1 Introduction

Technology transfer plays an important role in developing countries (Inkpen and Dinur 1998), and it has been as one of vital strategies for building an innovative nation in China. Transferring technology across borders is an expensive complex and difficult task because there are so many factors affecting the processes (Szulanski 2000; Szulanski et al. 2001; Simonin 1999; Kostova and Roth 2002; Cummings and Teng 2003; Reed and DeFillippi 1990). Among these factors, the host country’s institutional profile is a critical one (Kostova and Roth 2002). The country institutional profile reflects the institutional environment in that country and is defined as the set of all relevant institutions that have been established over time, that operate in that country, and that are transmitted into organizations through individuals (Kostova 1997, 1999; Kostova and Roth 2002). Institutional environments comprise three main components: regulatory, cognitive, and normative (Kostova 1997). In Scott’s institutional theory, the regulatory component reflects the existing laws and rules in a particular national environment which promote or restrict certain types of behaviors. The cognitive component reflects the cognitive structures and social knowledge shared by the people in a given country. The normative component consists of social norms, values, beliefs, and assumptions about human nature and human behavior that are socially shared and carried by individuals (Kostova 1997).

Though some studies have focused on this issue, most of those were conducted in western countries, and there are few with a Chinese context, so it is urgent to focus on the relationship between the Chinese institutional profile and technology transfer. To close this gap, the present study tries to extend this field of study based on a survey of 167 foreign ventures sampled in China. The following section addresses theoretical backgrounds and develops hypotheses; the third section introduces the methodology and sampling, and the fourth section presents the empirical results. The final section presents discussion, conclusion, theoretical and practical implications, and potential directions for future study.

2 Theoretical backgrounds and hypotheses

It is challenging to transfer technology, whether in developed or developing countries. There are so many factors that can affect the effectiveness of technology transfer across borders (Szulanski et al. 2001; Simonin 1999; Cummings and Teng 2003). Characteristics of technology transfer, include such factors as complexity, specificity, and tacitness (Reed and DeFillipi 1990); characteristics of the accepted party include such things as time involvement, organizational structure, organizational culture, and technology and knowledge strategy orientations (Cummings and Teng 2003); and the characteristics of relationship quality and level between the transferring and accepting parties include such things as trust level (Simonin 1999). For technology transfers across borders, in addition to those mentioned above, other special barriers exist, such as cross-cultural differences and institutional profiles (Kostova and Roth 2002).

Before reviewing the literature on host country institutional profiles, we will first address how to evaluate the effectiveness of technology transfer across borders, because the issue of efficacy of technology transfer is another “hot point” and would also be a dependent variable in this present study.

2.1 Effectiveness of technology transfer

There has been no consensus so far about how to evaluate the effectiveness of technology and knowledge transfer (Cummings and Teng 2003). The direct effect of technology transfer is obtaining the technology, including the amount received, and the advancement of technology. Here, we call this the technological performance of technology transfer. Technology transferring is a complex, even expensive process, although it is always welcome by many parties, especially by entities in the developing countries. For them, technology transfer means not only obtaining advanced technology and management experience, but also helping in developing and improving their current technology and management experience. So in the present study, we are particularly concerned about the performance of technology transfer and would set it as a dependent variable for evaluating the effectiveness of technology transfer across borders.

2.2 Country institutional profile

Research suggests that organizational technology and practices vary across countries (Lincoln et al. 1986), since they are affected by the socio-cultural environments in which they have evolved and are being used. Cross-country differences have been found in a variety of organizational technology and practices such as negotiations, leadership and distribution of power and authority in organizations (Hofstede 1980), and HRM practices (Adler 1995). There are many factors from the country level that affect technology transfer and integration, and researchers have usually studied country-level effects by using the concept of national culture, which has been defined in various ways. For example, Kogut and Singh (1988) use a national cultural distance to evaluate the effects of national cultural difference on entry modes in the view of Hofstede’s famous four-dimension national culture model.

