The impact of the talent management mechanism and self-perceived talent status on work engagement: the case of Japan

ABSTRACT

This study uses a quantitative analysis to investigate the connection between the talent management (TM) mechanism and work engagement with a perceived moderating effect of the self-perceived talent status on work engagement. Using responses from 471 respondents in six Japanese firms, I propose and test hypotheses regarding the TM mechanism, work engagement and the moderating role of the self-perceived talent status between the TM mechanism and work engagement. The results demonstrate that the TM mechanism and the self-perceived talent status have mutually exclusive and positive effects on work engagement. This finding corroborates the previous studies that employed exclusive and inclusive approaches. However, this study failed to find any concrete evidence that supports the moderating role of the self-perceived talent status.

KEYWORDS:

• Exclusive approach
• inclusive approach
• Japan
• self-perceived talent status
• talent management mechanism
• work engagement

Introduction

Although no unanimous definition exists, both business and academic communities are rapidly accepting the importance of the concept of talent management (TM) (Lewis and Heckman 2006; ASTD 2009; Collings and Mellahi 2009). Both pundits and business practitioners have yet to agree on the definition of ‘talent’, instead of adding confusion by arguing that both exclusive and inclusive approaches to it can define the concept (Dries 2013; Gallardo-Gallardo, Dries, and González-Cruz 2013; Gallardo-Gallardo 2019). Dries (2013, 279), for instance, suggests that the central question of TM should be whether it offers ‘a commitment to only some employees’ (exclusive) or ‘a commitment to all employees’ (inclusive).

Previously, some scholars contended that the notion of strategic talent management (STM) should have an exclusive connotation that companies should ‘focus on only some employees’ (Collings and Mellahi 2009, 306). Their logic of explanation seems sound. Although global talent management (GTM) depends on the mechanism of STM, the GTM approach is interested in finding out how firms implement these policies for their global markets (Minbaeva and Collings 2013; Scullion, Collings, and Caligiuri 2010). Therefore, it is always possible for us to analyse the mechanisms of STM and GTM using an exclusive approach.

On the other hand, contenders to this exclusive approach have constructed what we call fully inclusive talent management (FITM), an approach that treats all employees as talented (Swailes, Downs, and Orr 2014; Savanevičienė and Vilčiauskaitė 2017). Facing these two contradictory approaches, Thunnissen, Boselie, and Fruytier (2013) argue that the concept of TM is too broad, while the majority of TM research wrongly deals with only organizational level variables. They underscore the fact that individual level research that banks on empirical evidence using quantitative analyses is few and far between (De Boeck, Meyers, and Dries 2018; Sparrow 2019).

I concur that individual level research is important, since many TM researchers have shown that employee engagement is technically an outcome of TM policies (De Boeck, Meyers, and Dries 2018; Pandita and Ray 2018). Unfortunately, however, research on TM that employs quantitative analysis has not yet corroborated any significant correlation between the mechanism of TM policies and employee engagement (De Boeck, Meyers, and Dries 2018). I also contend that we should take the talent status into serious consideration, as it directly asks the question of how and why we classify an employee into a talent (Björkman et al. 2013; De Boeck, Meyers, and Dries 2018). The same question also examines and measures the effectiveness of TM policies, since employees who consider themselves talents should play important roles in the TM mechanism (De Boeck, Meyers, and Dries 2018). This study therefore examines the moderating effect of the talent status between the TM mechanism and work engagement.

Furthermore, long term, if not lifetime, employment is a well-known feature of the Japanese employment system (Marsden 1999; Moriguchi 2013), which is still rampant in large companies (Moriguchi 2013; Tseng 2020). Under the Japanese employment system, firms treat all employees as talents (Marsden 1999; Moriguchi 2013), making TM irrelevant to their human resource management (HRM) practices. Therefore, despite the sheer number of studies on the Japanese employment system, few really deal with TM in Japan. The only exception would be Ishiyama and Yamashita (2017), who reveal the presence of TM as an STM practice (an exclusive approach) in several Japanese companies.

Considering the scant research on the systematic TM in Japanese companies, this research intends to build on the literature by clarifying the reality of TM in Japanese companies through the mechanism of TM policies and the talent status. In addition, findings from Japanese companies can be useful to other countries in the Asia Pacific, where the exclusive and inclusive TM approaches are concomitantly present. The next two sections of this paper provide a literature review and hypotheses development, while the fourth section justifies the methodology employed and test the hypotheses. I present findings in the fifth section, while the last two sections discuss the outcomes and implications of the study.

Literature review

The definition of TM

Myriad researchers take the term ‘talent’ for granted and therefore neglect to define the term explicitly (Tansley 2011; Gallardo-Gallardo 2019). Historically, the word ‘talent’ was used as a unit of weight or money. Since the Middle Ages, the meaning has changed to connotate ability or aptitude (Tansley 2011). Today, in the TM context, we use the term as an object or subject. As an object, talent refers to people who have valuable and exceptional characteristics, while, as a subject it, refers to people who have exceptional abilities and skills (Gallardo-Gallardo, Dries, and González-Cruz 2013; Gallardo-Gallardo 2019).

