This research investigated if succession planning was the mediating variable between self-leadership and turnover intention in a state-owned enterprise. The objective of the study was to determine if there was a significant relationship between succession planning, self-leadership and turnover intention at a state-owned enterprise.

Researchers define succession planning as an organisation’s method of continual recognition, appraisal and nurturing of talent in key roles so that those individuals are in place to ensure leadership stability over time (Barton, 2019; Olufemi, 2021). Succession planning is also understood as a deliberate and organised initiative for an institution to keep its leadership, preserve and advance its intellectual resources and encourage staff development (Budhiraja & Pathak, 2018). Succession planning practices in public institutions have been investigated to clarify their relationships with turnover intention and personal leadership (Ahmad et al., 2020). Karikari et al. (2020) researched successful succession planning models being used in the public sector, with an emphasis on the public universities in Ghana. The study concluded that the success could be ascribed to the understanding that universities had a duty to nurture academics to become experts in their fields so that they could occupy leadership roles.

Three succession planning approaches have emerged as influential in fostering a systematic and constructive process to groom, identify and recruit leaders to replace those leaving the organisation (Rosenthall et al., 2018). Strategic leader development, emergency succession planning and departure-defined succession planning are the three succession approaches (Barton, 2019).

Strategic leader development is described as an ongoing internal practice shaped by the goal of defining an agency’s strategic vision, determining the leadership and management skills needed to realise that vision, and recruiting and retaining talented individuals capable of improving those skills (Rosenthall et al., 2018). Strategic leader development succession planning entails developing career growth programmes to create a team of skilled people within the organisation that will fulfil the organisation’s potential leadership needs (Karikari et al., 2020). According to (Ahmad et al., 2020), organisations are exposed to a plethora of information on the internet or in archives, scholarly journals or books, explaining how knowledge economy companies craft and implement plans to stay competitive.

Emergency succession planning is a distinct approach to succession planning (Ahmad et al., 2020). The primary aim is to ready an organisation for the unexpected exit of key executives (Gordon & Overbey, 2018). Because succession planning can be a delicate topic, the process can be referred to as ‘emergency backup planning’ or ‘emergency leadership planning’ instead of ‘emergency succession planning’ (Gordon & Overbey, 2018, p. 60). This will help secure key individuals’ buy-in and challenge their perception of the mechanism as a threat to their positions (Ahmad & Keerio, 2020).

Departure-defined succession planning seeks to develop leadership capabilities in an organisation so that it can reduce its reliance on the incumbent’s talents, charm and relationships and maintain equilibrium without him or her (Chang & Busser, 2020). This strategy, if correctly planned, would ensure that enough time is allotted for either handover to the newly identified successor (Ahmad et al., 2020) or locating and training the newly identified successor in the skills required to carry out his or her responsibilities. Organisations may benefit from departure-defined succession planning in identifying successors (Gordon & Overbey, 2018) so that they do not face a disaster when a core member of staff exits. The studies that have explored succession planning approaches highlight the importance of succession planning in organisations (Ahmad & Keerio, 2020). However, they focus on private organisations which are run differently from the public organisations which are the subject of this study.

Employee turnover has posed a challenge to businesses. The success of the corporation is primarily dependent on those who work there. The happier the workers, the better for the company, as none of them is likely to leave anytime soon (Oh & Kim, 2019). Companies all over the world are struggling to retain workers, especially those who occupy key positions in the company. According to Belete (2018), one of the widely researched challenges of successful organisations is maintaining valuable staff. Today’s businesses still attempt to hold on to staff members considered important assets because they are resources for the company’s activities (Oh & Kim, 2019). This is of concern to all companies because turnover leads to substantial costs associated with the recruiting process as well as a lack of productivity (Xu et al., 2022).