Here we use an alternative way to conceptualize social or country-level effects, using a country institutional profile, rather than cultural characteristics like what Kostova proposed (1997, 1999). According to Kostova (1999), there at least are three considerations: (1) countries differ in their institutional characteristics; (2) organizational technology and practices reflect the institutional profile of the country where they have been developed and established and (3) when technology and practices are transferred across borders, they may not “fit” with the institutional profile of the recipient country, which may be an impediment to the transfer. According to institutional theory, Scott (1995) proposed that institutional profiles are composed of various types of institutions and are characterized by three dimensions: regulatory, cognitive, and normative. The regulatory component of an institutional profile reflects the existing laws and rules in a particular national environment that promote and restrict certain types of behaviors (Scott 1995). The cognitive component reflects the cognitive categories widely shared by the people in a particular country (Markus and Zajonc 1985; Scott 1995). Scott suggested that cognitive elements constitute the nature of reality and the frames through which meaning is made. Although carried by individuals, cognitive programs are elements of the social environment and are social in nature (cf. Berger and Luckman 1967). The normative component of an institutional profile focuses on normative systems; that is, the values and norms held by the individuals in a given country. Normative components introduce “a prescriptive, evaluative and obligatory dimension into social life” (Scott 1995). Norms specify how things should be done (Hosfstede 1980). The regulatory, cognitive, and normative components comprehensively comprise a country institutional profile (CIP) (Kostova and Roth 2002; Kostova 1997, 1999). Based on 600 samples from ten countries for the issue of quality management, Kostova (1997) constructed and developed a measurement instrument for country institutional profiles, including the regulatory, cognitive and normative aspects. Kostova (1997) called for two logical future extensions that would develop institutional profiles for other countries and for other key managerial issues. Recently, Busenitz et al. (2000) have measured the CIP construct and also evaluated the reliability and validity of entrepreneurial climates, but in the technology and knowledge transfer field, the CIP construct is little confirmed, particularly in the Chinese context.

Institutional theory holds that the values and implications of organizational practices overwhelm the technology of organizational practices because every specific practice has indicated specific institutional climates. So in order to improve the effectiveness of technology transfer across borders, the receiving country (that is the host country) should have an institutional environment that supports technology transferring. Though one country institutional profile would be depicted from the regulatory, cognitive, and normative aspects, the three aspects play different roles during the technology transfer process across borders (Kostova and Roth 2002). Regulatory component sometimes bring forces to both parties; for example, the legislation and regulations of some host countries clearly express when entering foreign ventures should be followed by advanced technology and management skills. This phenomenon is common in developing countries or areas (Inkpen and Diur 1998). Also, the host countries might set and pass some legislation or regulations to encourage the receiving parties to learn the transferred technology and management skills. But no matter how technology is transferred, ultimately the effectiveness of technology transfer depends on individual learning. Actually, institutional theory also holds this opinion, that the effects of the institution on the organization are the result of the effects on individuals because individual cognition and beliefs would affect on their perceptions and judgments, then their attitudes and behaviors. Social climates, norms and values would also affect individual attitudes and behaviors.

The following hypotheses are proposed specifically for determining the effectiveness of the technology transfer process across borders,

  1. H1:

    Regulatory institutions in the host country would be positively related to the technological performance of technology transferring across borders.

  2. H2:

    Cognitive institutions in the host country would be positively related to the technological performance of technology transferring across borders.

  3. H3:

    Normative institution in the host country would be positively related to the technological performance of technology transferring across borders.

3 Method

3.1 Sample and procedure

Since reforming and opening, China has been a “hot bed” for foreign ventures; most recently, the trend is that foreign direct investment (FDI) has been the main mode among foreign ventures (Liu and Wang 2005). From the industry perception, manufacturing and service are the two main bases with 73.48 and 24.57%, respectively. Most of foreign ventures are clustered in the southern and eastern China regions (Yangtze River Delta and Zhujiang River Delta) and regions across Bo Hai (including Beijing, Tianjin, and Hebei), amounting to 62.64%. So our samples are foreign ventures located in these economic regions and mainly in the manufacturing and service industries.