As the term talent is vaguely defined, the definition of TM has a wide scope for discussion, both academically and practically (Collings and Mellahi 2009). Originally, TM was concerned with how acquiring capable personnel directly produced a competitive edge for a company (Michaels, Handfield-Jones, and Axelrod 2001). This concept, called the war for talent approach, classifies employees into A, B, or C class talents based on their ability to perform. A company’s market position is critical for hiring as many A level talents as possible while aiming to reduce the number of C level applicants to a minimum (Lewis and Heckman 2006).

However, the war for talent approach does not provide a specific definition of A level talents (Lewis and Heckman 2006) making it difficult for firms to eschew deploying A level employees to nonstrategic positions, a typical case of HRM mismanagement that might be construed as overinvestment by firms (Minbaeva and Collings 2013). One strategy to overcome the problem of this approach is to find solutions by studying STM and GTM closely, both of which denounce the misconception that TM is simply another instance of HRM (Lewis and Heckman 2006).

Collings and Mellahi (2009, 304) define STM as:

Activities and processes that involve the systematic identification of key positions which differentially contribute to the organization’s sustainable competitive advantage, the development of a talent pool of high potential and high performing incumbents to fill these roles, and the development of a differentiated human resource architecture to facilitate filling these positions with competent incumbents and to ensure their continued commitment to the organization.

In this definition, talents are not universally defined as A level recruits. After all, talents in key positions play critical roles of giving organizations a competitive edge.

Building on the STM mechanisms, the GTM approach focuses on how firms implement these policies internationally (McDonnell et al. 2010; Scullion, Collings, and Caligiuri 2010; Minbaeva and Collings 2013). Specifically, GTM is defined as ‘the systematic identification of pivotal positions’, ‘the development of a talent pool of high potential and high performing incumbents’ and ‘the development of a differentiated HR architecture’ on a global scale (Collings, Mellahi, and Cascio 2019, 5).

Due to their emphasis on talents for key positions, we can discern STM and GTM as exclusive approaches. Conversely, FITM (an inclusive approach) is a new perspective that differs from STM or GTM (Swailes, Downs, and Orr 2014; Savanevičienė and Vilčiauskaitė 2017). FITM is defined as ‘the recognition and acceptance that all employees have talent, together with the ongoing evaluation and deployment of employees in positions that give the best fit and opportunity (via participation) for employees to use those talents’ (Swailes, Downs, and Orr 2014, 533). Briefly, the premise of FITM is that all individuals possess some sort of competency. FITM’s approach to competency is impacted by positive psychology, wherein individuals can build on their strengths and positive affirmations increase their future prospects and sense of wellbeing (Peterson and Seligman 2003). An individual’s talent could have a unique value that is appropriate for a company’s corporate culture or business model, which is based on its HR architecture (Lepak and Snell 1999). Buckingham and Vosburgh (2001) also point out that each individual has unique talents, and organizations should identify and bolster their employees’ unique talents.

The TM mechanism

As noted briefly above, some of the previous studies contend that TM is not really different from HRM, especially with regard to TM’s inclusive approach that takes all employees as talents (Mensah 2019; Schreuder and Noorman 2019). However, TM studies on business corporations have also shown that, firstly, firms must clearly define their business strategies, and, secondly, firms should also create profiles for each of the required personnel to demonstrate employees’ unique value propositions in the market (Ishiyama and Yamashita 2017).

As noted earlier, the core points of STM are to identify firstly firms’ key positions based on their corporate strategies; secondly, to develop high potentials and high performers within the talent pool; and lastly, to develop a differentiated HR architecture (Collings and Mellahi 2009). This differentiated HR architecture that centres on the resource-based view (RBV) model contributes to an organization’s sustainable competitive advantage (Collings, Mellahi, and Cascio 2019).

The RBV model includes causal ambiguity, path dependency, social complexity and tacit knowledge, all of which emphasize the firm’s competitive advantage realized by its unique resources and values (Barney 1991). In accordance with the RBV model, Bowman and Hird (2014) argue that it would be ineffective to bank on the vaguely defined concept of TM. Instead, through a precise understanding of a firm’s specific strategies in current and future directions, they suggest that firms fit talents to firm-specific knowledge and values. Similar to STM and GTM, The RBV model suggests that firms retain individual talents unique to a company’s corporate culture or business model.

Finally, Ishiyama and Yamashita (2017) propose that the inclusive approach allows firms to create an ideal personnel profile for each talent. If firms complete each HRM stage of selection, promotion and training with this ideal personnel profile in mind, then each employee can develop his or her capabilities as an individual, making it easier for employees to feel a sense of organizational wellbeing (Swailes, Downs, and Orr 2014).

To sum up, common to the exclusive and inclusive approaches, the core points of TM’s mechanism are to establish a clear definition of a firm’s business strategy and create profiles of the personnel required. Through the TM mechanism, firms can acquire a sustainable advantage and set up a structural barrier against competitors.

