Supervisor–subordinate kankei,job satisfaction and work outcomes in Japanese firms

Abstract

In this study, we revisit a research model in which the relationships between Japanese supervisor–subordinate kankei (S-S kankei) and organisational commitment, job promotions and turnover intentions are mediated through job satisfaction. Based on the data randomly collected from a sample of 201 Japanese subordinates in three manufacturing firms in Tokyo, Japan, the results of structural equation modelling tests indicated that job satisfaction does partially mediate the relationships between Japanese S-S kankei and organisational commitment, job promotions and turnover intentions. The theoretical implication is that S-S relationship–job satisfaction–work outcomes model, supported in the Chinese context is equally valid in the Japanese context, suggesting the robustness of this mediation model. The practical implication is that Japanese firms are advised to encourage the development of S-S relationship and make sure that employees are satisfied with their jobs in order to enhance an employee’s commitment to the organisation, facilitate job promotions and reduce his/her intentions to leave.

Keywords

Japan job satisfaction organisational commitment turnover intentions

Introduction

Year 2011 witnessed China’s overtaking Japan as the world’s second largest economy (BBC News, 2011). The two countries have many things in common. They are both located in Asia and share some Confucian cultural values. Some researchers have attributed Japan’s economic success to its development of unique management practices through the selective borrowing of Chinese Confucian cultural values such as benevolent love, righteousness, propriety, wisdom and faithfulness, and many of these values are still visible in modern Japanese human resource (HR) management practices (e.g. see Carr and Markus, 2006; Magoshi and Chang, 2009). Japan and China’s cultures are both characterised by high power distance, strong collectivism and long-term orientation (Minkov and Hofstede, 2011; Whitehill, 1991). The Japanese cultural values of collectivism, individual sacrifice, loyalty, commitment and centralised leadership are all deeply embedded in Chinese Confucian values (Zhu and Warner, 2000). Therefore, it is no surprise that some Japanese management practices are very similar to those of Chinese firms. For example, the Japanese seniority-based management system is deeply rooted in the principle of respecting the elderly people, which is also one of China’s core cultural values. Furthermore, Japan promotes lifetime employment, a practice quite similar to China’s ‘iron-rice bowl’, which has its hallmarks in various organisations, especially in the country’s state-owned enterprises.

Despite these similarities, cross-cultural researchers (Triandis, 1995; Triandis and Bhawuk, 1997) have denoted differences in the degrees of collectivism between China (horizontal collectivism) and Japan (vertical collectivism). To a certain extent, these differences may account for some of the differences in the cross countries’ supervisor–subordinate (S-S) relationships. For example, the Chinese S-S relationship emphasises subordinates’ rational commitment to and control by the leader, rather than their emotional ties, as found in Japan (Fukuda, 1993a; Silin, 1976). In addition, Chinese subordinates show little organisational loyalty (Alston, 1989) compared with their Japanese counterparts. The differences in subordinates’ loyalty targets between China and Japan can be explained by the influence of family orientation on Chinese subordinates, who stay loyal to supervisors who come from the same family line or clan (Lin, 1962). Japanese subordinates are society oriented in comparison. They believe their society is welded together like a piece of granite, and this belief drives them to stay loyal to their organisations. Regardless of the difference in loyalty orientation, both Chinese and Japanese supervisors not only hold power at the top and use personal criteria to distribute rewards and resources, but also emphasise the importance of collective work efforts (Fukuda, 1993b). Furthermore, similar to their Chinese counterparts, Japanese subordinates are expected to show gratitude for benevolent treatment in their total commitment and dedication to their supervisors (Whitehill, 1991).

In China, establishing an S-S relationship requires subordinates to cultivate guanxi with their supervisors by way of after-work social activities (Law et al., 2000). Given the salience of S-S guanxi, an increasing amount of empirical research has been conducted in the Chinese context (e.g. see Cheung et al., 2009; Cheung and Wu, 2011; Wei et al., 2010). However, little is known about how S-S guanxi operates and how it affects employee work outcomes in other countries, such as Japan.

This study contributes to the S-S relationship of cross-cultural literature in two important ways. First, this study tests the robustness of the mediation model used by Cheung et al. (2009). Second, the study also provides insights into the effect of S-S relationships on employee work outcomes through the mediating role of job satisfaction in the Japanese context.

Theory and hypotheses

Japanese management practices

There are two important management ideologies in Japan (Fukuda, 1993a). First, the Japanese advocate group orientation, in which the interests of a group are placed over those of the individual, and each individual is considered the body and soul of the group. Second, the Japanese also advocate community orientation, in which the company is concerned for the private lives of its employees and their work performances (Fukuda, 1993b). The Japanese have developed eight management practices based on these two ideologies (Fukuda, 1993a). First, they promote collective decision-making that allows employee participation and aims at unanimous decisions. Second, they emphasise group rather than individual duty and responsibility. For example, an individual’s job is usually vaguely defined. His or her duties are not spelled out clearly in written form, and his or her responsibilities are highly diffused. Third, the country’s lifetime employment practice guarantees employees jobs until retirement. Fourth, the Japanese practice seniority-based pay and promotion. This means that employees’ pay and promotions are contingent on their duration of company service rather than their work ability or performance. Fifth, Japanese companies provide comprehensive welfare programmes such as recreational facilities, medical and housing benefits and company loans to employees. Sixth, the Japanese prefer less detailed and restrictive contracts because they believe business terms are changeable and that flexibility in negotiations helps them cope with any unexpected issues. Last but not least, Japanese leaders provide training programmes to enhance employee work skills. They also practise job rotations, which allow employees exposure to different jobs and to become generalists rather than specialists.

