Impact of Repatriate’s Knowledge Transfer on Enterprise Performance: The Mediating Effect of Ambidexterity Innovation

2.1 Knowledge Transfer and Ambidexterity Innovation

Explicit knowledge transfer is process that repatriates in multinational corporation transfer and share coded, written and expressed knowledge. Explicit knowledge is easy to storage and transmission, and can be saved in structured knowledge base, which is advantageous to repatriate’s query and extraction. In order to accelerate knowledge reuse, multinational corporation must facilitate repatriate to effectively transfer explicit knowledge in organization, facilitate knowledge integration and creation, and then exploitative innovation can be occurred. For example, Kauppila[6] proposed that explicit knowledge transfer is favor to exploitative innovation. A similar point is made by Hernández-Espallardo, et al.[7], who study the influencing of distributors knowledge acquisition on innovation, and get the conclusion that knowledge acquisition has positively impact on exploitative innovation. Therefore, explicit knowledge transfer is positively related to exploitative innovation. In the process of exploratory innovation, it is very important for multinational corporation to expand the knowledge width of innovation network, enhance knowledge heterogeneity. Since, repatriates acquire many kinds of explicit knowledge in the international market, have a broad knowledge and technology scope, and can comprehensive grasp core knowledge and technology. Repatriates have the capability to explore innovation path so that multinational corporation can enter wholly new field. Repatriates transfer explicit knowledge that is more heterogeneity, the more easily to generating new innovative ideas. As a result, explicit knowledge transfer has a significant effect on exploratory innovation. Based on this, we put forward the following hypothesis:

H1a: Explicit knowledge transfer is positively related to exploitative innovation.

H1b: Explicit knowledge transfer is positively related to exploratory innovation.

Tacit knowledge transfer is a process that repatriates transfer and share knowledge which is not easy to express, highly individual and difficult to standardization. Exploitative innovation, which mainly pay attention to the extension of existing technology and knowledge, emphasize the optimization for current design, marketing channels and skills, and the innovation aim is to meet the needs of existing customers and market. Repatriates, who transfer tacit knowledge in organization, can facilitate members in multinational corporation to share and apply knowledge and know-how, and can help members to comprehensive understanding existing knowledge. In addition, repatriates who transfer tacit knowledge can facilitate members to teach each other about how to integrate knowledge, promote knowledge integration. Tacit knowledge transfer can improve existing knowledge from different ways, thus, boost exploitative innovation. Exploratory innovation is the activities that acquire and create wholly new knowledge, and try to exceed the existing knowledge base. Kauppila[6] suggested tacit knowledge transfer is benefit to exploratory innovation, and Cavusgil, et al.[8] found tacit knowledge transfer is advantage to enterprise’s product innovation. Subramaniam and Venkatraman[9], in the study of multinational corporation in the United States, found the capability transfer and use of tacit knowledge is stronger, and the innovation is also higher. Therefore, tacit knowledge transfer will promote exploratory innovation activities. Based on this, we put forward the following hypothesis.

H2a: Tacit knowledge transfer is positively related to exploitative innovation.

H2b: Tacit knowledge transfer is positively related to exploratory innovation.

2.2 Ambidexterity Innovation and Enterprise Performance

Exploitative innovation makes the multinational corporation better use of existing resources, can achieve the potential value of the existing resources, and can generate high performance. While, exploratory innovation can help multinational companies to find new markets, help enterprises in a timely manner to adjust its production process, so as to accurately grasp all kinds of market opportunities, and find the potential market demand that consumers themselves have not found, accordingly, improve enterprise performance. He and Wong[10] suggested that exploitative and exploratory innovations have relationship with higher sales growth. In addition, Benner and Tushman[11–14], according to the difference of innovation degree and knowledge base, put forward exploitative innovation and exploratory innovation, and think the key factors for organization’s acquiring performance and maintaining sustainable competitive advantage are exploitative and exploratory innovation. Similarly, Auh and Menguc[15] empirically study shows whether the performance is an effect index or efficiency index, exploitative and exploratory innovation all show a significant positive on it. At the same time, they adopt comprehensive performance indicators, and get the conclusion that the two kinds of innovation have positive effect on the enterprise performance by empirical study[16]. Therefore, we put forward the following hypothesis:

H3: Exploitative innovation is positively related to enterprise performance.