TM and work engagement

Regardless of whether TM acts through the exclusive or inclusive approach, the ultimate goal of TM is to improve firm performances. Both exclusive and inclusive approaches commonly predict that TM will have enormous impacts on organizations through talent performances that can be measured on an individual level. Using a meta-analysis, Pandita and Ray (2018), for example, analyse a firm’s process of motivating and retaining employees through TM by nurturing its talents and expanding their capabilities as a means of motivating employee engagement. Employee engagement on the other hand reduces employee turnovers and attracts more talents. Not only does it promote the fulfilment of job duties among employees, but it also facilitates enthusiastic process improvements (Saks 2006; Seijts and Crim 2006; Albrecht et al. 2015).

Highly engaged employees behave enthusiastically to produce low rates of absenteeism and continuously prompt performance improvements (Swathi 2014; Pandita and Bedarkar 2015; Pandita and Ray 2018). These employees build favourable long-term relationships with clients (Salanova, Agut, and Peiró 2005; Pandita and Ray 2018) and are likely to maintain professional relationships with their former companies even after severance (Seijts and Crim 2006; Pandita and Ray 2018). It requires no explanation why this type of employee engagement makes a positive, long-term impact on organizations.

In this study I treat employee engagement as work engagement. Albeit advocated by major business consultants, academic communities, however, have rarely endorsed employee engagement. Scholars highlighted the dearth of any new insights the concept provides to the study of HRM and organizations, other than coining such trite terms as organizational commitment and extra-role behaviour (Schaufeli and Bakker 2010). In other words, employee engagement is not any different from the existing concepts within the study of organizations and is nothing but ‘old wine in a new bottle’ (Macey and Schneider 2008, 10). The distinction between employee engagement and organizational outcomes is therefore all but ambiguous. Work engagement appears more academic and robust as a theoretical concept than employee engagement (Schaufeli and Bakker 2010).

As opposed to the concept of burnout, which occurs in an environment where stress levels are high, and jobs are overly complex, work engagement is a positive and fulfiling psychological state about one’s work that is highly favoured by the employee. Work engagement fosters physical and mental health, reinforces job satisfaction, reduces the desire to leave one’s job, facilitates the fulfilment of task requirements, augments the desire to fulfil duties outside one’s job and encourages leadership behaviour. A large body of research suggests that work engagement is a determinant of task performances (Schaufeli and Bakker 2010; Bakker and Albrecht 2018).

As noted earlier, the FITM approach (an inclusive approach) engenders positive psychologies, where individuals can build on their strengths, and positive affirmations about them can improve their future prospects and sense of wellbeing. The concept of work engagement is therefore related to a sense of wellbeing. In this sense, work engagement is well suited to be an outcome variable in the inclusive approach. In the exclusive approach, work engagement turns out to have long-term effects on corporate performances. For these reasons, work engagement is appropriate as an outcome variable that can measure the engagement effects in both the exclusive and inclusive approaches.

TM in Japan

Tansley (2011) points out that, at least in Japanese culture, the term ‘talent’ refers to accomplishments and resourcefulness that employees can acquire and nurture rather than being an innate quality. This suggests that talents in Japan are the consequence of mastery, which takes many years to attain. Tseng (2020) argues that long-term employment and firm-specific training are the solid pillars of the Japanese traditional employment system. Moreover, 86.3% of Japanese companies consider their employment system as traditional rather than a system based on job evaluations (Persol Research and Consulting 2019). In addition, Sato (2020) examines the speed of promotion in Japanese companies using the data ranging from 2002 to 2015 and finds that the speed has slowed down over time, partly due to the ageing of the workforce.

Therefore, the traditional employment system is a key element for Japanese companies. What is important in this regard is that the comprehensive characteristics of the traditional employment system makes an example of the inclusive approach, as it is a system that develops many employees as talents over a long period through long-term employment. However, at the same time, the Japanese systems also retains some features of the exclusive approach, such as covert fast track. Even though the overall promotion speed of all employees slowed down in the statistical analysis, in reality, a competing system was in place that favoured some employees over others in terms of task placements and the speed of promotions, as the preferred ones received special treatments as fast track employees (Koike 1981; Uehara 2007; Ichinose 2013).

Hirano (2011) points out that some Japanese companies indeed embrace a hybrid employment system that mixes long-term employment with performance evaluations. This hybrid system, which was unpopular during the previous periods, is an attempt to make the covert fast track visible through job evaluations. In the Japanese hybrid employment system, however, the competence of talents is still realized as an end result of a long-term learning and accomplishment. In other words, the hybrid system loosely incorporates the characteristics of the exclusive approach in a more visible way than the traditional system in order to perfect the comprehensive nature of the inclusive approach. Nevertheless, some STMs in Japan admittedly take more clearly exclusive approaches than others, despite the ongoing predominance of either inclusive or hybrid approaches.

From a theoretical standpoint, STM research warrants a close attention from the academic community. Anlesinya, Dartey-Baah, and Amponsah-Tawiah (2019), for example, point out that scholars have conducted STM research against the backdrop of the European and Asian contexts, and therefore it has not been US-centric at all. From a global perspective, essential TM processes are the same in emerging and emergent markets (Sparrow et al. 2018). In a similar vein, Ishiyama and Yamashita (2017) point out the presence of Japanese STM practices. Romans and Lardner (2005) also consider the integrated TM system at Becton Dickinson, Japan, as an STM practice.