The turnover intention has been described as an employee’s voluntary, conscious and reasonably strong recurring intentions to leave an organisation (Oh & Kim, 2019). These repeated feelings and plans to leave the organisation initiate a cognitive process that is compounded by influences arising both inside and beyond the organisation, resulting in the ultimate decision to leave the organisation (Oh & Kim, 2019). The following considerations are used in this withdrawal cognition process: (1) thinking about leaving the company, followed by (2) a successful quest for another job elsewhere and (3) final intention to resign or depart (Fyn et al., 2019). Turnover intention is generally recognised as the strongest indicator of workforce turnover (Oh & Kim, 2019). This purpose is always difficult for organisations to discover until the employee shares it with members of the organisation.

The causes of turnover intention in organisations have been subject to intense scrutiny. Windon, Cochran, Scheer and Rodriguez (2019) investigated factors affecting turnover intention among extension programme assistants at Ohio State University. They found a substantial correlation between employees’ age, years of service, work satisfaction, supervisor satisfaction, organisational loyalty and plans to leave.

The relationship between job satisfaction and the desire to quit is influenced by organisational engagement (Oh & Kim, 2019). Increased workplace satisfaction has an impact on workforce effectiveness and work efficiency, and it helps companies maintain their workers (Fyn et al., 2019). Oh and Kim’s study (2019) made an important contribution to identifying the connection between organisational engagement and intention to leave. However, it involved a university setting in a particular geographical location. The organisational culture of American state institutions varies greatly from that in South Africa. As a consequence, the current study offers a context-based analysis of the phenomenon. The contribution of this study lies in the fact that the relationship between the three variables was tested in a South African context.

Self-leadership is the deliberate influence of thoughts, feelings and behaviour on goals (Inam et al., 2021). In a professional context, self-leadership requires a transformation in behavioural and cognitive habits that results in interventions intended to increase sense of satisfaction, desire to work smarter and accomplish more, and ability to think constructively about problems that can contribute to the accomplishment of goals set for oneself (Byrant, 2021). The philosophy of self-leadership comprises two dominant fields: self-managing teams and inspiring leadership. According to Goodridge (2019), the essence of self-leadership is self-exploration, self-recognition, self-control and self-growth. These four elements do not work in isolation, but rather complement one another (Goodridge, 2019). To be a good self-leader, one must cultivate all four. The first foundation is self-awareness (Inam et al., 2021). This means that being a great leader requires first considering oneself. According to Goodridge (2019), all people have ideals and principles that characterise them. As people’s approach to life and relationships is driven by these principles they should be explored, together with what people are passionate about. This introspection will facilitate discovery of and adherence to principles and ideals – the true self.

Self-acceptance is the second most significant pillar of self-leadership. Self-acceptance is described by Goodridge (2019) as being completely frank with oneself and embracing it without self-criticism or self-sabotage. Self-acceptance is crucial for a leader because when things are not going well, it helps leaders first to realise and then to accept their part in the state of affairs (Thompson, 2018).

Self-management is the third and most significant pillar of self-leadership. Thompson (2018) suggests that leaders who are capable of self-management are more creative, focused and capable of functioning efficiently by themselves. Goodridge (2019) defines self-management as keeping yourself accountable and ensuring that you use your time and money efficiently. Boonyarit (2023) argues that a leader who can self-manage knows that the most precious tools are time and energy and that striking a reasonable balance between the two will produce the desired self-leadership. Self-management is correlated with a variety of essential characteristics and behaviours.

Finally, self-development is an integral part of self-leadership. According to Goodridge (2019), self-growth occurs when one is frank about oneself, but without self-criticism. Self-development is about knowing what one is doing, what works and what does not, and how one can make meaningful improvements (Stewart et al., 2019). As a result, the leader gains a sense of transparency and responsibility for doing the right thing. For self-growth to occur, the leader must be open to feedback and dedicated to self-growth.

Pila et al. (2016) describe self-leadership as the process by which people manage their behaviours, gain influence and lead themselves via the use of certain behavioural and cognitive tactics (Thompson, 2018). There is a link between self-leadership and the desire to quit.