S-S guanxi in China

S-S guanxi is defined as a critical interpersonal relationship between a supervisor and his or her subordinates in a Chinese organisational setting (Xu and Van de Vliert, 2006; Yan and Sorenson, 2004; Yang et al., 2005). As indicated by Yan (2004) and Xu (2006), the Chinese emphasise personalism, in which individuals pay a high amount of respect to superiors who are powerful and own valuable resources. S-S guanxi has been found to be related to supervisory administrative decisions (Law et al., 2000), subordinates’ trust in their supervisors (Han et al., 2012), career development (Wei et al., 2010), joint decision-making and open-minded discussions (Chen and Tjosvold, 2007), organisational citizenship behaviour (Liu and Wang, 2013; Wong et al., 2003) and organisational commitment (Cheung and Wu, 2011).

S-S kankei in Japan

Kankei describes the quality of the informal and emotional ties between Japanese supervisors and subordinates that focus on interpersonal harmony and social cohesion (Alston, 1989). Like their cultures, there are both similarities and differences in Chinese and Japanese S-S relationships. The key similarities are as follows. First, both Chinese guanxi and Japanese kankei supervisors are expected not only to help their subordinates learn a particular job’s social and formal requirements but also to look out for the interests of their subordinates, take responsibility for their subordinates’ errors and protect them from the discipline exercised by higher-ranking authorities (Cole, 1971). Second, the supervisors in both the countries rely on their subordinates to meet work goals, and encourage informal discussions with close subordinates before key decisions are made for the sake of reducing interpersonal conflicts (Clark, 1979; Fukuda, 1993a). Third, Japanese kankei subordinates and their Chinese counterparts usually expect protection from their supervisors if something goes wrong (Chen, 1995; Clark, 1979), and the subordinates are obligated to return the favour by working hard and displaying loyalty. Finally, both Japanese kankei and Chinese guanxi subordinates come to their supervisors in times of trouble or confusion in both their work and personal lives.

Nevertheless, there are several important differences between Japanese S-S kankei and Chinese S-S guanxi. The first difference lies in type of access. The use of kankei is to deny the access to outsiders and to grant access to the close kankei subordinates for retrieving valuable information. However, the use of guanxi is guided by rule of equity and norm of reciprocation. It usually favours the weaker party (e.g. subordinates) to have access to the rewards and support from the stronger party (e.g. supervisor).

The second difference lies in the nature of relationship. Japanese S-S kankei tends to be more emotional in nature and involves a concern for human feelings, dependence, morality and contractual obligations (Clark, 1979). Due to norms and unwritten law (Hirt and Schneider, 2003), the kankei subordinates are once committed to the supervisor, they will stay loyal to that supervisor. By contrast, Chinese S-S guanxi is more rational, emphasising an exchange of favours between a supervisor and subordinates with unspecified time and amount of reciprocation. Comparing to Japanese counterparts, Chinese subordinates perceive the guanxi relationships with their supervisor as less salient when they foresee the benefits of switching their loyalty.

The third difference is the time orientation. The Chinese guanxi does not have a strong sense of group connotation as Japanese kankei does. When Chinese subordinates believe that their guanxi relationship with their supervisor is no longer beneficial, these subordinates will readily move to a new company. However, kankei has a longer term time orientation than guanxi. Once Japanese subordinates are accepted as key kankei members, the supervisor will regard these subordinates as life long members and will protect their welfare by all means.

The last difference goes to the basis on which relational ties are formed. Researchers (see e.g. Fukuyama, 1995; Whitley, 1991) argued that there is subtle difference between social interaction in China and Japan. Despite the fact that both guanxi and kankei networks stress on social connections, Chinese guanxi is more family tie-oriented than Japanese kankei. Moreover, the relational ties in China between a Chinese supervisor and his/her subordinates are largely based on visible and common background. Once the relational bonds have been formed, the Chinese guanxi supervisors are considered the gatekeepers of all key resources and information. These supervisors exert tight personal control and have a low degree of delegation. In addition, these supervisors use an autocratic approach in decision-making. Unlike Chinese who primarily use Buddhism to frame the Confucius of thoughts, Japanese modified the doctrines of Confucianism to suit their own culture while keeping Buddhism as their main philosophy. Due to this modification, the forming of kankei relationships tends to be more idiosyncratic and situation specific. As a result, S-S kankei is largely formed on the basis of social factors rather than on the similarity in visible and common background (Xie, 2012) and Japanese supervisors facilitate decision-making rather than make the final decisions.