In a nutshell, TM is still an emerging concept in academia. As the Japanese traditional employment system is being adapted to a hybrid system, some crucial facets of the exclusive approach are gradually being incorporated into a hybrid system, although the inclusive approach is still dominant. Moreover, the existence of a clear STM (i.e. an exclusive approach) in Japan makes it difficult to deny the fact that both inclusive and exclusive approaches are visible in Japanese firms.

Hypotheses

Regardless of the exclusive and inclusive approaches, the core aspect of the TM mechanism is to ascertain a clear definition of a firm’s business strategy and create a profile for each employee. Furthermore, the main components of TM are selection, promotion and training based on the company’s desired personnel profile (Ishiyama and Yamashita 2017; Tafti, Mahmoudsalehi, and Amiri 2017). Therefore, in this paper, I define the particular TM mechanism as the selection, promotion and training of employees based on the company’s desired personnel profile.

As mentioned above, the quantitative analysis of the TM mechanism using tangible measurement tools is few and far between (De Boeck, Meyers, and Dries 2018; Sparrow 2019). For instance, Jayaraman, Talib, and Khan (2018) proposed an integrated TM scale that is comprised of four dimensions: identifying critical positions, competence training, development and reward management. However, the measurement does not address the TM mechanism. To ameliorate this lacuna, this study examines the effectiveness of the TM mechanism using an original scale.

To generate testable hypotheses, I start out by examining the direct impact of the TM mechanism on work engagement. One common element of the mechanism of STM and GTM is their ability to ensure employees’ continued commitment to the organization (Collings and Mellahi 2009; Collings, Mellahi, and Cascio 2019). The reason for their continued commitment is twofold. Firstly, they consider themselves competent incumbents who can fill key positions; therefore, they have continued commitment to the organization. Secondly, firms nurture employees based on the organization’s desired personnel profile, which, by extension, are in harmony with the positions they occupy. The literature on STM and GTM has clearly substantiated the second reason for employees’ continued commitment. In the literature on the inclusive approach, for example, Swailes, Downs, and Orr (2014) point out that all employees should be deployed in positions in which they can be the best fit and have the opportunity to use their talent. Keeping records about such fit and opportunities are pivotal in the creation of organization’s desired personnel profiles for positions, and this is valid for both exclusive and inclusive approaches.

Furthermore, many authors present the job demands-resources (JD-R) model as an antecedent of work engagement (Demerouti et al. 2001; Bakker and Demerouti 2017). On the one hand, in the JD-R model, job demands exacerbate burnout if the resources required for tasks are scarce. On the other hand, job and personal resources promote work engagement if their amount is appropriate for the demand.

Job resources (e.g. autonomy, social support, supervisory coaching and opportunities for professional development) are crucial assets that foster the development of employees’ abilities, which, in turn, lead to organizational commitment. Likewise, personal resources (e.g. self-efficacy, organization-based self-esteem and optimism) are vital assets that foster employees’ stress resistance and physical/emotional wellbeing (Xanthopoulou et al. 2007; Bakker and Demerouti 2017). This means that the mechanism of TM falls into the category of job resources, together constituting firm policies.

Individuals who are aware of these organizational standards are more likely to be satisfied with the procedure of selection, promotion and training processesm which in turn incurs a positive impact on the accomplishment of business tasks (De Boeck, Meyers, and Dries 2018; Tafti, Mahmoudsalehi, and Amiri 2017). Furthermore, with a clear understanding of the organization’s strategy for nurturing talents and deploying them to appropriate positions, employees will gradually eliminate conflicts among workplace members over goals, enabling them to focus on completing tasks in a smooth and harmonious manner (Ishiyama and Yamashita 2017). It is therefore beyond doubt that the whole mechanism will improve work engagement.

To sum up, this study focuses on the individual awareness of the TM mechanism, as it is common to both the exclusive and inclusive approaches. Therefore, the following hypothesis is conceivable:

Hypothesis 1: The individual awareness of the TM mechanism has a positive effect on work engagement.

The concept of the talent status, which refers to a system of classification that defines whether an employee is a talent or not (Björkman et al. 2013), is based on social exchange theory. Employees who consider themselves as talents react positively to their organization’s favourable treatment of them (Björkman et al. 2013; De Boeck, Meyers, and Dries 2018). For instance, employees who are identified as talents perceive stronger sentiments of organizational support extended to them than those who are not (Gelens et al. 2015). In addition, the former also demonstrate stronger commitment to building competencies than the latter (Björkman et al. 2013). Furthermore, employees who are recruited into a talent pool and identified as high potentials have exalted expectations regarding their future career (Dries and Pepermans 2008; Swailes and Blackburn 2016).

