Original Research

Investigating turnover intention in a financial organisation in Gauteng

Gloria S. Tshwane, Molefe J. Maleka, Portia M. Tladi

Received: 11 Nov. 2022; Accepted: 29 Mar. 2023; Published: 23 May 2023

Copyright: © 2023. The Author(s). Licensee: AOSIS.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Orientation: Losing employees prompts organisational competitiveness, which leads to increased financial organisation costs related to talent acquisition.

Research purpose: This study aimed to determine the relationship between authentic leadership and turnover intention and establish whether job satisfaction mediated the relationship between authentic leadership and turnover intention. Additionally, this study endeavoured to establish if the relation between authentic leadership, job satisfaction, and turnover intention is higher among post-Grade 12 employees than those with Grade 12 and lower qualifications.

Motivation for the study: There is limited research in the Gauteng financial sector concerning how job satisfaction is mediated and how education has diminished the relationship between authentic leadership and alleviating turnover intention.

Research approach/design and method: This study was influenced by the positivism paradigm and was quantitative, using an online survey to collect the data. The final sample consisted of 85 respondents.

Main findings: It was also established that authentic leadership did not significantly relate to turnover intention, while job satisfaction was a mediator between authentic leadership and turnover intention.

Practical/managerial implications: The significance of the study is that when employees are satisfied with leadership support, they provide honest and truthful feedback. This is especially true when employees work in an inclusive and harmonious work environment; thus, their turnover intention diminishes. This is very positive for finance organisations, which will incur talent acquisition costs as intention does not result in resignation.

Contribution/added value: This study developed a framework on how authentic leadership and job satisfaction indicators can diminish turnover intention.

Keywords: authentic leadership; educational level; financial organisation; job satisfaction; turnover intention.

Introduction

Human resource management (HRM) and business management literature reflect a situation whereby organisations worldwide deal with a significant problem referred to as turnover intention (Maloba & Pillay-Naidoo, 2022). Monyaki et al. (2022) defined turnover intention as an employee’s willingness to leave their current job voluntarily. Employee turnover has been escalating within the financial sector since 2012; one of its significant drivers is employee dissatisfaction due to increased workload (Al Jamil et al., 2022). Another predictor of turnover intention is related to poor and toxic leadership (Ntseke et al., 2022).

Toxic leaders significantly and negatively impact employees and work outcomes (Hattab et al., 2022; Yavas, 2016), which include low satisfaction and higher employee turnover intentions (Paltu & Brouwers, 2020). It has been established that a toxic leader can be a positive driver of employees’ turnover intention (Redelinghuys et al., 2020). A study conducted in Indonesia determined that toxic leadership enhanced turnover intention by 33% (Iqbal et al., 2022). Furthermore, a similar trend was observed among South African non-financial service organisations (Engelbrecht & Samuel, 2019). However, it also emerged in other research that the finance industry concluded that employees did not intend to resign because they reported to authentic leaders who were transparent and fostered constructive engagement while being receptive to diverse views (Olckers et al., 2020).

Moreover, toxic leaders are inclined to have a negative impact on employees and organisations (Zaabi et al., 2018); thus, the presence of a toxic leader tends to create a stressful work atmosphere. According to Coban (2021), employees working under toxic leaders often become extremely dissatisfied with their jobs and dread going to work; thus, these toxic leaders facilitate an environment whereby employees’ intention to leave increases. Similarly, Labrague et al. (2020) discovered that employees who work for supportive leaders tend to have lower turnover intentions when compared to those who work under toxic leaders. A South African study conducted in the non-financial sector presented that job satisfaction indicators, such as supervisor support, reward and compensation, and recognition, reduced turnover intention by 11% (Lesenyeho et al., 2018).

From an HRM perspective, a range of different indicators result in employees being either satisfied with their job. Barkhuizen and Gumede (2021) included indicators such as supervision, rewards, policies, recognition, fringe benefits, job security, talent development and compensation that influence employees’ intention to leave or remain with an organisation. Another study revealed that employees who were satisfied with their jobs tended to perform better, leading to increased job security in the organisation for a sustainable period of years. However, at the same time, dissatisfied workers experienced intentions to leave the organisation. In essence, satisfied workers demonstrate positive attitudes, feel more comfortable and become loyal to the company (Gebregziabher et al., 2020). Hennicks et al. (2022) mentioned that the nature of the work also appears to be a significant indicator of job satisfaction. Moreover, workers who actually receive feedback perceive their work as being exciting and challenging, while those who receive adequate remuneration are often presented with promotional opportunities. Therefore, these employees develop positive relationships with co-workers (Coetzee & Bester, 2021) and tend to have a high job satisfaction level, which prompts them to remain with the organisation.