H4: Exploratory innovation is positively related to enterprise performance.

2.3 The Mediating Role of Ambidexterity Innovation

Repatriates have rich overseas experience, and possess the knowledge and ability to respond to the international environment change, through knowledge transfer, enterprise can gain more opportunities and method to solve the problem. Therefore, explicit and tacit knowledge transferred by repatriates is benefit to the multinational corporation taking some innovation, and improve enterprise performance. Report and technical documentation transferred by explicit knowledge, can be assimilated by members in multinational corporation, which can not only improve efficiency of communication and coordination, but also shorten product’s development cycle, decrease product’s research and development cost, as such enterprise performance can be improved. In addition, explicit knowledge transfer can make multinational corporation to rapidly response environment need, and decrease uncertainty of product development. For example, Schulz[17] thought that explicit knowledge transfer can bring potential profit for organization. Therefore, repatriate’s explicit knowledge transfer is helpful to promote the development of exploitative and exploratory innovation, and thus enhanced the performance of multinational corporation.

According to Jaffe, et al.[18], knowledge transfer, especially tacit knowledge transfer, is the basis of innovation. Zahra and George[19] pointed out that innovation by tacit knowledge transfer has path dependence, difficult to imitation, and new product development by innovation often show better market performance. Therefore, repatriate’s tacit knowledge transfer can promote exploitative innovation and exploratory innovation, form a new technology and new solution, so as to solve the technical problems smoothly in the process of product development, consolidate new products’ technology capability and market development ability, improve enterprise performance. Based on the above analysis, we can conclude that repatriate’s knowledge transfer has a positive effect on innovation, and innovation has a positive influence on enterprise performance, that is to say, innovation plays an intermediary role between repatriate’s knowledge transfer and enterprise performance. Based on this, we put forward the following hypothesis:

H5: Exploitative innovation mediates the relationship between explicit knowledge transfer and enterprise performance.

H6: Exploitative innovation mediates the relationship between tacit knowledge transfer and enterprise performance.

H7: Exploratory innovation mediates the relationship between explicit knowledge transfer and enterprise performance.

H8: Exploratory innovation mediates the relationship between tacit knowledge transfer and enterprise performance.

According to the above discussion, a theoretical model is established, see in Figure 1.

Figure 1

Theoretical model

3.1 Sample

This study obtain data by survey, the sample are multinational corporations located in Shaanxi, Beijing, Shanghai, Guangdong, Sichuan and other regions of China. Formal questionnaire of this study are issued during August 2014 to October, and three kinds of distribution channel are used. First, selecting the students of MBA, EMBA, MEM in our university, and students contact self company’s manager, and manager fill in the questionnaire, a total of 50 questionnaires are issued, and 50 are returned, effective questionnaire are 48. Second, depend on the authors and their friends’ social network relationship, questionnaires are issued to managers in multinational corporation by E-mail, 100 questionnaires are send out, and received 72, valid questionnaires are 56. Third, through access to relevant business listing, we determine the eligible sample enterprises, and contact with the top managers by telephone, confirm that they can accept investigation, then the questionnaire mailed to relevant personnel directly or through the EMS, we send out questionnaire 100, 58 are received, and effective questionnaire is 52. Through the above way, total 250 questionnaires are send out, 180 valid questionnaires are received, and effective questionnaire is 156, with a response rate of 62.4%, meet the needs of the data processing.

3.2 Measures

Explicit knowledge transfer: A four-item scale developed by Cavusgil, et al.[8] and Subramaniam, et al.[20] is used to measure explicit knowledge transfer. Measures are rated on a five-point Likert-type scale, and response options ranged from 1, “strongly disagree”, to 5, “strongly agree”. The sample item are “Repatriates often provide work reports and official documents to others; Repatriates often provide the operation manual, methods and models to others; Repatriates often write what they know and give others reference; Repatriates are willing to answer other member’s problems as much as possible”. Cronbach’s alpha for this measure was 0.813.