Against this backdrop, the talent status is mainly defined in the literature as formal identification as a talented employee by an organization. Although few scholars talk about the self-perceived talent status, their theoretical and practical contributions are not yet significant in the literature (De Boeck, Meyers, and Dries 2018). However, in reality, the concept of the self-perceived talent status exists in the inclusive approach. As Swailes, Downs, and Orr (2014) point out, the inclusive approach requires that work be performed consistently with the position. This observation entails that employees will soon develop a perception of whether they are talents, based on their personal evaluation of whether they fulfilled the task performance required for their positions. In terms of the exclusive approach, talents can be identified clearly in line with the TM programmes. Therefore, using the formal, organizational identification is an appropriate and useful research tool. Meanwhile, since all employees are defined as talents in the inclusive approach, it is difficult to carry out research about such organizations, where the distinction between talents and non-talents is impossible to obtain.

On the other hand, social exchange theory found that employees with different talent statuses would respond differently to the organizational treatment of employees (Björkman et al. 2013; De Boeck, Meyers, and Dries 2018). If employees perceive that their recruiters classified them as talents with a fully operational TM mechanism, the talents will feel that the TM mechanism will boost up their performances. As a consequence, they will also expect to be treated well by the organization. All in all, the employee’s work engagement is likely to increase. However, if employees do not perceive that they are not classified as talents, even with a fully operational TM mechanism in present, social exchange theory will predict that they will not expect to be treated well by the organization. Therefore, a positive impact on work engagement is unlikely to occur, which leads us to the second hypothesis:

Hypothesis 2: The self-perceived talent status will moderate the positive relationship between individual awareness of the TM mechanism and work engagement in such a fashion that the relationship gets stronger for the higher self-perceived talent status than the lower.

Figure 1 shows the conceptual model drawn from the two hypotheses. In this model, the TM mechanism is an independent variable, the talent status is a moderator variable, and work engagement is a dependent variable.

Figure 1. Analysis model based on the hypotheses

Methodology

Assessment scale

In this study, I created a set of 13 questions with a five-point scale to assess individual awareness of the TM mechanism. Myriads of previous studies have already constructed comparable questions independently of mine, using different theoretical and qualitative research methods (see inter alia, Lewis and Heckman 2006; Collings and Mellahi 2009; Swailes, Downs, and Orr 2014; Ishiyama and Yamashita 2017). These questions intend to identify a company’s definition of its business strategy, its business communication and how the personnel profiles are shared among key stakeholders of the organization. The questions, therefore, assess an individual’s awareness of the selection, promotion and training policies and whether they are properly aligned with the company’s desired personnel profile.

This study measures thee self-perceived talent status using four items from the 13-item ‘Self-Perceived Employability Scale’ developed in Japanese by Yamamoto (2014). In a rare study, Björkman et al. (2013) also examined the self-perceived talent status quantitatively. However, they defined the self-perceived talent status as personal perceptions of belonging to a talent pool. It is simply awkward to use this type of self-perception as a crucial concept in the inclusive approach because it is unclear whether the employee is formally considered as a talent by his employers. Therefore, in this study, I measure the self-perceived talent status by employability. Yamamoto’s (2014) Self-Perceived Employability Scale introduces two-dimensional concept by which we can measure an employee’s market value in both the internal and external labour markets. In the meantime, the talent status signals an employee’s perception that he or she is valued in the internal labour market. Therefore, in this study I selected four items from the scale, which are most relevant to the perception of one’s competence in the internal labour market: ‘I am promising in this organization because my employer appreciates my contribution’, ‘Because of my skills and experience, the company I am currently working for considers me a value-added resource’, ‘I have a high reputation in my company for my work performance and reputation’ and ‘My company sees me as a valuable asset to the organization’.

Next, I operationalized work engagement into nine items (on a seven-point scale) from a Japanese version of the Utrecht Work Engagement Scale (Shimazu et al. 2008). For control variables, I used dummies for gender (male = 1), position (1 = managerial class) and job changes, while concomitantly putting a control on age. In addition, since I constructed the sample from the six companies, I defined company dummies to address differences in work engagement among the employees from different companies. Furthermore, I assigned leader-member exchange (LMX) as a control variable because more studies than before increasingly treat leadership as an influence in a relationship (Uhl-Bien 2006).

LMX, the impact superiors and subordinates have on each other, is one example of such relationships that has attracted numerous studies (Graen and Uhl-Bien 1995). Moreover, if LMX is deemed favourable by employees, work engagement increases through support from superiors as a job resource (Shimazu et al. 2008). Therefore, it is not fortuitous to predict that LMX enhances work engagement. Furthermore, the TM mechanism is taken for granted that it primarily operated by supervisors. Therefore, if LMX is not controlled, it is difficult to verify the impact of the TM mechanism. Hence, I added LMX as a control variable. The seven items on a five-point scale collected from the LMX-7 questionnaire (Graen and Uhl-Bien 1995) intend to measure LMX. I developed and validated the LMX-7 questionnaire in English first, which was then translated into Japanese and finally back-translated into English to verify that the meaning of the Japanese version was in line with the English version (Brislin 1970).

