HIGH PERFORMANCE WORK SYSTEM AND ORGANIZATIONAL CITIZENSHIP BEHAVIOR IN MULTINATIONAL COMPANIES IN VIETNAM: THE MEDIATION EFFECT OF CAREER SUCCESS

In this study, the relationship between high performance work system (HPWS) and organizational citizenship behavior (OCB) in multinational corporations (MNCs) in Vietnam, a fast-developing country with highly economic growth in Asia, was investigated. Besides, the underlying mechanism of this relationship was also explored. From the social exchange approach, an underlying mediated mechanism of career success in the relationship between HPWS and OCB was hypothesized. Data collected from individual employees working in multinational companies in Vietnam was used to test the hypotheses. The research findings supported the partial mediating role of career success in the positive relationship between HPWS and OCB. In addition, research implications as well as suggestions for future research were also


1.
Introduction HRM literature has witnessed a lot of studies exploring the impact of separate components of HRM system on employee behaviors at individual level (Boselie et al., 2005;Wright & Boswell, 2002). However, some recent HRM scholars argued that HRM practices need to be treated as a set or bundle of individual HRM practices rather than single practices to assess their influence on employee outcomes (Gould-Williams & Mohamed, 2010;Snape & Redman, 2010a). According to the definition of Combs et al. (2006), HPWS is "a system of human resource practices designed to enhance employees' skills, commitment, and productivity in such a way that employees become a source of sustainable competitive advantage" (Datta et al., 2005). A coherent "system" or "bundle" (Delery & Doty, 1996;MacDuffie, 1995) of HRM practices which can bring to higher effectiveness, as in the synergistic relationship, any HR practice can support and enhance the effectiveness of one another; therefore, the integrated HR practices generate more effect than the accumulated effect of separate component of HR practice (Becker et al., 1997;Ichniowski et al., 1997). This recommends that employee perceptions of HR systems rather than single HR practices may be appropriate for identifying employee behaviors and attitude Guest, 2004;Paauwe, 2009). Hence, there is a need to study the effect of the whole HR practice system on employees' behaviors in the organization.
Besides, the recent years have witnessed increasingly the emphasis on the importance of organizational citizenship behavior (OCB), which is employees' behaviors and actions beyond the basic requirements of the job. Recently empirical HRM research indicated the positive relationship of HPWS with employee attitude and behaviors such as OCB, which resulted to organizational performance (Delery, 1998;Wright et al., 1994). However, there exists a debate on the underlying mechanism through which HPWS impacts on OCB. Some recent studies have paid more attention to explore mediators such as organizational commitment, job satisfaction, employee engagement, career success in the positive HPWS -OCB relationship; the results though have been varied. In addition, career success, especially subjective career success, which focuses on employee selfevaluation toward their work that would be beneficial to their own prosperity, is still not fully received attention as mediator variable in the influence mechanism of HPWS on OCB. Hence, based on social exchange theory, this study aims to explore the mediating mechanism through which HPWS positively impacts OCB in order to fill this research gap. By collecting responses from Vietnamese employees working in MNCs in Hanoi and Hochiminh, the study further proposes an individual framework in which career success mediates the positive HPWS -OCB relationship.

2.
Literature Review 2.1 The relationship between high performance work system and organizational citizenship behaviors Social exchange perspective gives a conceptual framework to explain how employee perceptions of HR system are related to employees' attitude and behaviors. Social exchange is based on imperfectly specific terms and norms of reciprocity (Blau, 1964). Social exchange perspective shows that "discretionary benefits offered to the exchange partner are paid back in a discretionary way in the longer term" (Blau, 1964;Rhoades & Eisenberger, 2002). If employees receive socio-emotional and economic benefits from their organizations, they will have positive attitude towards their career in term of career satisfaction, the competitiveness in both internal and external working opportunities, because they believe that their competency, knowledge, and skills are improved by the organizational investments. In addition, they will return discretionary role behaviors to their organizations. Hence, HR practices may be viewed as an input into the social exchange process (Snape & Redman, 2010a). HR systems can develop environment which can encourage or even discourage OCB (Morrison, 1996). When employees receive long-term investment from organizations through HR practices, they feel obligated to reciprocate with discretionary role behaviors that are typically conceptualized in terms of OCB (Gong et al., 2010;Shaw et al., 2009;Sun et al., 2007). As Hannah and Iverson (2004) asserted "HR practices are viewed by employees as a "personalized" commitment to them by the organization which is then reciprocated back to the organization by employees through positive attitudes and behavior".
Sharing the same stance, Leana and Van Buren (1999) indicated that HR practices foster the development of highquality exchange relationship, leading employees assume the role of good organizational agents. Indeed, they stressed that "if individuals believe that their efforts are an integral part of a collective, they are more likely to spend time doing things the organization and/or its members find useful (e.g., engaging in extra-role behavior or working cooperatively) and less time doing things benefiting the individual but not the organization (e.g., social loafing, selfpromotion, or unwillingness to cooperate)" (Leana & Van Buren, 1999). Further, within employee-organization relationship, the mutual investment reflected in HPWS will stimulate employees to perform out of their main responsibility and also offer their citizen behaviors to the organizations (Tsui et al., 1997).
HPWS offers employees with multiple social resources such as appreciation, recognition, prestige, fairness, growth, and empowerment, which fosters social exchange relationship with employees. Thanks to HPWS, employees are now likely to perceive investment from their organizations and reciprocate by expanding their job roles and engaging more in OCB (Morrison, 1996). In agreement with this argument, the findings of (Sun et al., 2007) from cross-level analysis reported that high performance HR practices were positively related to service-oriented OCB. Moreover, based on a collective social exchange approach, a study conducted on the research samples of the middle managers at group level in China enterprises found that HPWS positively influenced OCB through the intervening role of collective affective commitment (Gong et al., 2010). As a result, we predict that employees with positive perceptions of their HPWS will demonstrate more OCB and hypothesize as follows: H 1 : HPWS is positively related to OCB of employees in MNCs in Vietnam.