S-S kankei and job satisfaction in Japan

Owing to strong community orientation (Fukuda, 1993a), Japanese supervisors take on the responsibility of caring for the personal lives of their close kankei subordinates and their subordinates’ work performance. For example, kankei supervisors often provide job training to their subordinates and help them solve work-related problems. With the valuable resources and strong emotional support of their kankei supervisors, subordinates obtain constructive job feedback and are more equipped to accept challenging work assignments. As a result, the subordinates are more likely to enjoy their jobs.

Furthermore, the Japanese management practice of lifetime employment also facilitates a long-term, cooperative and trustful relationship between kankei supervisors and their subordinates. Owing to the lifetime employment practice, both Japanese supervisors and their kankei subordinates have higher initiatives to build cooperative and long-lasting relationships. As a result of these relationships, kankei subordinates are more willing to obey the work orders of supervisors and to avoid any interpersonal conflicts with supervisors at work. As highlighted by Fukuda (1993a), Japanese subordinates tend to be more satisfied with their jobs if they can use good kankei with their supervisors to obtain more resources and support and favourable (i.e. meaningful or autonomous) assignments. As such, we make the following hypothesis.

Hypothesis 1: S-S kankei is positively related to employee job satisfaction in Japan.

Job satisfaction and work outcomes in Japan

As stated by Fukuda (1993a), the lifetime employment practice may reduce an employee’s desire to job hop, particularly when the employee is satisfied with his or her job as a result of favourable assignments and attractive conditions. In this context, Japanese subordinates are indebted to repay the goodwill of their kankei supervisors by staying loyal to the organisations, particularly when comprehensive welfare package are provided and subordinates are kept employed in the midst of downsizing.

In addition, Japanese subordinates who are content with their work are driven by a strong community orientation to stay loyal to their organisations. These subordinates have strong feelings of dependence on and believe they are the souls of their organisations. If an organisation takes on the responsibility of taking care of its subordinates’ lifelong job security, subordinates should bear the duty of staying loyal to that organisation in return. Hence, if subordinates are given satisfying jobs that fully protect them from the potential risks of economic downsizing and redundancy, they will be encouraged to commit to their organisations.

As highlighted by Triandis and Bhawuk (1997), Japanese subordinates emphasise conservatism and harmony due to high collectivism. Employees may need to incur the high opportunity cost of losing job benefits if they quit. Hence, subordinates who are satisfied with their jobs have lesser intentions to leave. Aside from the opportunity cost, satisfying job conditions such as comprehensive job benefits and on-the-job training opportunities may help fulfil subordinates’ personal job achievement needs (George and Jones, 1996). Subordinates who have satisfactory jobs are motivated to put in greater time and effort to keep those jobs. Indeed, Luu and Hattrup (2010) found that individuals’ turnover intentions are strongly predicted by job satisfaction in high-collectivism countries such as Japan.

Furthermore, subordinates who are satisfied with their jobs are more likely to put extra effort into performing their duties. For example, they are more willing to work overtime or enrol in job-related courses to improve their job skills. By enhancing their job skills and improving their work performances, employees are more likely to get job promotions. Carr-Ruffino (1993) argued that an employee’s job promotion chances are increased if his or her outstanding work performance is recognised by management. As such, we make the following hypotheses.

Hypothesis 2a: Job satisfaction is positively related to organisational commitment.

Hypothesis 2b: Job satisfaction is negatively related to turnover intentions.

Hypothesis 2c: Job satisfaction is positively related to job promotions.

Mediating role of job satisfaction

The preceding hypotheses implicitly indicate that job satisfaction is a mediator between S-S kankei and an employee’s work outcomes. In good S-S kankei relationships, Japanese supervisors look out for the interests of their kankei subordinates, take responsibility for any errors made by their subordinates and protect them from the discipline exercised by higher-ranking authorities. These actions may make Japanese kankei subordinates more willing to reciprocate favours not to their organisations, but to their supervisors, who due to the good S-S kankei provide them with additional resources, valuable information and strong personal support at work. In addition, the nature of S-S kankei is emotional rather than rational. As suggested by Lin (1962), the relationship between Japanese supervisors and subordinates is largely characterised by human feelings, dependence and moral and contractual obligations. These characteristics influence the Japanese to develop strong group orientations, put the interests of a group over those of individuals and consider each individual the indentured body and soul of that group. Therefore, job satisfaction provides a leveraging mechanism that transforms the effects of S-S kankei into an employee’s work outcomes. Only when Japanese employees are happy with their current jobs will their kankei relationship with their supervisors be translated into positive work outcomes such as greater organisational commitment, higher promotion chances and lesser intentions to leave. Thus, we make the following hypotheses.

Hypothesis 3a: Job satisfaction mediates the relationship between S-S kankei and organisational commitment.

Hypothesis 3b: Job satisfaction mediates the relationship between S-S kankei and turnover intentions.

Hypothesis 3c: Job satisfaction mediates the relationship between S-S kankei and job promotions.