Educational level has also been positively linked with turnover intention because employees who are more educated are often more likely to leave (Sabei et al., 2020). Elian et al. (2020) noted that employees with a higher level of education who work in financial organisations are more likely to resign than employees with lower educational levels. More specifically, employees who are more highly educated often have higher expectations of their current employer, which means that fulfilling their needs becomes much more difficult. Leadership and job satisfaction are, therefore, critical to reducing employee turnover intention within a financial services organisation. Accordingly, employees who were well-educated were the ones who intended to leave or left the organisation. This observation is similar to the findings of a study conducted by De Gieter and Hofmans (2015) at a Belgian financial institution. The researchers’ observations and De Gieter and Hofmans’ findings provided the other motivations that drove the researchers to conduct this study so as to provide a practical or managerial and theoretical contribution that relates to the situation in the South African context. People with a higher education level generally exhibit higher expectations regarding financial rights, benefits and auditing than employees with a lower educational level.

The research mentioned here reflects contradictory findings; however, these findings still encourage further research into how leadership and job satisfaction influence turnover intention. Moreover, it can be argued that most of these studies were not conducted in relation to finance organisations operating in Gauteng. Hence, this study was conducted to close the gap and contribute to the body of knowledge involving the intent to quit. To create new insights and knowledge related to turnover intention, the authors conducted this study involving a finance organisation in Gauteng, which was having a problem relevant to intentions leading to the actual resignation. This organisation’s annual report, which was not cited in this study due to a confidentiality clause signed by the authors, did indicate that educated employees were leaving. However, prior to this study, it was unclear to what extent managers, job satisfaction indicators (i.e. working conditions, benefits and job security) and education levels contributed to turnover intention.

Objectives of the study

The study objectives are as follows:

• To investigate the relationship between authentic leadership and turnover intention.
• To establish if job satisfaction has a mediating impact on the relationship between authentic leadership and turnover intention.
• To establish if the relationship between authentic leadership, job satisfaction and turnover intention is higher in the group of employees with a post-Grade 12 education as compared to those with Grade 12 and lower qualifications.

Literature review

Human resource management is defined as a field that involves talent acquisition, onboarding, rewards and benefits, employment relations, organisational development, talent development and retention (Maleka, 2012). Human resource management is also related to social justice, which means that leadership must treat employees fairly and equitably when interacting with subordinates. An HRM turnover intention framework proposed for this study is depicted in Figure 1.

FIGURE 1: Proposed study model.

Various authentic leadership examples have been included in this study, which also represent a variable for this study (refer to Figure 1). In addition, ‘authentic leader’ was chosen for this study, because authentic leaders are ethical and possess technical and interpersonal competencies; they are also trustworthy (Chen & Sriphon, 2022). In addition, previous leadership literature and practitioners emphasised the significance of authentic leaders being true to themselves, primarily because they possess positive attributes, such as honesty and sincerity, and deliberative of their values (Ramalu & Janadari, 2020). It has also been found that authentic leaders are very resilient when overcoming hardships and instil these characteristics in their followers (Smith, 2019). Due to these attributes, authentic leaders keep employees invigorated, dedicated, committed and satisfied (Arsawan et al., 2022).

Relationship between authentic leadership and turnover intention

Authentic leadership has been described as a leader’s competencies and capacity to work with employees to foster positive development (Iqbal et al., 2020). This involves focusing on their leadership style. Meanwhile, they can enhance employees’ commitment and behaviours, which might have a positive effect on their performance (Iqbal et al., 2018).

Additionally, authentic leadership is relevant to the leader’s competencies and abilities to work with employees and foster positive development (Iqbal et al., 2018). Moreover, authentic leaders create a supportive work environment, which, in turn, reduces employees’ turnover intentions and instil positive attitudes and behaviour in subordinates (Alkadash, 2020). Previously, it has been established that employees reporting to authentic leaders experience less turnover intention (Hwang et al., 2022) and are transparent in making decisions (Fox et al., 2020). Additionally, authentic leaders have been found to diminish employees’ turnover intentions because they treat employees well (Puni & Hilton, 2020). Such managers tend to connect with employees emotionally (Ribeiro et al., 2020), have excellent listening skills and base their decisions on data, not hearsay (Kaya & Karatepe, 2020; Wong et al., 2020). Lastly, Iqbal et al. (2018) ascertained that employees who worked under authentic leadership reported no intention to resign because they were supported even when they were under pressure.