Tacit knowledge transfer: A four-item scale developed by Szulanski[21] and Li[22] is used to measure tacit knowledge transfer. Measures are rated on a five-point Likert-type scale, and response options ranged from 1, “strongly disagree” to 5, “strongly agree”. The sample item are “Repatriates often share experience and skills with others; Repatriates are willing to help others to seek the learning opportunity; When discussing work problems, repatriates often try their best to provide opinions; For difficult explaining thing, repatriates are willing to demonstration to the others”. Cronbach’s alpha for this measure was 0.793.

Exploitative innovation: A four-item scale developed by Jansen et al.[23] is used to measure exploitative innovation. Measures are rated on a five-point Likert-type scale, and response options ranged from 1, “strongly disagree,” to 5, “strongly agree.” The sample item are “Enterprises improve existing products and services; Enterprise improve the flexibility of the existing products; Enterprises reduce the cost of production of existing products; Enterprise improve the production of existing products or reduce the material consumption”. Cronbach’s alpha for this measure was 0.806.

Exploratory innovation: A four-item scale developed by Jansen, et al.[23] is used to measure exploratory innovation. Measures are rated on a five-point Likert-type scale, and response options ranged from 1, “strongly disagree”, to 5, “strongly agree”. The sample items are “Enterprise introduces new products; Enterprise increases the new product type; Enterprises explore new market field; Enterprise enter new technology areas”. Cronbach’s alpha for this measure was 0.803.

Enterprise performance: A four-item scale developed by Baker and Sinkula[24] is used to measure enterprise performance. Measures are rated on a five-point Likert-type scale, and response options ranged from 1, “strongly disagree” to 5, “strongly agree”. The sample items are “Compared with competitors, enterprise’s sales growth rate is very high; Compared with competitors, enterprise’s assets profit margin is very high; Compared with competitors, enterprise’s market share growth rate is high; Compared with competitors, enterprise’s overall performance and success degree are high”. Cronbach’s alpha for this measure was 0.854.

Previous studies have indicated that the enterprise background variables (age, size, industry) will affect multinational corporation’s innovation and performance. Therefore, this research will select enterprise’s age, size and industry as control variables.

4.1 Confirmatory Factor Analysis

Confirmatory factor analysis (CFA) is conducted to analyze the distinctiveness validity among all constructs. We first examined a five-factor CFA model that included explicit knowledge transfer, tacit knowledge transfer, exploitative innovation, exploratory innovation, and enterprise performance. The model’s overall Chi-square, the root mean square error of approximation, the Tucker-Lewis index and the comparative fit index are used to assess model fit. Convention suggests that a value below 0.08 for RMSEA and a value over 0.90 for TLI and CFI are indicative of a good fit between the proposed model and the observed data. The proposed five-factor model demonstrate acceptable fit (χ²/df(160) = 284.686, p < 0.01; RMSEA = 0.071, TLI = 0.928, CFI = 0.939). In addition, all factor loadings are significant, demonstrating convergent validity.

The discriminant validity of the five proposed constructs is tested by contrasting the five-factor model against alternative models. Model comparison results, which are reported in Table 1, suggest that the four-factor and the one-factor models yield poor fits to the data. Thus, the distinctiveness of the key constructs used in this study is confirmed. Given these results, all five constructs are applied in subsequent analyses.

4.2 Common Method Issues

As the data of dependent and independent variables are collected from the same source, common method variance may occur. To minimize common method variance (CMV), following Podsakoff and Organ’s (1986) recommendations, we conduct Harman’s single-factor test. Principal factor analysis with varimax rotation is performed to determine whether one single factor explain a majority of variance. A factor analysis of the dependent and independent variables yields a factor solution that accounts for 67.65% of the total variance, and the first factor only accounts for 18.23% of the variance. Because a single-factor solution does not emerge and the first factor does not explain the majority of the variance, common method variance does not pose a pervasive problem in the data.