Methodology

Sample

One of the authors made personal contact with three manufacturing firms in Tokyo, Japan. After seeking the consent of these firms, their HR managers randomly selected 500 respondents using two criteria: they must have (1) worked with their supervisor for at least 3 months and were thus able to give a relatively clear judgment of the quality of their work relationship, and (2) been required to report to the same supervisor on a daily basis. The employee respondents were reminded of the voluntariness of participating in this survey and that their responses are highly confidential. In order not to affect daily work operations, the subordinates were grouped in different batches and invited to complete the questionnaires in a conference room. However, these subordinates were also reminded not to discuss with and/or disclose their responses to other subordinates and their supervisors.

Of these 500 respondents, 201 completed and returned the questionnaires, yielding a 40 percent response rate. Of the 201 respondents, 72.6 percent came from foreign-invested firms, 16.4 percent came from collectively owned firms and 6.5 percent came from private firms. The respondents comprised male (92.5 percent), female (7 percent), primary-educated (0.5 percent), secondary-educated (53 percent), college-graduate (6.6 percent), matriculated (19.2 percent) and university-graduate (20.7 percent) participants. Almost half of the respondents (44.2 percent) had worked with the same kankei supervisor for 1 year or less. The mean organisational tenure was 9.9 years. Prior to collecting the data for the formal survey, we ran a pretest with a sample of 20 employees from a Japanese firm. The information obtained from the pretests was used to rephrase some ambiguous items, ensure the clarity of the instructions and ensure that the latent variable measures were satisfactory.

Measurements

A 7-point Likert scale (1 = strongly disagree; 7 = strongly agree) was used for all the scales in the study. Other than the job promotions scale, we adopted all the measurements from an article by Cheung et al. (2009).

S-S kankei

The three items were ‘I share leisure activity interests (e.g. sports and movies) with my supervisor’ (Liden and Maslyn, 1998); ‘My supervisor has invited me to his or her home for a dinner, party or informal evening’ (Liden and Maslin, 1998) and ‘We often go to social activities together’. The Cronbach’s alpha of the S-S kankei scale was 0.88.

Job satisfaction

The 10-item job satisfaction scale used by Cheung et al. (2009) was originally adapted from the Illinois Job Satisfaction Index (Chernyshenko et al., 2003). One sample item included ‘I am satisfied with my work’. The Cronbach’s alpha of the scale was 0.84.

Organisational commitment

We adopted the 10-item scale from Cheung et al. (2009). An example item of emotional commitment is ‘I would be very happy to spend the rest of my career in this organization’. An example item of normative commitment is ‘Right now, staying with my job at this organization is a matter of necessity as much as desire’. An example item of instrumental commitment is ‘This organization desires my loyalty’. The Cronbach’s alpha of the scale was 0.89.

Intentions to leave

The 10-item intention scale was adopted from Cammann et al. (1979) used in Cheung et al. (2009). One sample item included ‘I am actively looking for another job’. The Cronbach’s alpha of this scale was 0.91.

Job promotions

We decided to use a new scale, job promotions, to replace the participatory management scale used in Cheung et al. (2009), as job promotions have been found to relate directly to S-S guanxi (Cheung and Wu, 2011). A five-item job promotions scale was borrowed from Law et al. (2000). One sample item included ‘I have the confidence to get promoted in the near future’. The Cronbach’s alpha of the job promotions scale was 0.70.

Control variables

Because the gender, education level, organisational tenure and company type variables have been found to exert influence on work outcomes (e.g. see Wei et al., 2010), they were controlled to minimise their confounding influence on job promotion estimates, organisational commitment and turnover intentions.

The questionnaire was translated from English into Japanese by one researcher and then back translated into English by another researcher to prevent meaning discrepancies (Brislin, 1979). To reduce the effects of socially undesirable responses, the participants were guaranteed anonymity and confidentiality. One research assistant familiar with our research helped answer the enquiries of respondents and collected data on site.

Validity and common method variance

We used a two-step approach to test validity (Murray et al., 2005). In the first step, exploratory factor analyses were examined to confirm the unidimensionality of the measurement items. An exploratory factor analysis with varimax rotation was performed on all five constructs (S-S kankei, job satisfaction, organisational commitment, job promotions and turnover intentions) to explore the underlying factor structure. The results revealed five factors with eigenvalues greater than 1, accounting for 55.89 percent of the total variation. All the items loaded cleanly on the expected factors, indicating no significant cross-loadings.