Existing research has looked into the link between self-leading leadership and employee intention to resign. Tetteh and Asumeng (2020), for example, studied the influence of affective commitment on self-leadership and employee turnover intention. According to the findings of the study, affective commitment completely mediates the association between self-leadership and employee turnover intention. This suggests that successful commitment reduces an employee’s desire to resign, while increasing trust and readiness to follow the organisation’s leaders’ vision, philosophy and ideology (Pita & Dhurup, 2019).

From the literature on self-leadership and turnover intention, it can be deduced that employees should continuously work towards self-leadership as a way to generate job satisfaction, growth and fulfilment of goals.

The relationship between self-leadership and turnover intention is highlighted by the following hypothesis:

Research indicates that succession planning mediates self-leadership and turnover intention (Okwakpam, 2019; Stewart et al., 2019). Boonyarit (2023) confirms the relationship between self-leadership and turnover intention and recommends that organisations come up with strategies to promote self-leadership and succession planning as a particular point of departure to address problems related to turnover intention. Stewart et al. (2019) highlight that self-leadership can be enhanced by celebrating small wins as a method of achieving even greater success. Organisations must know how to balance leadership development and succession planning as this can help to reduce turnover intention (Pita & Dhurup, 2019).

A performance management system is often described as a structured approach to measuring an organisation’s employees’ performance (Winarto & Chalidyanto, 2020). During this phase, the organisation aligns its tasks, goals and targets with existing resources (such as workforce, materials and so on), processes and priorities (Chandra & Saraswathi, 2018). The approach to desirable performance evaluation is the collection and application of objective performance data to achieve competitive organisational objectives (Chandra & Saraswathi, 2018).

Leroy et al. (2018) highlight that performance management systems mediate self-leadership and turnover intention. The lack of performance management systems may result in low employee motivation and push employees to think about leaving the organisation (Nagel & Green, 2021).

Performance management systems help the organisation to identify challenges such as internal and external forces which impede the capacity of employees to perform (Okwakpam, 2019). These forces can be mitigated by implementing succession planning and nurturing employees who are loyal to the organisation and who can provide a competitive advantage to the organisation (Oduwusi, 2018). Employee development programmes are a means of achieving this. If employees are happy that their organisation is concerned about their growth, they are likely to have a higher level of satisfaction in their jobs. They also tend to be committed to the company as they believe that the company offers them a bright future and will take good care of their contributions. Luedi (2022) argues that employees who show high levels of commitment to their jobs are more likely to be successful and to not consider leaving the organisation.

The mediation by succession planning (performance management system) of self-leadership and turnover intention is highlighted by the following hypothesis:

Supervisor support is described as how much leaders respect their employees’ efforts and care for their well-being (Budhiraja & Pathak, 2018). A leader who receives proper guidance from his or her bosses listens to, values and cares for the opinions of his or her employees (Abid et al., 2019). Supervisor support (SVS) from the organisation’s management has a positive influence on employees’ affective commitment (Ahmad et al., 2020). Affective commitment to the supervisor, which denotes a solid bond between workers and their managers, is very likely to have a significant impact on employees’ work-related attitudes and behaviours and reduce intention to leave the organisation (Luedi, 2022).

The concept of fairness is often related to supervisor support. This definition applies to work effort–reward equity (Eisenbeger et al., 2016). The concept relates to how employees measure the worth of their contributions against the incentives and benefits they earn. Job efforts include a variety of investments, like the time, resources, expertise, experience and intelligence required for the completion of job tasks. Pita and Dhurup (2019) highlight attentive supervisor assistance as a way to increase employee job satisfaction. By implication, supervisor support entails offering employees opportunities for development that foster self-leadership. In turn, employees who feel the organisation supports them are less bound to think of leaving the organisation. Winarto and Chalidyanto (2020) investigated the importance of supervisor support in forecasting employee job satisfaction in a private hospital in Indonesia. They found that workers who had a high degree of work satisfaction were in good physical and mental health. Employees perceived that their jobs provided them with material, meaning and growth opportunities; hence, they felt attached to the organisation (Winarto & Chalidyanto, 2020).

