relates to the open sharing of information and exhibiting true emotions to others, resulting in elevated trust between leaders and employees (Iqbal et al., 2020). The sharing of information and emotions will create a psychologically safe environment where the employee will become more open about their feelings and needs. By feeling comfortable, employees will easily share positive experiences and concerns, leading to healthy trustful relationships. When employees experience that there are no hidden agendas, their TL will increase. Trust is likely to influence individual cooperation levels within a relationship (Norman, Avolio, & Luthens, 2010). Such behaviour can result in developing positive follower qualities such as optimism and confidence in the work situation. Authentic leaders may also instil enhanced admiration in their followers by demonstrating their internalised moral perspective through applying high moral standards and ethical behaviours (Ilies et al., 2005) rather than being influenced by external pressures and opinions. These leaders behave transparently towards their followers and lead by example while guided by characteristics such as honesty, fairness, and accountability. When employees perceive that they are being treated fairly and morally by their supervisor, they might trust their leader while being more engaged at work (Cropanzano & Mitchell, 2005; Norman, 2006). If followers perceive that they are being treated unfairly, they will trust that leader less.

Awareness of the leader’s values, and even more critical, alignment between the leader and employee’s values, creates a climate in which there are fewer reasons not to mistrust the leader. Lastly, an authentic leader’s ability to consider all available information in an unbiased and balanced manner before reaching a final decision (Penger & Černe, 2014) can reinforce the sense of confidence employees have in their leader’s decisions. This approach may result in actions that are well-planned and thought through. Although the decision-making processes that authentic leaders follow might be a bit more time-consuming, the methodology they follow might be beneficial in new and uncertain situations. A balanced approach will reinforce the employees’ perceptions that the leader acts in the organisation’s best interest and its employees, without bias towards specific individuals.

Flavian, Guinaliu and Jordan (2019) mentioned that when employees trust their leader, they are likely to have positive expectations about their intentions and behaviours since they have confidence that their leader will have their best interest at heart. When high levels of TL exist, followers are more likely to exhibit a readiness to face vulnerability during times of uncertainty as a result of their positive perception of the supervisor’s intentions and behaviour (Zhang & Zhou, 2014). The findings of a study conducted by Bligh (2017) underlines the importance of this result as it emphasises the vital role TL can play in improving organisational performance. The trust between leaders, and team members notably influenced employee-related outcomes such as turnover intentions and engagement through the reduction of workplace uncertainty and strengthening the social exchange relationship (Skiba & Wildman, 2019):

In contrast to expectations, no evidence could be found that job insecurity will moderate the direct and indirect associations between AL and TL. This implies that a variation in job insecurity levels will have no influence on the magnitude or direction of the relation linking AL and TL.

When employees regard their leader as trustworthy, it is plausible that they will feel safer and more optimistic about the leader making crucial decisions (Jiang & Probst, 2019) despite the lack of guarantees. One can assume that authentic leaders are able to demonstrate their concern for employee wellbeing through their lack of bias while considering all available and relevant information before reaching a final decision (Leroy et al., 2012), even during restructuring and workforce reduction initiatives.

Drawing from the Social Exchange Theory (Blau, 1964), it might be said that when employees trust their leader, these employees may deem it a relationship of social exchange while being convinced that the leader will reciprocate the trusting behaviour. The inherent principle of reciprocity suggests that an individual’s behaviour depends on the give and take relationship between one person (leader) and another (follower). It might thus be possible that because employees trust their leader and are dedicated to their work, they expect that the supervisor will ‘return the favour’ by considering their wellbeing when making decisions during an organisational optimisation and workforce reduction process. Moreover, because team members have trust in their leaders and their decision-making ability, they might have faith that the leader will make a decision that will have their best interest at heart. The unfortunate reality that the manufacturing organisation where the study was conducted had to undergo numerous restructuring and down-sizing initiatives over the past number of years might also have resulted in employees exhibiting a certain level of reduced fear for the unknown and willingness to face the risks of the potential loss or change in the content of their jobs.

This study addressed the identified research gap by firstly confirming the relationship between a positive leadership style, such as AL and TL. Secondly, the study added theoretical value by suggesting that job insecurity as a form of business-related precariousness does not have a moderating effect on the relationship between AL and TL.

Leaders can be successfully developed provided the leadership development intervention used has been proven effective and that both the leader and the organisation are developmentally ready to embrace leadership development (Avolio & Hannah, 2020). Organisations should thus consider including AL elements in their selection, training, and rating activities to benefit from AL. As part of their succession planning and leadership development processes, organisations might want to consider the inclusion of AL training in order to prepare future leaders. Organisations should afford leaders on all levels the opportunity to develop their AL capabilities. Aside from earning the trust of their followers, leaders must also be educated to increase perceptions of their authenticity by behaving in line with their beliefs or by ‘practising what they preach’ and being open and honest towards their team members as well as themselves.

Various study limitations should be noted when interpreting the study results. Firstly, using a cross-sectional design limited the ability to make causal inferences between research variables. Employing self-report questionnaires as the only source of information for the study may result in common method variance. On account of the research being conducted at only one manufacturing organisation in the steel manufacturing industry in South Africa, the extrapolation of results to different environments should be made with caution.

Regardless of the mentioned research limitations, the following proposals could be considered in future studies. The fact that the study was conducted by only involving managers leaves the question as to whether involving employees on lower levels (e.g. less educated, blue-collar) would have resulted in the same outcome. New insights into the potential influence of a precarious work environment could also be gained by conducting a longitudinal or mixed-methods study. Future studies might want to include trust in the organisation, since the possibility exists that job insecurity might have a closer relation to trust in the organisation than to AL.

The results of this study underscore that promoting AL is a promising pathway for improved follower trust levels in their leader, which may ultimately benefit individual and organisational performance. The benefit of promoting trust within an organisation is highlighted in a study by Koohang, Paliszkiewicz and Goluchowski (2017), who found that trust is a critical factor in social and economic relationships as well as a significant contributing factor to organisational performance. The non-moderating effect of job insecurity may imply that even employees with high job insecurity will have trust in their leader. The researchers are of the opinion that organisations will benefit from developing authentic leaders.