A confirmatory factor analysis was then tested on the measurement model using AMOS 16.0 with maximum likelihood estimation. A five-factor measurement model was tested against a four-factor model by merging job satisfaction and organisational commitment into a single factor, and another four-factor model by merging organisational commitment and intentions to leave into a single factor. These four factors were selected based on the high intercorrelations between job satisfaction and organisational commitment on one hand and between organisational commitment and intentions to leave on the other hand. The results indicated that the hypothesised five-factor model (χ ²/df = 2329.20/856, CFI = 0.97, TLI = 0.98, IFI = 0.93 and RMSEA = 0.08) had a better fit than the four-factor model (that merges organisational commitment and job satisfaction into a single factor) (χ ² = 2398.8, df = 857, CFI = 0.71, TLI = 0.70, IFI = 0.70 and RMSEA = 0.09), and another four-factor model (that merge organisational commitment and intentions to leave) (χ ² = 2738.7, df = 857, CFI = 0.62, TLI = 0.60, IFI = 0.62 and RMSEA = 0.10), the single-factor model (χ ² /df = 3330.70/860, CFI = 0.69, TLI = 0.67, IFI = 0.68 and RMSEA = 0.10). The difference in the χ ² change between the five-factor and single-factor models was significant (Δχ ²/df = 1001.50/4, p < 0.01), between the five-factor model and the first four-factor model (merging job satisfaction and organisational commitment into a single factor was also significant (Δχ ²/df = 69.6/1, p < 0.001) and between the five-factor model and the second four-factor model (merging organisational commitment and intentions to leave into a single factor) (Δχ ²/df = 409.5/1, p < 0.001). As such, the five-factor model was deemed acceptable. In addition, the composite reliability of each construct ranged from 0.71 to 0.91, all greater than the recommended threshold of 0.70. The average variance extracted (AVE) for each construct ranged from 0.51 to 0.73, above the recommended threshold of 0.50, providing support for convergent validity (Anderson and Gerbing, 1988; Fornell and Larcker, 1981). The discriminant validity was determined by comparing the squared correlation between two constructs and their AVEs. All the squared correlations were less than the AVEs, indicating sufficient discriminant validity (Fornell and Larcker, 1981).

The common method variance was tested using two methods. Harman’s one-factor test was used first, and the analysis resulted in a five-factor structure explaining 56 percent of the variance. No single factor was apparent in the unrotated factor structure. The first factor explained 28.86 percent of the total variance (Kaiser–Meyer–Olkin = 0.89, p < 0.001) and no single factor accounted for the majority of the variance. No general factor was apparent, and there was little common method variance to confound the interpretation of the findings. We also followed the recommendations of Podsakoff et al. (2003) closely to reduce the possibility of common method bias by inter-mixing the different construct items, and reduced the potential effects of response pattern biases by including negatively worded items. Hence, a common variance bias was not visible in this study.

Results

The means, SDs and correlations among variables are reported in Table 1. The means ranged from 2.68 to 3.71 with low SDs. As shown in this table, S-S kankei was positively related to job satisfaction (r = 0.33, p < 0.01), organisational commitment (r = 0.21, p < 0.01) and job promotions (r = 0.31, p < 0.01) and negatively related to turnover intentions (r = −0.21, p < 0.05). In addition, job satisfaction was positively related to organisational commitment (r = 0.71, p < 0.01) and job promotions (r = 0.53, p < 0.01) and negatively related to turnover intentions (r = −0.51, p < 0.01).

Table 1.

Means, SDs and intercorrelations.

Variables	Means	SD	1	2	3	4	5	6	7	8
1. Gender (female = 0)	1.07	0.27
2. Education	3.07	1.26	−0.17*
3. Organisational tenure	1.09	2.46	−0.04	0.74**
4. Company types	3.69	0.59	0.12	−0.35**	−0.26**
5. Supervisor–subordinate kankei	2.68	1.31	−0.01	−0.03	0.00	0.06
6. Job satisfaction	3.71	0.92	0.13	0.05	0.08	−0.05	−0.09
7. Organisational commitment	3.47	1.12	0.50	0.29**	−0.04	0.14*	−0.14**	0.21**
8. Intentions to leave	3.25	1.34	−0.07	−0.10	0.06	0.04	0.02	−0.21*	−0.51**
9. Job promotions	3.01	0.99	−0.16*	0.04	0.27**	−0.05	−0.19**	0.31**	0.53**	0.49**

Notes: n = 201; *p < .05, **p < .01.

We used structural equation modelling (SEM) to test the hypotheses. As indicated in Table 2, Model 2 (the partial mediation model) had an overall better fit (χ ² = 1962.40, df = 853, GFI = 0.97, CFI = 0.98, IFI = 0.98, TLI = 0.96 and RMSEA = 0.08) than Model 1 (the full mediation model) (χ ² = 1993.3, df = 856, GFI = 0.76, CFI = 0.77, IFI = 0.77, TLI = 0.75 and RMSEA = 0.10). Our SEM results showed that S-S kankei was both directly and indirectly related to employees’ organisational commitment, job promotions and turnover intentions through the mediating role of job satisfaction. These results are shown in Figure 1.

Figure 1.

Results of the final mediating model. Notes: n = 201, GFI = 0.97, CFI = 0.98, IFI = 0.98, TLI = 0.96, RMSEA = 0.08; *p < 0.05; **p < 0.01; ***p < 0.001.

Table 2.

Comparison of alternative models.