The mediation by succession planning (supervisor support), self-leadership and turnover intention is highlighted by the following hypothesis:

Perceived organisational support (POS) is defined as the general belief among employees that their employer values their contribution and cares about their well-being. Research indicates that there is a positive relationship between organisational support and self-leadership and turnover intention (Li et al., 2022). These findings are supported by Chang and Busser (2020), who established that there was a strong link between POS and employee job satisfaction.

Satisfied employees have lower turnover intention than disgruntled ones. In addition, organisational support facilitates self-leadership as employees are inspired when they see the organisation expressing interest in them, paying attention to them, honouring them and supplying them with resources for promotion within the organisation (Abid et al., 2019). In turn, they feel obliged to improve their behaviour and develop positive attitudes to work and the organisation (Suárez-Albanchez et al., 2022). Employees will be motivated to help the company achieve its goals if they feel they have positive POS. Perceived organisational support will result in an improvement in in-role and extra-role average efficiency and a reduction in depression and habits such as absenteeism and turnover (Li et al., 2022). Perceived organisational support affects employees’ long-term dedication (Li et al., 2022) and elicits well-known affective reactions such as process pleasure and good temper (Chang & Busser, 2020).

The mediation by succession planning (organisational support), self-leadership and turnover intention is highlighted by the following hypothesis:

Figure 1 is a graphic representation of the hypotheses.

Positivism was used as the research paradigm. A quantitative approach was used to evaluate the factors contributing to turnover intention and self-leadership while determining if succession planning was the mediating variable between self-leadership and turnover intention in a state-owned enterprise. Quantitative research allows for a broader study, involving a greater number of subjects and enhancing the generalisation of the results; there is also greater objectivity and accuracy (Aspers & Corte, 2019). The research was done at a state-owned company in the freight logistics sector of one of the specialist units in South Africa. The company had five operating divisions and specialist units.

The study targeted 545 permanent employees classified as executives, top management, senior management, professionally qualified employees, experienced specialists, middle management, skilled technicians, academically qualified workers, junior management, supervisors, foremen and superintendents within a state-owned company in the freight logistics sector of one of the specialist units. Purposive sampling was used in this investigation. The strategy allowed the researcher to choose a sample based on personal assessment (Benoot et al., 2016). The sample size was derived from one of the operating divisions dealing with infrastructure development. The company had five operating divisions and specialist units. Furthermore, the study only focused on C-F levels in one of the special units. The categories of managers were senior management, management, industry experts, professionals, specialists, middle management, skilled technicians, employees with an academic qualification, employees in junior management positions, supervisors, foremen and superintendents.

The questionnaires were distributed to 300 of the total population of 545. A basic random sample strategy was used in this investigation.

Self-administered questionnaires were used to collect data from participants. In this study, the data gathering instrument was laid out as follows:

The following questionnaires were used to measure the relationship between the three variables in this study:

Part A asked for biographical information in the form of a choice question, but Parts B through D used a Likert scale to express assertions. A 6-point Likert scale was employed, with 1 indicating strongly disagree, 2 indicating disagree, 3 indicating somewhat disagree, 4 indicating slightly agree, 5 indicating agree and 6 indicating completely agree.

In addition, self-administered questionnaires are cost-efficient as the researcher does not have to hire surveyors and they safeguard the anonymity and privacy of respondents who in turn feel freer to give honest responses.

Descriptive statistics, exploratory factor analysis and inferential statistics in the form of Spearman’s correlation coefficients and multiple regression analysis were used to analyse the data. For the analysis, SPSS 24 was utilised. This software was chosen because it provides an efficient and organised way to manage large and complex data sets and perform advanced statistical analysis, making it an essential tool for measuring the relationships between variables.