Data collection

This study collected responses from 471 participants in six cooperating companies in Japan from August 2016 to January 2018. The companies¹ were named Company A (distribution firm: 95 responses), Company B (consulting firm: 75), Company C (consulting firm: 18 responses), Company D (IT firm: 108 responses), Company E (IT firm: 38 responses) and Company F (pharmaceutical firm: 137 responses).

As noted aforementioned, both the inclusive and exclusive TM approaches are present and observable in Japanese companies. Therefore, to clarify the reality of TM using the TM mechanism and the talent status, Japanese companies are an appropriate setting to investigate both approaches. However, it is also noteworthy that a large proportion of companies maintain the traditional Japanese employment system (Persol Research and Consulting 2019). Therefore, to select an appropriate sample for this study, it was mandatory to verify whether the company adopts the traditional or the TM employment strategy. For this purpose, meetings were held in advance between the author and the six participating companies, who agreed to cooperate with the author towards a goal of incorporating feedback from the results of the study into their TM policies. As long as they remain willing to adopt new TM policies based on my findings, it was deemed appropriate to use these companies for this study because they considered their employment system as TM rather than traditional. The type of industry they are in, size and age are not homogenous among these companies due to the scarcity of Japanese companies that embrace TM as the employment system.

Employees in the six companies were asked by the author to respond to the survey through a URL created by the author’s research laboratory. Employees were informed by me that their personal data would not be disclosed to their company and that their responses would remain confidential. I used caution not to create any bias by relationship building with the participating companies. Furthermore, scholars still regard talent competency as a long-term accomplishment in Japanese companies (Hirano 2011). Therefore, I reserved my data collection to full-time employees for this study. The demographics of the 471 responses analysed are as follows: 349 respondents were men, and 119 were women, and the average age was 39.31 (with a standard deviation of 8.20).

This study derived its dependent and independent variables from self-reports of the same pool of respondents, thus making it necessary to account for the common method bias. However, the self-reports were deemed ideal to me for assessing the psychological concepts in this study, such as work engagement and TM, because they are usually affected by personal perceptions (Conway and Lance 2010). I also conducted Harman’s single-factor test to examine the presence of potential bias across the same respondents (Podsakoff and Organ 1986). An exploratory factor analysis (i.e. principal factor analysis) of all variables revealed four factors with eigenvalues greater than one with a cumulative contribution ratio of 62.06%. However, the contribution ratio of the first factor was 38.77% and did not pose a serious issue.

Data analysis

Factor analysis, descriptive statistics and correlations

The entire analysis that I carried out in this paper starts from a factor analysis of the TM mechanism to measure the significance of each factor dimension. The result clearly indicates that no item corresponds to the ceiling or floor effect, ratifying my assumption that all items have correctly been included in calculating the mean values and standard deviations. Subsequently, I completed an exploratory factor analysis using IBM SPSS 21.0 for all 13 items that are pertinent to the TM mechanism. The one-factor framework therefore turned out appropriate for interpretation, based on changes in the eigenvalues. Consequently, I could easily single out one important factor from the thirteen items. The cumulative contribution ratio is 60.86%, as Table 1 males the ultimate factor pattern.

Table 1. Factor analysis results: TM mechanism

	Factor loadings
The company’s desired personnel profile is clear	.83
Personnel evaluations are conducted in accordance with the company’s desired personnel profile	.82
Personnel are trained in accordance with the company’s desired personnel profile	.81
Personnel are trained based on the business strategy	.80
Personnel are selected for promotion in accordance with the company’s desired personnel profile	.80
The company’s desired personnel profile is communicated clearly	.79
The company’s desired personnel profile is in accordance with the company’s job duties	.78
New hires are chosen appropriately	.78
Hiring is conducted in accordance with the company’s desired personnel profile	.78
Personnel are appropriately selected for promotion	.77
After hiring, personnel receive support until they are established in their jobs	.68
After hiring, personnel receive thorough follow-up support until they adjust to the company environment	.66
I have a clear understanding of how I should improve my capabilities based on the company’s desired candidate profile	.53

As noted earlier, the TM mechanism is defined as the selection, promotion and training of employees based on the company’s desired personnel profile. The above single factor is related to this definition, and the Cronbach’s alpha produced an acceptable number (α = 0.95). Therefore, the factor is not only appropriate and can be named ‘TM mechanism’.

Correlation coefficients, means, standard deviations and correlations

In addition to the scales for the factor analysis, I also calculated the mean values to learn the mean scores for the LMX scale (α = 0.92), the Work Engagement Scale (α = 0.93) and talent status (α = 0.88). The mean values, standard deviations and correlations for each variable are shown in Table 2.

Table 2. Mean values, standard deviations, and correlations for each variable

Variables	N	Mean	SD	1	2	3	4	5	6	7
1.Gender	468	0.75	0.44	-
2.Age	454	39.31	8.2	.26***	-
3.Position	471	0.16	0.37	.15**	.19***	-
4.Job change	471	0.17	0.37	−.10*	−.22***	−.07	-
5.LMX	471	3.68	0.88	.04	.07	.10*	.05	-
6.TM mechanism	471	3.1	0.85	−.04	−.03	.10*	.06	.54***	-
7.Talent status	471	3.02	0.86	−.02	−.02	.27***	.02	.36***	.40***	-
8.Work engagement	471	3.4	1.03	.07	.10*	.19***	−.04	.41***	.41***	.44***

***p < .001,**p < .01,*p < .05.

Gender: 1 = male, 0 = female; Position: 1 = managerial class, 0 = other classes; Job change: 1 = change, 0 = no change.