Model	χ ²/df	Δχ ²/df	GFI	CFI	IFI	TLI	RMSEA
Model 1. S-S kankei > job satisfaction > OC, job promotions and ITL	1993.30/856	30.90/3 (p < 0.001)	0.76	0.77	0.77	0.75	0.10
Model 2. S-S kankei > job satisfaction > OC, job promotions and ITL; S-S kankei and OC; S-S kankei and job promotions and S-S kankei and ITL	1962.40/853	–	0.97	0.98	0.98	0.96	0.08

Model 2: baseline model; S-S kankei: quality of supervisor–subordinate kankei; OC: organisational commitment; ITL: Turnover intentions. CFI = Comparative Fit Index; GFI = Goodness of Fit Index; TLI = Tucker-Lewis Index; IFI = Incremental Fit Index; RMSEA = Root Mean Square Error of Approximation.

Notes: n = 201.

As indicated in Figure 1 (the final model), the standardised path between S-S kankei and job satisfaction was positively related (β = 0.41, p < 0.001), lending support to H1. In addition, job satisfaction was positively related to job promotions and organisational commitment, and negatively related to turnover intentions (β = 0.43, p < 0.001), (β = 0.70, p < 0.001) and (β = −0.12, p < 0.001), respectively, giving support to H2a, H2b and H2c. Moreover, S-S kankei was also directly linked to job promotions (β = 0.17, p < 0.05), turnover intentions (β = -0.45, p < 0.001) and organisational commitment (β = 0.10, p < 0.05). Hence, H3a, H3b and H3c were supported, demonstrating that job satisfaction is a partial mediator in the relationship between S-S kankei and work outcomes.

Discussion

This study tested the robustness of the mediation model of Cheung et al. (2009) and investigated whether and how job satisfaction mediates the relationship between S-S kankei and employee work outcomes in Japan. Our results showed that job satisfaction partially mediates the links between S-S kankei and organisational commitment, turnover intentions and job promotions. As a result, the results were clearly quite similar to those found by Cheung et al. (2009) in the Chinese context.

The findings generally provide strong support for the role of job satisfaction in S-S relationships and work outcomes. Moreover, our results also testified to the robustness of the mediation model of Cheung et al. (2009). Although the Japanese have selectively drawn upon Chinese Confucianism, the similarity in the two countries’ collective cultures and management practices (e.g. providing full employment security) may explain why there are only minor differences between our results and those of Cheung et al. (2009).

In light of the similarity between the results of current study and those of Cheung et al. (2009), our results indicate that job satisfaction can be predicted by a good kankei relationship between Japanese supervisors and their subordinates. For example, Japanese subordinates can make use of good kankei to obtain additional support and resources that facilitate work and to bargain for more job related benefits, whereas supervisors can use it to motivate their subordinates to meet their high expectations by offering additional resources, information and personal protections at work. Also, the partial mediation is largely in line with our expectations. The findings show that the quality of S-S kankei can either be directly and indirectly related to subordinates’ organisational commitment, job promotions and turnover intentions via job satisfaction as the mediator. Good S-S kankei is not only a precondition for job satisfaction, but also encourages Japanese subordinates to commit to and stay with their organisations and to improve their chances of receiving promotions. As a result, good S-S relationships are important in both Chinese and Japanese contexts.

Our results also highlight that the cultivation of S-S kankei can be either directly or indirectly related to subordinate’s level of turnover intentions and organisational commitment through job satisfaction as a mediator. Since Japanese are society oriented, they believe society is welded together like a piece of granite, and this belief drives them to stay loyal to their organisations. A Japanese subordinate may simply consider his or her supervisor as a representative of the organisation. As a result, good S-S kankei can have a direct effect on an employee’s organisational commitment and intention to leave. Specifically, of the partial mediating role of job satisfaction in the relationship between Japanese S-S kankei and organisational commitment is in line with the findings of Cheung et al. (2009) that job satisfaction partially mediates the relationship between S-S guanxi and organisational commitment.

The indirect impact of S-S kankei on work outcomes through the mediating role of job satisfaction can also be partly explained by Japan’s prolonged economic depression and stagnation. To cut the increasing operational cost, some Japanese firms have begun to use a merit-based system in replacement of a lifetime employment system. Rather, some Japanese firm adopted an extensive use of reduced hiring and transfer of employees to related firms as alternative to lay-offs. Yet, Japanese culture has a strong emphasis on specific time dimension. Therefore, lifetime employment is still prevalent in many Japanese firms for the sake of attracting and retaining high calibre candidates. Due to the poor economic situation, the role of lifelong employment becomes less salient nowadays. Japanese employees of strong group connotation have begun to resort to developing close relationships with their supervisors to obtain job security and other work-related support, resulting in the direct effect of S-S kankei on their work outcomes.

Nevertheless, our results are slightly different from those of Cheung et al. (2009) in that a strong guanxi with supervisor alone is not sufficient to have a direct impact on subordinates’ intentions to leave the organisation because Chinese guanxi subordinates display personal loyalty towards the guanxi supervisor (Lin, 1962). Rather, the potential benefits of good guanxi need to be transformed into reduced intentions to leave via job satisfaction in China. But in the case of Japan, in addition to job satisfaction, there may be other mediators in the relationship between S-S kankei and intentions to leave.