2.2
The mediating role of career success mediates in the relationship between high performance work system and organizational citizenship behaviors Some empirical studies have the findings supporting for the notions that HR practices have positive impact on individual and organizational outcomes; however, more recent SHRM scholars have sought to investigate the mechanism linking between HR practices and individual as well as organizational outcomes. Further, Guest (1997) proposed a theoretical model on HR practicesperformance relationship and suggested that employee awareness in term of HR practices result in attitudinal outcomes, which consecutively produce behavioral outcomes such as OCB (Guest, 1997). For example, organizational commitment and job satisfaction are considered as these attitudinal variables linking HR practices with employees' behaviors. Although the empirical research asserted the positive link between high performance HR practices and OCB (Alfes et al., 2013;Gong et al., 2010;Snape & Redman, 2010a;Sun et al., 2007), this link likely relies on some intervening attitudinal explanation. Particularly, the relationship between HPWS and OCB can be clarified with the process through the influence of HPWS on career success by social exchange perspective (Guan et al., 2015;Ng & Feldman, 2014;Simo et al., 2010).
Explaining by the view of social exchange theory, when employees perceive the organizational investment, like HPWS, in heightening their abilities, knowledge, and motivation, they are more likely to pay back with a good attitude toward their work and their competency; consequently, they will take initiatives in OCB. One of the most common overall actions organizations have taken in response to pressing workplace trends is investing more in supporting training and development to boost skill levels of employees.
Likewise, employee development has become something that is very important to workers as part of their careers and working lives. The career success helps to sustain behavioral responses in the absence of rewards. Consequently, employees who are positive evaluation towards their career are willing to contribute more in the direction of the organizational objectives accomplishment. Moreover, they are also more motivated than others to exert considerable efforts in extra-role performance to further organizational effectiveness, even if they do not expect to be directly compensated for this behavior on the basis of formal HPWS. Moreover, there are some previous studies indicated that career success has the positive relationship with perceived organizational supports and is significant with organizational outcomes (Shockley, Ureksoy et al., 2016). Thus, we expect that there exists the mediated role of career success in the HPWS-OCB relationship. According to Baron and Kenny (1986) research on testing the role of mediating variable, these below hypotheses need to be tested:

Research Method 3.1 Sample and Procedures
In this study, the 220 usable data was collected by the convenience sampling method. The questionnaires were sent to the target respondents who were employees working in MNCs in two biggest cities in Vietnam including Hanoi, Hochiminh.
The information of participants in this research was assured to be confidential and anonymous. The questionnaire firstly was developed in English and was then independently translated into Vietnamese by two Vietnamese bilingual academics who were in charge of conventional translation and back-translation. Then, the questionnaires were reviewed by two HR managers of MNCs to test the accuracy of the professional terms and perspicuity. Then the mass survey was conducted to collect the data from the target populations. After collecting and cleaning the data, SPSS 22.0 software was used to analyze the data with following steps: testing reliability of scale and validity of questionnaire through Cronbach's alpha coefficient, analyzing correlation, and regression to test mediation.

Measures
To measure both independent and dependent variables, a questionnaire of multiple items was used. The target respondents were asked to evaluate their opinion about the statements with five levels of disagree-agree Likert scale. After testing the reliability of scales, all of the scales achieved an alpha reliability of high value more than .90 (HPWS: α=.938, career success: α= .932, and OCB: α= .904). This result is a consequence of a well-designed, clear questionnaire, wellgrouped, and good samples (Hair et al., 1998).