In this study, the following was hypothesised:

H₁: Authentic leadership negatively predicts turnover intention.

Job satisfaction as a mediator between authentic leadership and turnover intention

Hair et al. (2022) opined that a mediator is a third variable that affects the direct relationship between two variables. The same authors added that mediation occurs when the relationship between the independent, mediator and dependent variables is significant. Suifan et al.’s (2017) theoretical framework forms a comparative and positive relationship between authentic leadership and job satisfaction and employee development. It emphasises the central role of authentic leaders in facilitating employees’ development opportunities and enabling satisfaction with work. Moreover, it has been found that authentic leaders have a positive impact on employees’ turnover intention, thus abating turnover within the financial sector (Novianti & Fuadiputra, 2021). Additionally, authentic leadership has been considered an essential component for creating a motivational and productive work environment. In short, authentic leadership has a highly significant impact on job satisfaction (Yamak & Eyupoglu, 2021). Recent research revealed that empowering and supportive leaders have been related to improved work effectiveness and job satisfaction outcomes. Similarly, Sultana et al. (2018) demonstrated that authentic leadership positively and significantly influenced job satisfaction, suggesting that authentic leaders increase job satisfaction and significantly lower turnover intention.

Job satisfaction is one of the primary determinants that influences turnover intention (Gebregziabher et al., 2020). Accordingly, studies revealed that employees who are satisfied with their development and advancement opportunities are less likely to consider leaving the organisation (Zhang & Fang, 2016). In a study examining the relationship between job satisfaction and turnover intention, the researchers found a significant relationship between job satisfaction and turnover intention (Andresen et al., 2017). Similarly, Suifan et al. (2017) reported a significant negative relationship between job satisfaction and turnover intention.

There are several job satisfaction indicators that tend to decrease employees’ turnover intention. For example, Ekhsan (2019) established that strategies such as job rotation and appropriate workload are essential to reducing employee turnover intentions. In some instances, it has been discovered that managers who create a conducive work environment and reward employees for performing well reduce turnover intention (Ekhsan, 2019). Moreover, recognising employees and offering them training opportunities are indicative of effective strategies for reducing employee turnover intentions (Pratama et al., 2022). Similarly, managers who create a conducive work environment and encourage effective teamwork activities have also significantly minimised turnover intention (Hidayat et al., 2021).

Research showed a mediated inverse relationship between employee satisfaction and turnover. High job satisfaction seems to help maintain a low turnover, while, on the other hand, high job dissatisfaction is likely to result in high turnover (Asriani & Riyanto, 2020). In addition, Munyaka et al. (2017) discovered that job satisfaction mediated the relationship between authentic leadership and turnover intention.

Hence the following hypothesis was made:

H₂: Job satisfaction is a mediator between authentic leadership and turnover intention.

Education as a moderator between authentic leadership and turnover intention

Leite (2022) defined education as a deliberate action to impart knowledge, skills and attitudes or structured learning through intentional or unintentional efforts. In this study, education relates to the General and Further Education and Training sub-framework (Grade 9 to Grade 12) and the Higher Education Qualification sub-framework (higher certificate to doctorate) (Maleka, 2020). This study categorised education as Grades 9 to 12 and certificate to doctorate. Previous studies have indicated that the degree of education is one of the demographic factors that consistently correlates with turnover intentions. According to a survey by Naidoo (2018), education moderates leadership and employee intent; thus, employees who were educated either intended to leave or left the organisation. This result is similar to a study conducted at a Belgian financial institution by De Gieter and Hofmans (2015).

The third hypothesis of the study is, therefore, as follows:

H₃: The relationship between authentic leadership, job satisfaction and turnover intention is higher among employees in the post-Grade 12 group as compared to those with Grade 12 and lower qualifications.

Figure 1 shows the proposed study model of the research.