Our study also offers two practical implications. First, given the important mediating role of job satisfaction, Japanese supervisors are encouraged to provide attractive, challenging, autonomous and meaningful jobs, improve working environments and extend comprehensive welfare benefits from employees to their family members or provide comprehensive training to enhance their subordinates’ work skills. Supervisors can also enhance their subordinates’ job satisfaction by changing their HR practices and spelling out duties and responsibilities clearly in written form. In addition, HR practitioners may consider encouraging Japanese subordinates to have regular job rotations that will improve their specific skills. After-hours social gatherings are also important for the development of S-S kankei, and Japanese firms should provide more socialisation opportunities for both supervisors and subordinates. Understanding the mechanism and cultural differences that affect the Japanese subordinates’ perception of interpersonal relationship with their supervisor are necessary for cross-cultural management. In the long-term, Japanese HR managers are advised to reduce the cost of lay-offs by redesigning the jobs that develop the potential in employees at work.

This study is subject to several limitations. While some of its constructs were employee rated, the results of a Harman’s one-factor analysis revealed no common method bias problem. Data were collected using only the cross-sectional method, limiting our examination of how S-S kankei influenced job satisfaction and, in turn, affecting employee work outcomes longitudinally. Since our study is cross-sectional in nature and thus we cannot directly rule out the reverse causal relationship from job satisfaction to S-S Kankei. Nevertheless, the mediation model is a good statistical design for testing the causal relationships (Baron and Kenny, 1986). Future research may use experimental or longitudinal design to directly test the causal relationship between job satisfaction and S-S kankei. In addition, future researchers may consider testing the robustness of the mediation model on other employee work outcomes, such as turnover rate and task performance. We also recommend future researchers to revalidate this model using a large sample size. In addition, many American management approaches have been implanted into Japanese organisations since World War II, such as the leader–member exchange (LMX), which is conceptualised as a work-related exchange relationship between a supervisor and his or her subordinate (Graen, 1976). Future researchers may find it useful to compare the respective roles of both LMX and S-S kankei and their effects on employee work outcomes and job satisfaction.

Conclusion

Our findings provide strong support for the robustness of the mediating model used by Cheung et al. (2009) by revealing that job satisfaction partially mediates the relationship between S-S kankei and employee work outcomes. Furthermore, the findings provide theoretical and managerial implications that are valuable to researchers who are making theoretical developments on S-S kankei, job satisfaction and work outcomes and to Japanese practitioners who are designing and implementing effective HR management practices.

Footnotes

Funding

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

References

Alston

(1989) Wa, guanxi, and inhwa: managerial principles in Japan, China, and Korea. Business Horizons 32: 26–31.

Anderson

Gerbing

(1988) Structural equation modelling in practice: a review and recommended two-step approach. Psychological Bulletin 103(3): 411–23.

BBC News (2011) China overtakes Japan as world’s second-biggest economy. 4 February 2011.

Baron

Kenny

(1986) The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51: 1173–82.

Brislin

(1979) Back translation for cross-cultural research. Journal of Cross-Cultural Psychology 1: 185–216.

Cammann

Fichman

Jenkins

Kiesh

(1979) The Michigan Organisational Assessment Questionnaire. Unpublished manuscript, University of Michigan, Ann Arbor, Michigan, USA.

Carr

Markus

(2006) Convergence of management practices in strategy, finance and HRM between the USA, Japan, and Germany. International Journal of Cross Cultural Management 6: 75–100.

Carr-Ruffino

(1993) The Promotable Woman. Belmont, CA: Wadsworth Publishing Company.

Chen

(1995) Asian Management Systems: Chinese, Japanese and Korean Styles of Business. London and New York: Routledge.

10.

Chen

Tjosvold

(2007) Guanxi and leader member relationships between American managers and Chinese employees: open-minded dialogue as mediator. Asia Pacific Journal of Management 24(2): 171–96.

11.

Chernyshenko

Stark

Crede

Wadlington

Lee

(2003) Improving the measurement of job attitudes: the development of JSI. In: Rup

(ed) New Frontier in Job Satisfaction, Job Performance, and Their Linkage. Symposium conducted at the 18th Annual Meeting of the Society for Industrial and Organisational Psychology, 11-13 April 2003, Orlando, Florida.

12.

Cheung

(2011) Participatory management and employee work outcomes: the moderating role of supervisor–subordinate guanxi. Asia Pacific Journal of Human Resources 49(3): 344–64.

13.

Cheung

Chan

Wong

(2009) Supervisor–subordinate guanxi and employee work outcomes: the mediating role of job satisfaction. Journal of Business Ethics 88: 77–89.

14.

Clark

(1979) The Japanese Company. New Haven, CT: Yale University Press.

15.

Cole

(1971) Japanese Blue Collar: The Changing Tradition. Berkeley/ Los Angeles/ London: University of California Press.

16.

Fornell

Larcker

(1981) Evaluating structural equation models with unobservable variables and measurement errors. Journal of Marketing Research 18: 39–50.

