Formal voice mechanisms and portfolio career workers’ prosocial voice in Japan and Korea: the mediating role of managers’ issue-related leadership activities

ABSTRACT

This study analyses the relationship between formal voice mechanisms and prosocial voice among portfolio career workers (PCWs) in Japan and Korea. We particularly focus on the leadership activities of managers as human resource management agents and issue sellers. Under similar conditions, data on 400 and 409 PCWs in Japan and Korea, respectively, are gathered through web-based longitudinal surveys conducted in 2017 and 2018. The findings are threefold. First, when PCWs perceive that formal voice mechanisms are activated, they also rate the levels of their managers’ issue-related leadership activities more highly. Second, when PCWs evaluate the issue-related leadership activities of managers as being at a high level, they perceive that employment relations are based on a social exchange relationship. Third, when PCWs perceive employment relations based on a social exchange relationship, they provide their prosocial voice more actively. These results are discussed relative to the internal labour market models of Japan and Korea.

KEYWORDS:

• Formal voice mechanisms
• Japan and Korea
• longitudinal analysis
• managers’ issue-related leadership activities
• portfolio career worker
• prosocial voice

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The various systems that constitute FVMs are also referred to as ‘participatory employment practices’ (Kato and Morishima 2002), ‘participative human resource management’ (Lee and Chee 1996), ‘employee participation programs’ (Cooke 1994), ‘the communication category in HPWPs’ (Evans and Davis 2005), ‘team voice’ (Kim, MacDuffie, and Pil 2010), and ‘direct voice’ (Holland et al. 2011). For consistency, this study refers to all of these systems as ‘voice mechanisms’ based on the terminology used in Spencer’s (1986) pioneering study of FVMs. Furthermore, given that informal voice mechanisms have received emphasis in recent discussions of employee voice activation (Morrison 2011), we specifically use the term ‘FVMs’ to denote that we are investigating FVMs within personnel systems.

2. The goal of this study is to explain the ‘within variation’ of prosocial voice. In terms of variables, the total variation of prosocial voice is divided into ‘between variation’ and ‘within variation’. ‘Between variation’ refers to variation that maintains a stable value over time. A typical example of individual-level ‘between variation’ is a personality variable. ‘Within variation’ refers to a variable whose value changes as time passes. Taking the PCWs discussed in this study as an example, as they adapt to a new organization and, thus, better understand its business processes, they may become more easily able to express prosocial voice. Alternatively, if PCWs cannot smoothly construct interpersonal relationships in new workplaces, their expression of prosocial voice may feel constrained. This example shows that the ‘within variation’ of prosocial voice is liable to change over time. The need to distinguish ‘within variation’ and ‘between variation’ arises from the criticism that existing OB research on voice has over-specialized in ‘between variation’ (Kaufman 2015). For example, a significant relationship between the big five personality traits and prosocial voice has been reported in the OB field (LePine and Van Dyne 2001), but such approaches based on ‘between variation’ are limited by the fact that they cannot explain variation that changes over time. As far as we have been able to determine, ‘within variation’ in prosocial voice has not been demonstrated previously.

3. Unlike other studies focusing on events as the unit of analysis that remain in theoretical review stage, Kim, Baik, and Kim (2014) provide operationally definable and measurable constructs.

4. There are two types of mediation models: the single mediation model, which has one mediating variable, and the multiple mediation model, which has two or more mediating variables. Multiple mediation models can be further classified as the parallel mediation model, which presumes no causal relationship between the mediating variables, and the serial mediation model, which presumes a causal relationship between the mediating variables. For the purposes of this study, ‘single mediating effect’ shall refer to the single mediation model, and ‘double mediating effects’ shall refer to a serial mediation model with two mediating variables.

5. These mediators are not the only mediating mechanisms of the channel between FVMs and prosocial voice. For example, trust in management (Hu and Jiang 2016) and expected positive performance outcomes (Yuan and Woodman 2010) may act as significant mediators. Therefore, our mediators are expected to have only a partial mediating effect.

6. We excluded women for two reasons. First, when we conducted a preliminary investigation, we found that women were a tiny minority because only a small percentage of managers at or above the section chief level in Japan and Korea are women. Second, it is a distinct possibility that women may be unable to utilize prosocial voice due to pregnancy and childrearing concerns. For example, a female PCW who wishes to avoid long workdays because of her parenting duties probably will not offer opinions on issues outside of her job duties, even if she has ideas about them, because out-role behaviours, including prosocial voice, come with personal costs such as work/life imbalance or stress (Bolino and Turnley 2005).

7. The Sobel test is normally used to verify a single mediation effect. However, several methodological limitations to the use of the Sobel test for serial mediation effects, including the double-mediating effect in this study, have been identified (Preacher and Hayes 2004). The bootstrapping approach has been proposed as a countermeasure to these limitations when dealing with regression equation-based mediation models (Preacher and Hayes 2004, 2008).

Additional information

Funding

This research is supported by the grants from the Ministry of Education Culture, Sports, Science and Technology Japan Program for the Strategic Research Foundation at Private Universities, 2014–2018.

Notes on contributors

Seonjo Kim

Seonjo Kim is a doctoral student in graduate school of business administration at Rikkyo University. His research interests include interdisciplinary approaches to portfolio career workers voice and participation in new organization. He is particularly concerned how these workers perceive and react to institutional and organizational environments, with an emphasis on the social identity processes that influence their status evaluation and discretionary work behaviours.

Jun Ishikawa

Jun Ishikawa, Ph.D., is a Professor of organizational behaviour and human resource management at Rikkyo University. His research interests include leadership, R&D management, and strategic human resource management.