Ordinary least squares (OLS) regression results

Table 3 shows the results of the OLS regression analysis. The variance inflation factor scores for each variable did not reveal any serious problems (VIF = 1.096–2.312). In Step 1, I could affirm that position, the dummy for companies (E and F) and LMX, among the control variables, were positively correlated with work engagement. The results show that managers had higher work engagement than rank-file employees; that differences in work engagement existed among companies; and that LMX had a positive effect on work engagement, as predicted by previous studies. In Step 2, even after controlling for these variables, I could confirm that the TM mechanism was still positively related to work engagement. All these therefore mean that Hypothesis 1 is supported.

Table 3. Ordinary least squares regression results

	β:Work engagement
Variables	Step1	Step2	Step3
Control Variables
Gender	.05	.08	.08
Age	.01	.03	.03
Position	.16***	.09*	.09*
Job change	−.01	−.02	−.02
Company dummy (standard: Company A)
Company B	.08	.08	.08
Company C	.08	.03	.03
Company D	.05	.03	.03
Company E	.21***	.16**	.16**
Company F	.30***	.26***	.26***
LMX	.33***	.10**	.10**
Independent variables
TM mechanism		.29***	.29***
Talent status		.25***	.25***
TM Mechanism × Talent status			.04
Adjusted R^2	.26***	.39***	.39***
ΔR^2		.13***	.00

***p < .001,**p < .01, *p < .05 N = 450,VIF = 1.096–2.312.

Gender: 1 = male, 0 = female; Position: 1 = managerial class, 0 = other classes; Job change: 1 = change, 0 = no change

Next, I analysed Hypothesis 2 by utilizing the method Frazier, Tix, and Barron (2004) recommended applying for studies of this scale/magnitude. To avoid multicollinearity, I centred the TM mechanism and the talent status. I inserted the control variables, the TM mechanism and the talent status, into Step 2. In Step 3, I also added the interaction between the TM mechanism and the talent status. Steps 2 and 3 present the result of a hierarchically moderated regression analysis. From Step 2, I could affirm that both TM mechanism and talent status were positively related to work engagement. However, in Step 3, I failed to confirm any significant correlation of the interaction of the TM mechanism and the talent status with work engagement.

The results, however, demonstrate no significant evidence that corroborates the moderating role of the talent status. The TM mechanism and the talent status have mutually exclusive effects on work engagement only without any concomitant effect on the dependent variable. Therefore, Hypothesis 2 is not supported.

Discussion

In the extant literature, many scholars have criticized TM, especially the inclusive approach, for its similarity to HRM. They found it difficult to embrace the premise of TM that the inclusive approach tends to assume that all employees are talented. Furthermore, these previous studies treated the talent status as formal recognition of employee talent by firms, eliding the significant aspect of TM that incorporates the self-perceived talent status in its theoretical framework. Using the self-perceived talent status, this study identified the role of the talent status as a predictor of work engagement.

This study elicited responses from 471 participants in six cooperating companies in Japan. The OLS regression analysis shows that the TM mechanism and talent status had a positive effect on work engagement but found no evidence in support of the moderating role of the talent status. In other words, this study identified the TM mechanism and talent status as independent predictors of work engagement.

The results suggest that the TM mechanism and the talent status are common to both exclusive and inclusive approaches. Therefore, the TM mechanism can be a differentiator between TM and HRM, even though all employees are deemed talented.

Furthermore, this study clarified the real differences between the Japanese employment system and the TM mechanism, which includes both exclusive and inclusive approaches. Though a large number of Japanese companies maintain the traditional Japanese employment system, a few of them are now adopting TM, including both inclusive and exclusive approaches in their employment system. This study therefore revealed the presence of the TM mechanism within Japanese companies that differ from the traditional employment system.

Theoretical implications

The theoretical implications of this study are threefold. Firstly, in terms of the individual awareness of the TM mechanism, I can notice that the desired personnel profile based on the company’s business strategy is an important factor. For individual awareness of the mechanism, the factor analysis also verified the one-factor framework. In addition, the OLS regression analysis and covariance structure analysis showed that the TM mechanism had a positive effect on work engagement. As per the hypotheses, the paper found that employees’ understanding of the standards for selection, promotion and training improves, leading to increased work engagement, if a business strategy is well defined, and the personnel profile required to execute it is properly disseminated. This means that, in the TM context, understanding the ideal personnel profile is an essential factor for job performance and employee wellbeing. Previous studies have not provided any quantitative validation of the TM mechanism, unlike HRM. However, in this study, a scale of the TM mechanism, constructed based on theory and qualitative research, shows its positive impact on work engagement. The theoretical significance is the identification of the concrete scale that can examine both exclusive and inclusive approaches.