The ethical tenets relevant to conducting research were observed in this study. They were:

The questionnaires were distributed to 300 of the total population of 545. Two hundred and twenty-seven questionnaires were completed and returned, which represents a response rate of 78.67%. Of the 227 respondents, male respondents made up a bigger proportion (59.5%) than female respondents (40.5%). This is because the company focuses on capital projects for engineering, a field historically dominated by males in the workforce. The age group 26–35 years had the largest percentage of participants (41.4%). Those aged 36–45 years made up 35.7% of the sample, while 40 respondents (17.6%) were between the ages of 46 and 55 years, followed by those aged 56–65 years (4.0%); those aged 25 and under made up the lowest number (1.3%).

Most participants held positions as skilled technicians, academically qualified employees, junior management, supervisors, foremen and superintendents (F level) (61.2%). This was followed by those with professionally qualified, experienced experts and middle management (E level) roles (22.5%). The senior managers (D level) constituted 11.9% and several participants held top management (B&C level) positions (4.4%). This shows that the vast majority of employees were competent. Managers were targeted to complete the questionnaire due to the sample frame (levels A–F).

To study the grouping of the items and their connection to original theoretical scales and to assess concept validity, factor analysis was used. According to Saunders et al. (2009), construct validity assesses whether a measuring tool measures the constructs that it is designed to measure. It is critical for determining a method’s overall validity.

A principal axis factor analysis with direct Oblimin rotation was used with 0.4 cut-off loadings. The Kaiser-Meyer-Olkin measure of sampling adequacy (KMO) can be used to determine if data are adequate for factor analysis (Kaiser, 1970). Kaiser (1970, p. 402) states that ‘the KMO should be 0.60 or higher to proceed with factor analysis’. Kaiser (1970, p. 402) proposes a 0.4 cut-off value and a preferred value of 0.8 or above. According to Field (2009, p. 660), Bartlett’s sphericity test should be significant. The KMO was first conducted to ensure sample adequacy. The sample adequacy measure for succession planning was 0.920, and Bartlett’s test of sphericity was significant (p = 0.000), indicating sampling adequacy. Furthermore, the KMO sampling adequacy metric for self-leadership was 0.819 and Bartlett’s test of sphericity was significant at (p = 0.000), indicating sampling adequacy. It was high enough in all three situations to proceed with the factor analysis. Several factor solutions for succession planning were thus investigated, and the best factor solution was found to be a three-factor solution, explaining a total of 55%:

The KMO value was 0.632, which is an acceptable level, and only 1 factor was suggested by the Kaiser criterion, explaining 72.280% of the variance. The turnover intention was therefore retained as a unidimensional scale.

The KMO measure of sampling adequacy for self-leadership was 0.819, which is acceptable, and Bartlett’s test of sphericity was significant at (=0.000), which indicates sampling adequacy for self-leadership. A two-factor solution was the result of the factor analysis:

Cronbach’s alpha was used as a measure of internal consistency to assess variables. A Cronbach’s alpha of 0.70 or above suggests adequate dependability (Bryman & Bell, 2007). All three factors of succession planning showed high reliability of 0.70, which point to acceptable reliability. Inter-item correlations all fell within the guideline of 0.1 to 0.5.

A reliability test was done using Cronbach’s alpha to assess the dependability of the survey. According to Roller and Lavrakas (2015), the dependability of a study is proved when a research instrument used for data collection can offer similar findings when employed frequently under similar settings across time.

The reliability of the current study’s questionnaire was calculated using Cronbach’s alpha. Cronbach’s alpha is a measure of reliability that measures the internal consistency or median correlation of items in a survey instrument. Middleton (2016) points out that Cronbach’s alpha has a range of 0 to 1, with 0 being the least internal consistency and 1 representing the most internal consistency. A Cronbach’s alpha of 0.7 or higher indicates strong reliability. The researchers confirmed the content and face validity of the research instrument by pilot-testing it in 17 distinct positions encompassing all areas in the specialist unit where the study was conducted. Before use, this was done to verify that it was aligned with what it was designed to measure and to determine if the questions were understood and interpreted efficaciously and in the manner that they were expected to be interpreted. Roller and Lavrakas (2015) emphasise the importance of instrument validation in determining if an instrument measures what it is meant to assess.