High Performance Work System
A 27-item HPWS scale adapted from Sun et al. (2007) was used to measure HPWS. There are eight subscales in this set of scale (including selective staffing, extensive training, internal mobility, employment security, clear job description, results-oriented appraisal, incentive reward, and participation). Respondents were required to choose option among 05 points with one representative for "strongly disagree" and five representatives for "strongly agree" for evaluating statement on all of the aspects of HPWS practices. Some item examples are "Employees in customer contact jobs will normally go through training programs every few years", "Employees in this job are often asked by their supervisor to participate in decisions", or "Employees do not have any future in this organization (Reversecoded)".

Organizational Citizenship Behavior
Following Organ (1988Organ ( , 1990) and Podsakoff et al. (2000), a set of 22 items with a response scale ranged from 1 -"strong disagree" to 5 -"strongly agree" was used to measure employees' OCB. In the measurement scales, there were some reversed items used such as "I consume a lot of time complaining about trivial matters", "Attends functions that are not required, but help the company image", "I always ready to lend a helping hand to those around him/her". OCB was divided into five dimensions, including civic virtue, sportsmanship, altruism, courtesy, and conscientiousness (Organ, 1988(Organ, , 1990Podsakoff et al., 2000)

Career Ruccess
There were three dimensions measured by 11 scale items. Five-item scale was used to measure perceived career success adapted from Greenhaus et al.
(1990). Perceived external marketability and perceived internal marketability were measured by scale adapted from Eby et al. (2003). The anchors for these items ranged from 1strongly disagree to 5 -strong agree. In which, there are some items such as "I am satisfied with the progress I have made toward meeting my goals for advancement", "There are many jobs available for me given my skills and experience" (Eby et al., 2003).

Results and Discussion 4.1 Results
Prior to the linear regression analysis, the Pearson correlation coefficient (r) is examined with the intention of evaluating the correlations among variables. The analysis results in a correlation matrix among variables are shown in

Mediation Analysis
To examine the mediating role of career success in the relationship between HPWS and OCB, the four-step regression analysis was used (Baron & Kenny, 1986;James & Brett, 1984;Judd & Kenny, 1981) to identify whether career success enhances the effect of HPWS on OCB or not. According to this method, there are three conditions need to achieve. Firstly, the causal and effect relationship between independent variable (HPWS) and dependent variable (OCB) is tested. Secondly, the correlation relationship between independent variable (HPWS) and mediating variable (career success) must be significant. The relationship between mediating variable (career success) and dependent variable (OCB) also must be demonstrated. Finally, the multiple regressions between independent variable (HPWS), mediating variable (career success) and dependent variable (OCB) is investigated to identify the role of mediating variable. In case the influence of independent variable (HPWS) on dependent variable (OCB) is reduced but still significant, this result fits with partial mediation model. On the other hand, the relationship between independent variable (HPWS) on dependent variable (OCB) is not significant anymore; the result fits with full mediation model.
To analyze the role of mediating variable, the SPSS PROCESS is employed (Preacher & Hayes, 2004). The results of standard regression analysis are shown in table 2 below. As shown in table 4.2, the result of linear regression showed that the coefficient of determiner R2 (R-Square) is 0.2436 at the significance of Sig. = 0.000 < 0.05, indicating that the regression model was appropriate. The value of Rsquare is 0.2436 which means that 24.36% of the variance in OCB is explained by HPWS. This model suggests that HPWS plays an important role in strengthening OCB with b = 0.609 and p < 0.01. Upon the result, Hypothesis 1 is supported at the level of significance of 1% or the confidence level is above the 99%. This finding is also supported by studies conducted by Morrison (1996), Sun et al., (2007), and Gong et al., (2010).
Hypothesis 2: HPWS is positively related to career success of employees in MNCs in Vietnam. The result of model 4 showed that the coefficient of determiner R2 (R-Square) is 0.5639 and b = 0.8791 at the significance of Sig. = 0.000 < 0.01, indicating that the regression model was appropriate. The value of R-square means that 56.39% of the variance in career success is explained by HPWS. The results support the hypothesis 2 at the level of significance of 1%. This finding is also aligned with previous studies (Simo et al., 2010). Hypothesis 3: Career success is positively related to OCB of employees in MNCs in Vietnam.
Using the linear regression analysis to test the hypothesis 3, the result is presented in model 2 (Table 4.2). The Rsquare is 0.2114 and b = 0.3183 at the significance of Sig. = 0.000 < 0.01. That the adjusted R-Square was 0.2114 meaning that the linear regression models were constructed 21.14% in accordance with the data. In other words, 21.14% of the differences of observed level of OCB can be explained by the difference in the career success. The value of the coefficient is positive indicating the positive relationship between career success and OCB. Thus Hypothesis 3 is supported. Hypothesis 4: Career success mediates the relationship between HPWS and OCB of employees in MNCs in Vietnam.
The hypothesis 1, 2, 3 are supported and thereby satisfying two first requirements for mediation examination. The final condition is that the multiple regression analysis between the independent variable (HPWS), mediating variable -career success and the dependent variable (OCB) is tested to identify the role of mediating variable. When both of independent and mediating variables are put into the regression equation, as shown in Table 4.2, results in model 3 reveals that when both independent variables -HWPS and mediating variables -career success are simultaneously entered into the model, HPWS is significantly related to OCB at b = 0.2756, Sig. = 0.000 < 0.01. Besides, career success is still significantly related to OCB at b = 0.1416 and sig.< 0.05. Comparing to the value of the coefficient in model 1, the strength of the relation between HPWS and OCB is reduced (0.1416 < 0.4001). These results indicate that career success partly mediates the HPWS -OCB relationship. Hence, hypothesis 4 is supported. The figure 2 provides the conceptual model that was tested above.

Discussion
The primary objective of this study was to examine the relationship between HPWS and OCB of employees working in MNCs in Vietnam. Besides, this study attempted to explore the "black-box" of this relationship, the potential mediating role of career success. The findings of this research clarified the relationships among these factors. The result of this study also supported the previous research.
Firstly, this study indicated that the HPWS is positively related to OCB of Vietnamese employees in MNCs. In the other word, the more HPWS, a set of HR practices which comprise "selectively staffing", "extensive training", "clear job description", "result-oriented appraisal", "incentive reward", and "participation", organizations invest in; the more OCB of employees is enhanced. From the reciprocate approach of social exchange theory, when the employees receive the investment from their organization through HPWS, they have a tendency to react with more beyond expected behaviors as well as engage more in OCB. This finding is consistent with previous research and reinforces for the prior theories of strategic HRM (García-Chas et al., 2016;Gong et al., 2010;Muduli et al., 2016;Sun et al., 2007;Way & Johnson, 2005;Wright & Nishii, 2007). Secondly, the research findings supported the partial mediation role of career success in the positive relationship between HPWS and OCB. It means that HPWS influences OCB partially through career success of employees. The psychological outcomes of employees towards their jobs play intervening role in the path HPWS transfer the impact on employees' behaviors. This finding contributes to the opening the black-box of HPWS. In addition, the partial mediation role of career success in the positive HPWS -OCB relationship also suggests a different view for both theoretical and practical implications. For managerial implication, the manager should not only pay attention to the HR practices to enhance the employee's outcomes but also take consideration of the intrinsic factors of the employees. Boosting the positive and subjective evaluation of employees towards their career path is a way to gain the valuable outcomes within the organization.
Thirdly, this study along with some recent research that was conducted in Asian countries like China, India. (Gong et al., 2010;Jiang & Liu, 2015;Muduli et al., 2016;Snape & Redman, 2010b;Sun et al., 2007), provides sufficient evidence against the assumption that strategic HRM with integrated HR practices is only applicable in Western context. This finding contributes to extending the generalizability and applicability of Western theories to the Asian context. Thus the HPWS theory is completely considered as a universal theory that can be applied in multinational context especially in some emerging and developing economy like Vietnam.
Although the presented study basically accomplished the research objectives is investigating on HPWS in multinational companies in Vietnam. However, it certainly contains several limitations because of the limited resources and competencies. Firstly, it is the limitation of sample size. This study tested the hypotheses based on data collected from individual employees in two largest cities in Vietnam (Hanoi City and Hochiminh City). Therefore, future research may take a larger and broader sample with the more reliable sampling method to enhance the general ability and robustness of the proposed model. The second limitation is the validity of collected data. In this study, a crosssectional study was conducted by obtaining data at one point in time, which can limit our conclusion related to causeeffect relations inferred from our results. Moreover, the employees were required to evaluate both independent and dependent variables, which can bring to possible common method bias. However, to avoid common method bias, we designed questionnaire with the order of independent variables firstly assessed and then followed by dependent variables. Nevertheless, we still encourage future research might be based on a longitudinal design with multilevel analysis to explore our findings more. In addition, this study only focused on exploring mediating mechanism through which HPWS practices influenced the collective OCB, ignored contextual factors which may moderate HPWS-OCB linkage, and thereby not explore black-box fully. Takeuchi et al. (2007) indicated that "Previous empirical research has encountered difficulties when examining the moderating effects of contextual factors such as business strategy on the HR system-firm performance relationship" (Takeuchi et al., 2007,).
Hence, future research may explore the other factors as mediators of the causality relation and contextual factors as moderators of mediating relationship.

Conclusion
In conclusion, the research results have contributed to open the "black-box" of influence mechanism of HPWS on employee behaviors. Particularly, the study introduced and explored career success as a mediator of positive HPWS -OCB relationship. This finding is consistent with the central propositions of social exchange theory, which suggests that organizations that can invest in their employees through HPWS practices will be likely paid back positive behavioral outcomes from their employees.