4.3 Descriptive Statistics

Table 2 presents the means, standard deviations, and Pearson correlations of all key variables. As shown, explicit knowledge transfer is positively correlated with exploitative innovation (r = 0.442, p < 0.01), exploratory innovation (r = 0.322, p < 0.01), and enterprise performance (r = 0.427, p < 0.01). Tacit knowledge transfer is positively correlated with exploitative innovation (r = 0.469, p < 0.01), exploratory innovation (r = 0.431, p < 0.01), and enterprise performance (r = 0.423, p <0.01). In addition, exploitative innovation (r = 0.447, p < 0.01) and exploratory innovation (r = 0.553, p < 0.01) are positively correlated with enterprise performance. Finally, age and scale have no significant correlation with explicit knowledge transfer, tacit knowledge transfer, exploitative innovation, and exploratory innovation, which means knowledge transfer and innovation have no significant difference in multinational corporation in terms of age, and scale.

4.4 Hypotheses Testing

Hierarchical multiple regression analyses are applied to analyze the relationship of variables. In order to test the relationship of knowledge transfer and ambidexterity innovation, we enter the variables into the regression analysis, dependent variable of exploitative innovation and exploratory innovation are entered first, followed by control variable (age, scale, and industrial) and independent variable of explicit knowledge transfer and tacit knowledge transfer. Table 3 presents the results. As shown in Table 3, explicit knowledge transfer is positively related to exploitative innovation (M₂, β = 0.394, p < 0.05), and exploratory innovation (M₄, β = 0.366, p < 0.01), thus supporting hypothesis 1a and hypothesis 1b. Tacit knowledge transfer is positively related to exploitative innovation (M₂, β = 0.461, p < 0.05), and exploratory innovation (M₄, β = 0.482, p < 0.01), which support hypothesis 2a and hypothesis 2b. In the meanwhile, we can get the conclusion that explicit knowledge transfer has bigger effect on exploitative innovation (0.394 > 0.366), while tacit knowledge transfer has bigger effect on exploratory innovation (0.482 > 0.461).

To test hypothesis 3 and hypothesis 4, dependent variable of enterprise performance are entered first, followed by control variable (age, scale, and industrial) and independent variable of exploitative innovation and exploratory innovation. As Table 3 presents, exploitative innovation and exploratory innovation have significant positive effect on enterprise performance (M₇, β = 0.531, p < 0.01; β = 0.294, p < 0.05), hence, H3 and H4 are confirmed. We predict that ambidexterity innovation mediates the relationship between knowledge transfer and enterprise performance. According to the suggestion of Baron and Kenny[25], the full mediation is supported if four conditions are met: 1) The independent variables significantly related to the mediator; 2) The independent variable is significantly related to the dependent variable; 3) The mediator is significantly related to the dependent variable; and 4) When both independent variable and mediator are entered into the model, the relationship between the independent variable and the dependent variable becomes nonsignificant. The results in Table 3 indicate that: 1) Explicit knowledge transfer and tacit knowledge transfer are positively related to exploitative innovation (M₂, β = 0.394, p < 0.05; β = 0.461, p < 0.05), and exploratory innovation (M₄, β = 0.366, p < 0.01; β = 0.482, p < 0.01); 2) Explicit knowledge transfer and tacit knowledge transfer are positively related to enterprise performance (M₆, β = 0.408, p < 0.01; β = 0.496, p < 0.01); 3) Exploitative innovation and exploratory innovation are positively related to enterprise performance (M₇, β = 0.531, p < 0.01; β = 0.294, p < 0.05); and 4) The relationship between explicit knowledge transfer, tacit knowledge transfer and enterprise performance (M₈, β = 0.182, p < 0.05; β = 0.212, p < 0.05), still significant but decrease when knowledge transfer and ambidexterity innovation are entered into the model, while exploitative innovation and exploratory innovation still is positively related to enterprise performance (M₈, β = 0.265, p < 0.01; β = 0.391, p < 0.01). Thus, exploitative innovation play a partial mediatory role in the relationship among explicit knowledge transfer, tacit knowledge transfer and enterprise performance, hypothesis 5 and hypothesis 6 are supported. In the mean while, exploratory innovation play a partial mediatory role in the relationship among explicit knowledge transfer, tacit knowledge transfer and enterprise performance, supporting hypothesis 7 and hypothesis 8.