Research design

Research approach

A quantitative and cross-sectional research design was selected for this study. The quantitative approach was utilised to test correlations between variables in this research (Du Plooy-Cilliers et al., 2019), while this study used the census sample approach because the sample size was less than 500 (Bothma & Roodt, 2013).

Research method

Research participants

One hundred (n = 100) employees at the selected financial services organisation responded to the online survey. Fifteen (n = 15) respondents were deleted as they failed to complete the questionnaire in its entirety. Hence, the final sample consisted of 85 employees, which comprised 34% of the target population within the 10% to 50% response rate, as suggested by Neumann (2000). The respondents were collectively categorised as totalling 76.5% female and 22.4% male. The majority of respondents (74.1%) were between the ages of 18 and 30, while 2.4% were between the ages of 51 and 65. Most respondents (56.5%) had a Grade 12 educational level, and 5.9% had bachelor’s degrees. As regards the tenure, 62.4% of respondents had 1 to 5 years of experience, while 5.9% had over 6 years of experience.

Research instruments

This study used a web-based online questionnaire to collect data. The research questionnaire was divided into two sections. Section A contained demographic data such as gender, age, length of service and education. Section B was used to measure authentic leadership, job satisfaction and turnover intention. Authentic leadership was measured through a scale adopted by Walumbwa et al. (2008) consisting of 19 items (α = 0.89). ‘I often consider quitting my job’ is one authentic leadership item measured in this. The assessment related to job satisfaction was provided through the Minnesota Satisfaction Questionnaire, consisting of 20 items. One of job satisfaction items measured is, ‘I am satisfied about the working conditions’.

The turnover intention was measured through the scale adopted by Sjöberg and Sverke (2000) and composed of five items (Beukes, 2019). Herewith is one of the turnover intentions measured in this study, ‘I often consider quitting my job’. All the items were rated on a five-point Likert scale from 1 (strongly disagree) to 5 (strongly agree). Prior South African research on turnover intention and job satisfaction showed Cronbach’s alphas of 0.93 and 0.96 (McCallaghan et al., 2020).

Research procedure

The Faculty of Management Sciences issued ethical research clearance at Tshwane University of Technology (TUT). The selected organisation permitted the authors to conduct the research study. In compliance with ethical principles, the researcher ensured that the respondents’ briefing communication regarding the study’s objective and methods was fair, honest and open. Respondents indicated their voluntary participation by clicking through the online questionnaire (Bester & Engelbrecht, 2009). They were also informed about the research objectives before participating (Connaway & Powell, 2010). Finally, the respondents’ anonymity was guaranteed, because the type of online questionnaire used made it impossible to identify individual respondents.

Statistical analysis

This study utilised descriptive statistics to determine the respondents’ demographic variables by using frequencies. Diverse inferential statistics were conducted. Firstly, reliability and validity statistics were conducted. In order to achieve the former, Cronbach’s alpha and composite reliability were calculated, and a cut-off of 0.7 was used (Hair et al., 2017). In addition, the average variance extracted (AVE) and the cut-off was used to ascertain convergent validity. In addition, an AVE score of 0.5 demonstrates that convergent validity has been achieved (Henseler et al., 2015). The heterotrait-monotrait (HTMT) was calculated to determine discriminant validity. Heterotrait-monotrait is defined as the ‘mean of all correlations of indicators across constructs measuring different constructs’ (Hair et al., 2022, p. 312).

Secondly, confirmatory tetrad analysis (CTA) was conducted to determine whether the structural equation model (SEM) was reflective or not. Hair et al. (2022) explained that CTA is a statistical procedure used to ascertain whether indicators are formatively and reflectively set up. In a reflective SEM model, the arrows point from the construct to the indicators (Hair et al., 2017). In the former, the arrows point from the indicators to the construct, while in the reflective model, the arrows point from the latent construct to the indicators or items. In this study, the Bonferroni-adjusted bootstrap confidence levels included zero for determining that the model was reflective.

Conversely, model indicators are formative if they are 90% bias-correlated, while the Bonferroni-adjusted bootstrap confidence levels should include zero (Hair et al., 2018). In order to calculate the tetrads, 5000 samples were bootstrapped. Field (2018, p. 1009) opined that bootstrapping is a sampling distribution statistic ‘estimated by taking the repeated sample from the dataset’.

Once it was determined through CTA that the model was reflective, this assessment procedure adhered to the following. Firstly, the variance inflation factor (VIF), which is a statistical technique to determine multicollinearity, was calculated. Indicators with values above five were deleted. Secondly, p-value, t-statistics and confidence intervals were used to assess model significance. If the relationship was 0.05 or 5% or less (Hoyle, 1995; Kline, 2010), the relationship was deemed significant. A threshold of −2 and + 2 indicated that the data were not significant in terms of confidence intervals. The relationship was deemed insignificant if it included zero, meaning that the upper and lower intervals had opposite directions. Conversely, if the results did not include zero, it indicated that the data were significant (Tabachnick & Fidell, 2013).

Thirdly, to assess the explanatory power, the R-squared was calculated. Fourthly, to determine the model’s predictive power, blindfolding, which is a resampling technique that computes Q², was used to calculate the reflective model’s predictive power (Hair et al., 2022).

Structural equation model was calculated by determining the direct and indirect effects. The former relates to bivariate relationships, where an arrow with a single pointer points from an independent to dependent variable. An independent variable is a variable that influences the other variable (Saunders et al., 2019). This study’s independent variable was authentic leadership, and the dependent variable represents turnover intention. Hair et al. (2022) opined that full mediation occurs when the direct effect is insignificant, and the indirect is significant. The same authors are of the view that partial mediation occurs when the direct and indirect effects are significant. When the indirect effects are not significant, mediation does not take place. Therefore, job satisfaction is the mediator in this study.

The other inferential statistical conducted was multigroup analysis (MGA). Cheah et al. (2020) postulated that MGA is a form of moderation where the moderator that is being used is dichotomous. The MGA approach was employed to assess moderation across multiple relationships. Multigroup analysis tests predefined data groups to ascertain significant differences across specific group parameter estimates (Hair et al., 2017). The groups that were compared were Grade 9 to Grade 12 and certificates to doctorate. The data were analysed in Statistical Packages for the Social Sciences (SPSS) version 26 and SmartPLS version 3. The former was used to conduct the sample distribution discussed in the section ‘Research participants’, while the latter was used to address the study’s objectives.

Ethical considerations

Ethical clearance to conduct this study was obtained from the Tshwane University of Technology Faculty of Management Sciences Research Ethics Committee (FCRE-ECO) (no. FRCE2020/FR/04/004 – MS[2]).

Results

The researchers conducted exploratory factor analysis (EFA) in SPSS, and because the sample was less than 150, as Pallant (2016) suggested, the data did not converge. The rationale for conducting EFA was to extract factors that could be used to conduct bivariate or correlational analysis. That being the case, the data were analysed in SmartPLS, which has the following capabilities: calculating the measurement model (i.e.. indicator loading, Cronbach’s alpha composite reliability and AVE); the SEM (i.e. to test the direct and mediation relationship) and MGA (i.e. moderation analysis). Thus, calculating the direct (bi-variate correlation analysis) relationship, mediation relationship and moderation analysis can be classified as inferential statistics.

Inferential statistics

Measurement model

In order to determine validity and reliability, a measurement model was used (refer to Table 1; it also lists the indicator loadings). The authentic leadership indicators with the highest loading (0.82) were B18 and B9, while job satisfaction indicators with the highest loading (0.84) were B25 and B37. The turnover intention indicator with the highest loading (0.95) was B43. The constructs, Cronbach’s alpha, and composite reliability scores were above the 0.7 threshold as suggested by Hair et al. (2022). Based on the data in Table 1, it can be argued that convergence validity was achieved.

Heterotrait-monotrait is a statistical technique employed to measure discriminant validity. According to the data in Table 2, none of the correlation values exceeded the threshold of 0.90, and none of the upper-bound confidence intervals was above the threshold of 0.85. This infers non-discriminant validity issues. Because the data showed that convergence and discriminant validity were achieved, the researchers discuss the structural equation modelling next.

Structural equation model

Structural equation model was conducted to address study objectives 1 and 2. The average VIFs shown in Table 3 are the five thresholds, which suggest no collinearity issues. The same authors opined that, in determining the model fit, the standardised root mean square residual (SRMR) value should be less than 0.12. The SRMR in Table 3 is 0.08, suggesting model fit.

Model specification

It was visibly clear from the data listed in Table 4 that no job satisfaction tetrads included zero. Therefore, the job satisfaction indicator did not include zero.

The direct effects are displayed in Table 5. Only the relationship between authentic leadership and turnover intention was not significant. It was insignificant because the t-statistic (1.80) was within the –2 to 2 range, and the confidence intervals included zero [−0.57, 0.21]. Based on this result, H₁ is not supported.

Mediation is calculated by multiplying the coefficient of a times coefficient b. In Table 5, the coefficient of a is 0.41 (authentic leadership -> job satisfaction), and the coefficient of b is −0.54 (job satisfaction -> turnover intention). The product of multiplying them (0.41 * −0.54) is −0.22. The mediation score is significant because the t-statistics (3.23) falls outside the –2 to 2 range and the confidence interval did not include zero [−0.38, −0.11]. It can be argued that full mediation occurred because the direct effects were insignificant. Based on these findings, H₂ is supported.

The calculation of the total effects entails adding the direct and indirect effects. This study calculated the total effects of authentic leadership -> turnover intention as follows: −0.12 + (−0.22) = −0.39. Not all the total effects resulting from the t-statistics (4.31) fell within the –2 to +2 range and the confidence intervals did not include zero [−0.58, −0.23]; thus, the results were deemed significant.

The statistical model of the study is depicted in Figure 2.

FIGURE 2: Job satisfaction as a mediator between authentic leadership and turnover intention.

Multigroup analysis

To address the third goal of the study objective, MGA was conducted. As can be observed in Table 6, the results are not significant. The path correlation scores not shown in Table 6 were as follows: authentic leadership to job satisfaction – r = 0.402, M = [−0.411, 0.372]; authentic leadership to turnover intention – r = −0.063, [−0.422, 0.423] and job satisfaction to turnover intention – r = −0.568, [−0.335, 0.324].

Before testing H₃, it was confirmed that each group had a minimum of 10 respondents. Because there were 52 respondents from Grades 9 to 12 and 33 respondents in the certificate to doctorate group, MGA was conducted. A bootstrap with 5000 samples was calculated. As can be observed in Table 7, all p-values exceeded 5%, which means that H₃ cannot be supported.

Discussion

This study aimed to determine the relationship between authentic leadership and turnover intention and establish if job satisfaction mediated the relationship between authentic leadership and turnover intention. It also ascertained the moderating effect of education on the relationship between authentic leadership and turnover intention. This study also created insights using under-researched employees at an organisation within the financial sector. Moreover, the study revealed that authentic leadership diminished turnover intention, even though the relationship was insignificant. A plausible reason is that the relationship was very weak (−0.12). This finding is consistent with the literature (Sultana et al., 2018). It was also verified that authentic leaders diminished turnover intention because they treated employees with fairness and equity (Puni & Hilton, 2020), connected with employees emotionally (Ribeiro et al., 2020), had excellent listening skills and based their decisions on data, not hearsay (Kaya & Karatepe, 2020; Wong et al., 2020). In addition, other scholars established that employees reporting under authentic leadership did not have an intention to resign, because such leaders also supported them when they were under pressure (Iqbal et al., 2018).

Another contribution of this study is that it showed that job satisfaction mediated the relationship between authentic leadership and turnover intention. The literature revealed that employees reporting to an authentic leader were satisfied because this type of leader is motivational, empowering and supportive (Sultana et al., 2018). In addition, authentic leaders satisfy followers because they provide a clear strategic direction and possess technical competencies to solve operational challenges (Chen & Sriphon, 2022).

Moreover, the literature revealed that job satisfaction indicators, job rotation and proper workload are essential to reduce employee turnover (Ekhsan, 2019). Consistent with the literature, this study demonstrated that job satisfaction diminishes turnover intention (Gebregziabher et al., 2020). The relationship was negative and significant (−0.54). Another indicator that diminishes turnover intention is when managers create a conducive work environment and reward employees for performing well. Such managers diminish employees’ turnover intention because they encourage employees’ development and advancement opportunities (Dodanwala & Santoso, 2021; Pratama et al., 2022; Zhang & Fang, 2016). In addition, such managers create a conducive working environment in which culture is combined to achieve departmental objectives (Modaresnezhad et al., 2021).

It was previously determined that, within the financial sector, employees with the highest qualification had a high propensity to resign (Elian et al., 2020). Other than poor working conditions or working under toxic leaders, educated employees had a turnover intention due to the high demands imposed upon them within the financial labour market (De Gieter & Hofmans, 2015). The downside of losing educated and skilled workers is that the financial organisation loses its competitive edge and incurs the cost of attracting talented and educated employees (Cooke et al., 2022). Not only do they incur talent acquisition costs, but also talent development costs (Bussin, 2018).

Practical implications

One of the study’s practical implications is the determination that employees are less likely to resign when they are satisfied with the organisation’s policies. Factors such as retention and work-life balance policies, paying above the median or 50th percentile, authentic leaders who recognise and support employees, and offering benefits such as training and development reduce turnover intention (Bussin, 2018). Oh (2020) ascertained that employees were satisfied and did not have turnover intentions, because management implemented talent acquisition and diversity policies that were consistent and transparent. This information shows that policies alone are sufficient to reduce turnover intention; however, managers and supervisors still play a role in implementing such policies.

Another practical implication of this study is that it has been determined that giving employees regular feedback and supporting them when under pressure play a salient role in diminishing turnover intention. An additional positive spin-off is that such employees support managers’ strategic goals or objectives (Rawashdeh et al., 2022). Furthermore, subordinates of managers who are recognised and praised for performing well do not focus their efforts on looking for employment opportunities elsewhere (Chênevert et al., 2022).

An additional implication is that when employees feel satisfied and have high job security, they will not seek another position; thus, their intention to exit the company will be lower. Hu et al. (2022) discovered that fringe benefits, including paid annual leave, paid sick leave, maternity leave, insurance plans and retirement accounts, played a vital role in reducing turnover intention. The study’s final practical implication is that employees are less likely to leave their jobs for reasons other than retirement when they are satisfied with leadership support (Brewer et al., 2022).

Limitations of the study

Despite this study’s theoretical and methodological contribution, it has several limitations. Firstly, it cannot be generalised to the whole organisation because it was conducted only in the Gauteng branch. Secondly, the study was limited to a single financial services company; therefore, the results cannot be generalised to other organisations in the financial services sector. Thirdly, this study also used a small sample size; hence, a non-parametric SEM was used. This can be attributed to collecting the data during coronavirus disease 2019 (COVID-19). Further data could not be collected because the first author was retrenched when the organisation reduced labour costs. Finally, the data collection was cross-sectional and provided only a once-off picture of the relationship among the variables. It must also be noted that the authors are not claiming that SmartPLS Version 3 is not credible. In fact, it has advanced functions like CTA, which are not available in other data analysis software programs such as SPSS and STATA.

Recommendations

Similar studies can be performed in other sectors in the future. It is further recommended to generalise the results. This can be achieved if other researchers use a larger sample when testing the relationships between variables in this study so that external validity can be achieved and the results can be generalised. A mixed-method research design can further assist researchers in the future in triangulating results. The following recommendations are suggested for management at the selected financial services organisation:

• Managers should be exposed to leadership intervention with authentic leadership indicators.
• Managers should attend leadership interventions focusing on scenario planning and decision-making based on data.
• The rewards or fringe benefits should be reviewed at specific intervals to determine their relevance.
• The organisation should encourage managers to informally recognise employees if there is no budget to implement a formal recognition programme.
• Managers should delegate and rotate employees so that they can develop various competencies (cross-training) needed by the organisation.
• Managers should be encouraged to continuously furnish employees with feedback about organisational changes and performance.

Conclusion

The results of this study can empower HRM practitioners working in talent acquisition and line managers to reduce turnover intention. The findings further reflect that managers and leaders who listen, are objective and make decisions based on data, and are ethical and give continuous feedback actually enhance employee satisfaction and diminish turnover intention. Another conclusion that can be drawn from the data is that a leader with both technical and soft skills plays a salient role in maintaining employee satisfaction and diminishing their turnover intention. Furthermore, this study established that job satisfaction explained the variance of turnover intention by 41%. Finally, it is concluded that indicators such as supervisors who treat employees well, as well as praising, recognising, giving regular feedback and offering employees an opportunity to work with other employees, diminish employees’ turnover intentions.

Acknowledgements

This article is part of Gloria Setjwane Tshwane master’s dissertation.

The authors declare that they have no financial or personal relationships that may have inappropriately influenced them in writing this article.

G.S.T., M.J.M. and P.M.T. contributed to the design and implementation of the research, analysis of the results and writing of the manuscript.

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

The data that support the findings of this study are available on request from the author, M.M.

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of any affiliated agency of the authors.

References