In terms of the correlation between succession planning and self-leadership, it was found that behavioural awareness and volition showed small positive correlations with organisational support (r = 0.182), supervisor support (r = 0.256) and performance management system (r = 0.290)as seen in Table 1. These correlations were all significant at the 0.01 level of significance. Constructive cognition showed small to negligible correlations with organisational support (r = 0.160), performance management system (r = 0.151) and supervisor support (r = 0.090). While the first two correlations were significant at the 0.05 level of significance, the latter was insignificant. Overall, the relationship between self-leadership and succession planning was thus practically small. The interest of the study related more to the relationship between these two constructs and turnover intention. The turnover intention had a small yet significant correlation with supervisor support (r = −0.292), while its correlations with organisational support (r = −0.308) and performance management system (r = −0.341) were moderate. All were significant at the 0.01 level of significance.

Lastly, the correlations of turnover intention with self-leadership were non-significant in both behavioural awareness and volition (r = 0.001) and constructive cognition (r = 0.061). There was thus no relationship between self-leadership and turnover intention. It seems that not all the variables had significant relationships.

For the purposes of the first mediation, the following variables were used:

Y = Turnover intention

X = Self-leadership

M = Performance management system

The analysis establishes the total effect of X on Y. This represents path c in the model. Table 2 highlights the mediation model. The regression of self-leadership on turnover intention, ignoring the mediator, was not significant, b = 0.114, p = 0.487.

The regression of the self-leadership on the mediator and performance management system was significant with b = 0.437, p = 0.002. The table also depicts that the mediator affected the outcome variable (estimate and test path b). However, it needs to be controlled for the independent variable. In this case, the relationship between the performance management system and turnover intention was significant (b = −0.5204, p = 0.000), controlling for self-leadership (H4).

In the second mediation, supervisor support was used as mediator. The relevant output is provided below.

For the purposes of the first mediation, the following variables were used:

Y = Turnover intention

X = Self-leadership

M = Supervisor support

In table 3, the analysis revealed that controlling supervisor support was not a significant predictor of turnover intention (b = 0.2745, p = 0.0820). In this case, the relationship between supervisor support and turnover intention was significant (b = −0.4394, p = 0.000), controlling for self-leadership (H5). The results were consistent with the results of a study conducted by Thompson (2018) which confirmed that increased self-leadership where the employee had developed self-awareness and positive emotions was associated with higher job satisfaction, affective commitment and lower turnover intention.

In the third mediation, organisational support was used as mediator. The relevant output is provided below.

For the purposes of the first mediation, the following variables were used:

Y = Turnover intention

X = Self-leadership

M = Organisational support

In table 4, the analysis revealed that controlling mediator (organisational support), self-leadership was a marginally significant predictor of turnover intention (b = 0.3221, p = 0.0431). In this case, the relationship between supervisor support and turnover intention was significant (b = −0.4394, p = 0.000), controlling for self-leadership (H6).

The outcomes of the hypotheses are presented in Table 5.

It is clear from Table 5 that Hypotheses 1, 2 and 6 were marginally accepted while Hypotheses 3 and 5 were rejected. Hypothesis 4 was accepted.

The findings of the study can be used by the government and managers to develop policies that enhance self-leadership and succession planning and reduce the intention to leave among employees. Currently, South Africa is experiencing high youth unemployment and the government can start addressing the problem of the ageing population, particularly those with scarce skills, by bringing in young graduates and helping them develop self-leadership as part of succession planning. Incentivising government departments and managers that take measures to promote talent and skills development and reduce turnover intention will ensure that there is buy-in from managers who may be sceptical of the policy.

In conducting the study, some limitations were experienced. Firstly, a large proportion of the sample comprised respondents who worked in Johannesburg. This indicates that the results may not accurately reflect what other employees from other towns think. Secondly, information within state-owned enterprises is generally restricted and this prevented some respondents from freely sharing their views. Lastly, this study was cross-sectional and cannot be generalised to other contexts.

Based on the results of the study, the following recommendations are put forward:

Future research could be conducted in the following areas: