Although discrepancies occur regarding the components that constitute empowering leadership behaviours (Ahearne, Mathieu, & Rapp, 2005; Amundsen & Martinsen, 2014; Arnold, Arad, Rhoades, & Drasgow, 2000; Konczak, Stelly, & Trusty, 2000), studies generally concur that it sufficiently differs from other leadership constructs (Cheong, Yammarino, Dionne, Spain, & Tsai, 2019; Sharma & Kirkman, 2015). This confirms the scholarly relevance of leader empowering behaviour.

Konczak et al. (2000) conceptualised leader empowering behaviour according to six dimensions: delegation of authority, accountability, self-directed decision-making, skills development, coaching for innovative performance and information sharing. Thus, leaders who display this type of behaviour offer their followers a larger chunk of authority, responsibility and independence. Through knowledge sharing and the prioritisation of skills development, employees are sufficiently coached, novelty is encouraged and failures are regarded as building blocks (Konczak et al., 2000). When leaders empower their employees in a manner that satisfies their desire to be empowered, positive outcomes generally ensue (Lee, Cheong, Kim, & Yun, 2017).

Associations between leader empowering behaviour, person–environment fit, job satisfaction and intention to leave have been studied before. Redelinghuys and Botha (2016) found that person–environment fit indirectly affected the intention to leave via job satisfaction. Additionally, they established that leader empowering behaviour moderated the relationship between person–environment fit, job satisfaction and intention to leave. They, however, did not consider the potential direct impact leader empowering behaviour may have on the study outcomes. Therefore, the aim of the study was to assess the associations leader empowering behaviour have with person–environment fit, job satisfaction and intention to leave.

Person–environment fit refers to the degree to which individual characteristics (needs, abilities, values) are in line with environmental characteristics (supplies, demands, values) (Kristof-Brown & Guay, 2011). When employees experience a good fit with their work environment, their values correspond to those of the organisation (person–organisation fit), their needs are sufficiently addressed by the organisation’s reward structure (needs–supplies fit), and they have the necessary capabilities to deal with the demands of their job (demands–abilities fit) (Cable & DeRue, 2002).

Leaders play a pivotal role in employees’ work experiences (Tuckey, Bakker, & Dollard, 2012), where perceived good leadership usually equates to positive outcomes. Although studies assessing leader empowering behaviour as an antecedent of person–environment fit have been scarce, there is some evidence to suggest a positive association between the constructs (Cai, Cai, Sun, & Ma, 2018). In their study, Cai et al. (2018) established that leader empowering behaviour is significantly associated with person–job fit (consisting of demands–abilities fit and needs–supplies fit). Even though they assessed person–environment fit from a multidimensional perspective (person–job fit and person–group fit), they excluded person–organisation fit. In line with Cable and DeRue’s (2002) three-dimensional person–environment fit conceptualisation, we included person–organisation fit and differentiated between demands–abilities fit and needs–supplies fit. Therefore, the following hypothesis is posited:

Job satisfaction refers to the general satisfaction an employee has with his or her job (Spector, 1997). Numerous studies suggest a positive association between leader empowering behaviour and job satisfaction (Amundsen & Martinsen, 2014; Konczak et al., 2000; Vecchio, Justin, & Pearce, 2010). A meta-analytic study (Kim, Beehr, & Prewett, 2018) has also confirmed these findings. Thus, employees should be more satisfied in work environments where their leaders exhibit empowering behaviours such as involvement in decision-making, skills improvement opportunities, coaching and knowledge sharing. Therefore, the following hypothesis is posited:

Lastly, by creating a favourable work environment for employees, leaders may reduce the need for employees to explore alternative job opportunities. Intention to leave refers to the relative strength of an employee’s intention to seek employment opportunities elsewhere (Hom & Griffeth, 1991). The role superiors play in employees’ decisions to remain or depart from organisations is well supported empirically (Kiazad, Holtom, Hom, & Newman, 2015; Taplin & Winterton, 2007) and reflects the words of Buckingham and Coffman (1999, p. 32): ‘People leave managers, not companies’. In line with social exchange theory (Blau, 1964; Cropanzano & Mitchell, 2005), employees generally respond positively towards an entity (e.g. team, organisation, supervisor) that beneficially impacts them. Amongst others, these positive employee responses may entail organisational commitment (Settoon, Bennett, & Liden, 1996), which is indicative of an employee’s decision to remain with an organisation. The negative association between leader empowering behaviour and intention to leave has been confirmed by previous studies (Bester, Stander, & Van Zyl, 2015; Van Schalkwyk, Du Toit, Bothma, & Rothmann, 2010). A meta-analytic study (Kim et al., 2018) has also confirmed a relationship between leader empowering behaviour and withdrawal behaviour (e.g. intention to leave). Therefore, the following hypothesis is posited:

The Leader Empowering Behaviour Questionnaire (LEBQ; Konczak et al., 2000) assessed the leader empowering behaviour. The scale comprises 19 questions rated on a seven-point response scale, ranging from 1 (strongly disagree) to 7 (strongly agree). Five of the six subscales were measured by three questions: delegation of authority (e.g. ‘My manager gives me the authority to make changes necessary to improve things’), accountability (e.g. ‘I am held accountable for performance and results’), self-directed decision making (e.g. ‘My manager relies on me to make my own decisions about issues that affect how work gets done’), skills development (e.g. ‘My manager ensures that continuous learning and skill development are priorities in our department’) and coaching for innovative performance (e.g. ‘My manager focuses on corrective action rather than placing blame when I make a mistake’). Information sharing was measured by four questions: (e.g. ‘My manager provides me with the information I need to meet customers’ needs’). The scale has yielded reliable and valid findings in South African contexts (Bester et al., 2015).

The Perceived Fit Scale (PFS; Cable & DeRue, 2002) assessed the person–environment fit. The scale comprises nine questions rated on a seven-point response scale, ranging from 1 (strongly disagree) to 7 (strongly agree). All three subscales include three questions: person–organisation fit (e.g. ‘My organisation’s values and culture provide a good fit with the things that I value in life’), demands–abilities fit (e.g. ‘My personal abilities and education provide a good match with the demands that my job places on me’) and needs–supplies fit (e.g. ‘The job that I currently hold gives me just about everything that I want from a job’). The scale’s three-factor structure and reliability is well established in the South African context (Redelinghuys, Rothmann, & Botha, 2019a).

The Job Satisfaction Scale (JSS; Hellgren, Sjöberg, & Sverke, 1997) assessed the job satisfaction. The scale comprises three questions rated on a five-point response scale, ranging from 1 (disagree) to 5 (agree). Covering a solitary dimension, a sample item includes: ‘I am content with my job’. The scale has yielded reliable and valid findings in South African contexts (Smit, De Beer, & Pienaar, 2016).

The Turnover Intentions Scale (TIS; Sjöberg & Sverke, 2000) assessed the intention to leave. The scale comprises three questions rated on a five-point response scale, ranging from 1 (strongly disagree) to 5 (strongly agree). Covering a single dimension, a sample item includes: ‘If I was completely free to choose, I would leave this job’. The scale has yielded reliable and valid findings in South African contexts (Redelinghuys, Rothmann, & Botha, 2019b).

Study permission was granted by the retail organisation’s management team and the North-West University’s ethical committee. Participants had the opportunity to voluntarily complete the questionnaires in the comfort of their own workspace. The study was sufficiently introduced and explained by an information letter accompanying the questionnaires. Participants had ample opportunities to ask questions pertaining to the study.

Mplus 8.2 (Muthén & Muthén, 1998–2019) and SPSS25 (IBM Corp, 2017) were chosen as the preferred statistical software. Structural equation modelling was used to assess the study hypotheses. The following fit indices in Mplus were used to assess the suitability of the models: the Chi-square value, the standardised root mean square residual (SRMR), the root mean square error of approximation (RMSEA), Tucker–Lewis Index (TLI), comparative fit index (CFI), Akaike information criterion (AIC) and the Bayesian information criterion (BIC). On all the indices, lower values indicate better fit, except for the CFI and TLI, where higher values indicate better fit (Wang & Wang, 2012).

The study received ethical clearance from Optentia Ethics Committee, North-West University. Ethical clearance number: OPT-2014-001.

Table 2 reports the descriptive statistics, Raykov’s rho values and correlation coefficients of the measuring battery.

Sufficient Raykov’s (2009) rho values were found for each scale (> 0.70). All variables were practically and statistically significantly related to each other to varying degrees.

To assess the factor structures of the constructs under scrutiny, confirmatory factor analyses via Mplus was conducted. An initial measurement model (Model 1) was specified and evaluated against alternative models (Models 2–4) to establish the best-fitting model.

Model 1 comprised two second-order factors: leader empowering behaviour and person–environment fit and two first-order factors: job satisfaction and intention to leave (each measured by three observed variables). Leader empowering behaviour comprised six factors: delegation of authority, accountability, self-directed decision-making, information sharing, skills development and coaching for innovative performance. Each factor was measured by three observed variables, except for information sharing (measured by four observed variables). Person–environment fit comprised three factors: person–organisation fit, demands–abilities fit and needs–supplies fit (each measured by three observed variables).

Competing measurement models were also tested. In Model 2, person–environment fit comprised two factors: person–organisation fit (measured by three observed variables) and person–job fit (needs–supplies and demands–abilities fit) (measured by six observed variables). In Model 3, leader empowering behaviour comprised one factor (measured by 19 observed variables). In Model 4, person–environment fit comprised one factor (measured by nine observed variables).

The original measurement model yielded a χ² of 891.542 (df = 512). The CFI (0.95), TLI (0.94), RMSEA (0.04), χ²/df (1.74) and SRMR values (0.05) were all acceptable. The AIC and BIC values of the initial measurement model were also the lowest. Results are indicated in Table 3.

A structural model (Model 5) was specified and tested in line with the best fitting measurement model (Model 1). No difference in fit values between Model 1 and Model 5 were found, indicating adequate model specification. Model 5 produced the following fit statistics: χ² = 891.542; df = 512; p < 0.001; TLI = 0.95; CFI = 0.94; RMSEA = 0.04 (90% CI 0.060, 0.065); SRMR = 0.05; AIC = 44325.15; BIC = 44791.56.

Figure 1 illustrates the standard path coefficients found with leader empowering behaviour as independent variable and person–environment fit, job satisfaction and intention to leave as dependent variables.

For the model portion focusing on person–environment fit, leader empowering behaviour’s path coefficient (β = 0.69; p ≤ 0.01) was statistically significant and displayed the anticipated sign. Hypothesis 1 is therefore supported.

For the model portion focusing on job satisfaction, leader empowering behaviour’s path coefficient (β = 0.65; p ≤ 0.01) was statistically significant and displayed the anticipated sign. Hypothesis 2 is therefore supported.

For the model portion focusing on intention to leave, leader empowering behaviour’s path coefficient (β = -0.49; p ≤ 0.01) was statistically significant and displayed the anticipated sign. Hypothesis 3 is therefore supported.

Regarding effect sizes (Cohen, 1988), the model accounted for the following percentages of variance explained (R²): person–environment fit = 47% (large effect), job satisfaction = 42% (large effect) and intention to leave = 24% (medium effect).

The study had various limitations. As the study focused on a single retail organisation, one cannot generalise the findings to the larger retail industry or other contexts. Furthermore, this study measured person–environment fit in a static manner, despite contrary findings by previous studies (Boon & Biron, 2016; De Beer, Rothmann, & Mostert, 2016; Gabriel, Diefendorff, Chandler, Moran, & Greguras, 2014; Van Vianen, 2018).

Leaders need to be in sync with what their employees want and adapt their leadership style accordingly. Some employees may want more coaching and authority, whilst others may instead want more accountability, as the amount of coaching and authority they currently receive is in line with what they prefer. Others might want less authority and decision-making power, as it exceeds the need they have for it. Although empowerment, in essence, has a positive connotation, research has shown that one should be cautious of the ‘too-much-of-a-good-thing’ effect (Sharma & Kirkman, 2015). Accordingly, the diversity of employee preferences calls for balanced and informed leadership. When the right balance between under-empowerment and over-empowerment is achieved, leaders should bring out the best in their employees.

Future research may study the proposed relationships of this study in different contexts to enable generalisation of findings to a larger audience. This will simultaneously add to the thin literature base that exists regarding the relationship between person–environment fit and leader empowering behaviour. As various employee and environmental characteristics may change over time, it may also be worthwhile to assess the constructs from a daily diary level, weekly diary level or other longitudinal perspectives.