Secondly, this study found no evidence of the moderating role of the talent status. Hypothesis 2 predicted that the talent status would moderate a positive relationship between the individual awareness of the TM mechanism and work engagement because, based on social exchange theory, if the talent status were high, employees would expect to be treated well by the organization in line with the TM mechanism. By contrast, if the talent status were low, they would not anticipate fair treatment by the organization in line with the TM mechanism. However, as mentioned above, this study found no evidence of the moderating role of the talent status, which means that even if the talent status were low, the TM mechanism could be a predictor for work engagement. In other words, for all employees, the TM mechanism is effective as a reliable predictor of work engagement.

Why did the results differ from the hypothesized relationships? The reason may be that the TM mechanism aligns the company’s strategy with its desired personnel profile, so that employees can work smoothly with a clear understanding of the company’s strategy. This implies that the effectiveness of employees’ understanding of the corporate strategy, while performing their duties, is independent of the self-perceived talent status. In this sense, this study has clarified the value of the TM mechanism.

Thirdly, as noted earlier, most previous studies of TM defined the talent status as a formal classification of individual talents by an organization. However, this study clarifies the importance of the self-perceived talent status in investigating the common features of exclusive and inclusive approaches.

Practical implications

The practical significance of this study involves the way in which companies introduce the TM mechanism to their firms in the Asia Pacific. The study clarifies that the TM mechanism and the talent status are common to both exclusive and inclusive approaches and that the TM mechanism can be a differentiator between TM and HRM. In Japanese firms, both exclusive and inclusive approaches are adopted by firms; however, the exclusive approach has not been the mainstream. Therefore, TM, banking on the inclusive approach, is in the process of gradually permeating into Japanese firms. Similarly, for other countries in the Asia Pacific, where both approaches exist and especially for those where the inclusive approach is dominant, the use of the TM mechanism may be effective in disseminating and adopting TM.

Furthermore, clarifying and disseminating the desired personnel profile is an important factor, even as the comprehensive implementation of this practice throughout the company by line managers plays a large role. The importance of the role played by the line managers is also supported by the fact that LMX, regardless of its status as control variables, has a positive impact on Steps 2 and 3, as well as on the TM mechanism itself. Previously, Martin and Schmidt (2010) pointed out that line managers tend to be short-term focused in terms of talent development. Therefore, they concluded that talent nurturing should not be delegated to line managers, especially in the exclusive approach. However, the current findings offer a new perspective: Companies must be aware that it is not enough to define the desired personnel profile; it is also important to empower superiors to implement the TM mechanism.

Limitations and future research

This study obtained its independent and dependent variables from the self-reports from the same pool of respondents, thus making it necessary to take the common method bias into account. Nonetheless, I find the self-reports ideal for assessing the psychological concepts analysed in this study, such as work engagement and the self-perceived talent status, because these individual level perceptions are often affected by personal perceptions (Conway and Lance 2010). In addition, as noted earlier, Harman’s single factor analysis did not pose a serious issue, although discussions on this issue need to continue in the near future.

In addition, experimental research on the TM mechanism is still scarce. Using large samples that consist of a massive number of companies and divisions would facilitate more intricate analyses than sample with small numbers. This issue needs to be further explored by the academic community in the near future.

Conclusion

This study highlights the impact of the TM mechanism and the talent status on work engagement. Prior studies assumed that TM, especially the inclusive approach, is similar to HRM. The findings of this study show that the TM mechanism can be a differentiator between TM (both exclusive and inclusive approaches) and HRM because of employees’ understanding of the desired personnel profile that reflects a firm’s business strategy for selection, promotion and training.

Furthermore, this study finds no evidence of the moderating role of the talent status. This means that the TM mechanism and the self-perceived talent status can be mutually exclusive predictors of work engagement. In other words, for all employees, the TM mechanism is effective for work engagement.

Acknowledgements

I would like to acknowledge the constructive comments and encouragement received from Professor Ingyu Oh, Associate Editor of APBR, and other anonymous reviewers. I would also like to extend my gratitude to the six participating companies in Japan for their cooperation and contributions to the value of this work.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Nobutaka Ishiyama

Nobutaka Ishiyama is a Professor at the Hosei Graduate School of Regional Policy Design, Hosei University, Tokyo, Japan. He received his PhD in policy, planning and development from Hosei Graduate School of Regional Policy Design. He has published numerous books and articles such as Mechanisms of Cross-Boundary Learning Communities of Practice and Job Crafting (Cambridge Scholars Publishing) and ‘Role of Knowledge Brokers in Communities of Practice in Japan’ (Journal of Knowledge Management). He has received the JAHRD Award from the Japanese Academy of Human Resource Development (2018).

Notes

1. The response rate and the number of years in operation for each company are as follows: Company A (86.5%, 43 years), Company B (38.1%, 21 years), Company C (13.5%, 20 years), Company D (31.3%, 24 years), Company E (15.3%, 8 years) and Company F (27.0%, 59 years).