5.1 Conclusion

Repatriate’s knowledge transfer in multinational corporation is beneficial to the sustainable development of multinational corporation. This research puts knowledge transfer, ambidexterity innovation and enterprise performance into one framework, analyzes the mediation role of ambidexterity innovation in the relationship between knowledge transfer and enterprise performance. Using empirical method, study object, 156 multinational corporations, is selected, and data are collected, then common method variance, descriptive statistics, and hierarchical multiple regression analysis are made by applying SPSS 19.0. The results are discussed in the following. First, exploitative innovation and exploratory innovation play partial mediatory role in the relationship of explicit knowledge transfer, tacit knowledge transfer and enterprise performance, that is to mean, the effect of explicit knowledge transfer and tacit knowledge transfer on enterprise performance are not wholly direct, partial influence on enterprise performance through exploitative innovation and exploratory innovation. At the same time, partial effect also means knowledge transfer is not completely affect enterprise performance through innovation, maybe there are other mediators.

5.2 Theoretical Contributions

Combining the theory of knowledge transfer, innovation management and enterprise performance management, this study expands the research field of knowledge transfer, and discusses the relationship of repatriate’s knowledge transfer, ambidexterity innovation and enterprise performance. This study makes several theoretical contributions. First, according to ambidexterity theory, innovation in multinational corporation is divided into two dimensions: Exploitative innovation and exploratory innovation, analyze the effect of explicit knowledge transfer and tacit knowledge transfer on exploitative innovation and exploratory innovation, which further expand the innovation theory. Second, we propose that ambidexterity innovation play a mediatory role in the relationship between knowledge transfer and enterprise performance. The results show that ambidexterity innovation play partial intermediary role, which indicate repatriate’s knowledge transfer not only has direct positive effect on enterprise performance, but also has indirect influence on enterprise performance through innovation. Third, multinational corporation as the research object, we demonstrate that ambidexterity innovation plays an important role in the repatriate’s knowledge transfer and enterprise performance, and realize the integration of human resources management and innovation capability, which has theoretical contribution to multinational corporation’s knowledge transfer and innovation.

5.3 Practical Contributions

From a practical viewpoint, the results of our study have implications for managers. According to the interaction relationship between repatriate’s knowledge transfer, ambidexterity innovation and enterprise performance, managers can design policies that can promote multinational corporation’s performance. First, we find that knowledge transfer has significant influence on exploitative innovation and exploratory innovation. Hence, managers should take measures to facilitate repatriate’s to transfer knowledge. Such as, knowledge sharing culture can be established, which can facilitate knowledge communicating and integrating among members, and then knowledge transfer can be occurred. And managers should design corresponding incentive measures, and encourage repatriates to transfer knowledge, especially tacit knowledge transfer. Second, our findings indicate that exploitative innovation and exploratory innovation is positively related to enterprise performance. Hence, an effective way to improve multinational corporation is to improve repatriates innovation behavior. Managers should provide many kinds of training for repatriate, and facilitate repatriate to produce creative thinking, so that innovation speed and quality of multinational corporations can be promoted.

5.4 Limitations and Future Research

This study is an exploratory empirical research, there are some limitations. First, due to the limitation of research conditions, although the samples are selected from many industries, the sample size and industrial involved in the survey is still not wide, in such, the results of the study are not comprehensive, and future research can be further expanding the scope of the investigation and sample size. Second, because of cost, time and other conditions limitation, this research that use the cross sectional study can only get the correlation of variables, the causal relationship between variables also need inspection by tracing from longitudinal. Third, there may be common method bias, because the survey asked the same respondents about both independent and dependent variables. Future studies can consider longitudinal designs, which it is vital to understand the mutual relationship of knowledge transfer, ambidexterity innovation, and enterprise performance.