17.

Fukuda

(1993a) Japanese management at the crossroads. In: Fukuda

(ed) Japanese Management in East Asia and Beyond. Hong Kong: Chinese University Press, pp. 139–170.

18.

Fukuda

(1993b) Japanese and Chinese management practices: uncovering the differences. The Mid-Atlantic Journal of Business 21(2): 35–44.

19.

Fukuyama

(1995) Trust: Social Virtues and the Creation of Prosperity. New York, NY: Free Press.

20.

George

Jones

(1996) The experience of work and turnover intentions: interactive effects of value attainment, job satisfaction, and positive mood. Journal of Applied Psychology 81: 318–25.

21.

Graen

(1976) Role-making processes within complex organizations. In: Dunnette

(eds) Handbook of Industrial and Organizational Psychology. Chicago, IL: Rand, pp. 1201–45.

22.

Han

Peng

Zhu

(2012) Supervisor–subordinate guanxi and trust in supervisor: a qualitative inquiry in the People’s Republic of China. Journal of Business Ethics 108(3): 313–24.

23.

Hirt

Schneider

(2003) Risk and motivation in Sino-Australian joint-venture in China. In: Haak

Hilpert

(ed) Focus China: The New Challenge for Japanese Management. Iudicium Verlag.

24.

Law

Wong

Wang

(2000) Effect of supervisor–subordinate guanxi on supervisory decisions in China: an empirical investigation. International Journal of Human Resource Management 11(4): 751–65.

25.

Liden

Maslyn

(1998) Multidimensionality of leader-member exchange: an empirical assessment through scale development. Journal of Management 24(1): 43–72.

26.

Lin

(1962) My Country and My People. London, UK: Heinemann.

27.

Liu

Wang

(2013) Abusive supervision and organisationalcitizenship behavior: is supervisor–subordinate guanxi a mediator? The International Journal of Human Resource Management 24(7): 1471–89.

28.

Luu

Hattrup

(2010) An investigation of country differences in the relationship between job satisfaction and turnover intentions. Applied Human Resources Management Research 12(1): 17–39.

29.

Magoshi

Chang

(2009) Diversity management and the effects on employees’ organisational commitment: evidence from Japan and Korea. Journal of World Business 44(1): 31–40.

30.

Minkov

Hofstede

(2011) The evolution of Hofstede’s doctrine. Cross-Cultural Management: An International Journal 18(1): 10–20.

31.

Murray

Kotable

Zhou

(2005) Strategic alliance-based sourcing and market performance: evidence from foreign firms operating in China. Journal of International Business Studies 36(2): 187–208.

32.

Podsakoff

MacKenzie

Podsakoff

(2003) Common method biases in behavioral research: a critical review of the literature and recommended remedies. Journal of Applied Psychology 88: 879–903.

33.

Silin

(1976) Leadership and Values: The Organisation of Large-Scale Taiwanese Enterprises. Cambridge, MA: Harvard University Press.

34.

Triandis

(1995) Individualism and Collectivism. Boulder, CO: Westview Press.

35.

Triandis

Bhawuk

DPS

(1997) Culture theory and the meaning of relatedness. In: Earley

Erez

(eds) New Perspectives on International Industrial/Organisational Psychology. San Francisco, CA: New Lexington Press, pp. 13–52.

36.

Wei

Liu

Chen

(2010) Political skills, supervisor–subordinate guanxi and career prospects in Chinese firms. Journal of Management Studies 47(3): 437–57.

37.

Whitehill

(1991) Japanese Management: Tradition and Transition. London and New York: Routledge.

38.

Wong

Ngo

Wong

(2003) Antecedents and outcomes of employees’ trust in Chinese joint ventures. Asia Pacific Journal of Management 20(4): 481–99.

39.

Whitley

(1991) The social construction of business systems in East Asia. Organisation Studies 12(1): 1–28.

40.

Xie

(2012) Cooperation, relationship, and evolutionary process in institutions. Available at: http://www.jcer.or.jp/eng/pdf/Zhigang2012eng.pdf

41.

(2006) Job formalisation and cultural individualism as barriers to trust in management. International Journal of Cross Cultural Management 6(2): 221–42.

42.

Van de Vliert

(2006) Job formalization and cultural individualism as barriers to trust in management. International Journal of Cross Cultural Management 6(2): 221–42.

43.

Yan

(2004) The influence of Confucian ideology on conflict in Chinese family business. International Journal of Cross Cultural Management 4(1): 5–17.

44.

Yan

Sorenson

(2004) The influence of Confucian ideology on conflict in Chinese family business. International Journal of Cross Cultural Management 4(1): 5–17.

45.

Yang

Van de Vliert

Shi

(2005) Siding in a workplace dispute in China: the impact of legitimacy, sanction, and guanxi. International Journal of Cross Cultural Management 5(3): 329–47.

46.

Zhu

Warner

(2000) An emerging model of employment relations in China: a divergent path from the Japanese? International Business Review 9: 345–61.