The survey questionnaire is strictly developed. All items are abstracted from the public literature and revised according to Chinese characteristics. Respondents are Chinese and foreigners. So there are two identical versions (English and Chinese) of the questionnaire with the translation-back-translation procedure. Due to the difficulties and challenges confronted by empirical studies in a Chinese context, the random sample procedure is not always useful. Here we have used a convenient sampling technique, sending the questionnaire to the subjects contacted before. Information that the respondents provide is assumed to be quality information, so our questionnaires are limited to responses by middle management and above because above middle or top management levels would provide relatively real and complete information. This survey delivered 1,000 questionnaires (Chinese/English version is 800/200), and there are effectively 167 questionnaire responses, given the ratio of 16.7% (Chinese /English version is 165/2). According to Hambrick et al. (1993), the response ratio from top management is 10–12%. Among them, 65% of the respondents are vice presidents, human resource managers, marketing managers, research and development managers, and technology managers, and 21% of the respondents are supervisors and program leaders. The rest (14%) are professional engineers (average tenure more than 3 years). There are 31 organizations from the financial, consulting, and housing development industries, 136 organizations from the manufacturing industry (such as machinery, electronic and information products, and medical and pharmacy industry). The average age entering into China is 9.4.

According to Podsakoff and Organ (1986), we examined the common method bias error issues. The results of principal component factor analysis show that there were four factors with an Eigenvalue over one. And the four factors explained 82.4% of the total variance (the largest factor is 23.17%), so we could draw a conclusion that the common method bias error issue did not exist.

3.2 Key construct measurement

3.2.1 Technological performance of technology transferring (TPTT)

Based on Wong et al. (1998) and using a 5-point scale, we estimated the technological performance of technology transferring across borders (TPTT) by asking respondents about the following four items: “achieve the expected technology”, “corporate technological competence improvement”, “achieve the expected management and skills”, and “corporate management and skills improvement” (α = 0.929).

3.2.2 Country institutional profiles (CIP)

According to Kostova and Roth (2002), the original country institutional profile scale has been revised to take the Chinese characteristics into account. The revised scale includes nine items related to regulatory institutions (α = 0.689), cognitive institutions (α = 0.681), and normative institutions (α = 0.696).

3.2.3 Control variables

Here, we choose foreign ventures entering into China mode (Mode), foreign ventures’ size (Size), tenure of foreign ventures entering into China (Tenure), and industry (Industry) as control variables to explain the empirical studies in detail. Regarding Tenure, generally speaking, if tenure is longer, foreign ventures would more depend on themselves to achieve technology and management and skills and depend less on technology transfers across borders. As in much similar literature (e.g., Minbaeva et al. 2003), Tenure is calculated from the time when foreign ventures entered China. Regarding corporate size, larger corporations often have systematic and complete functions, including technology, research and development departments, and turn less often to technology transfers. As in the literature (e.g., Minbaeva et al. 2003), size is the natural logarithm of the total foreign ventures’ employees. Regarding industry characteristics, for international entrepreneurial entities, industry has an effect on entrepreneurial performance (Robson et al. 2003). That is, if foreign ventures are in the manufacturing industry, let Industry be equal to 1; if from the service industry, let Industry be equal to 0. Entering mode is another important variable because different modes accompany different control powers, resource relocation, risk and potential revenues (Liu and Wang 2005). So we control the Model variable. Because we just focus on joint ventures and wholly-owned ventures, if foreign ventures belong to joint ventures, let Mode be equal to 0; if foreign ventures belong to wholly-owned ventures, let Mode be equal to 1.

4 Results

All data are analyzed by SPSS 12.0 and AMOS 5.0 software. Table 1 is the descriptive statistical result of all variables. Table 1 shows the means, standard deviations, and correlations between the main constructs. The reliabilities for the main constructs are also given in Table 1. We find that all of the Cronbach’s alphas coefficients are above 0.68, which is close to the original reliabilities in the works of Kostova (1997). That is, an acceptable internal consistency of the constructs is conducted overall.

Table 1 Descriptive statistics and correlations (N = 167)
Full size table

To confirm the country institutional profile and technological performance of technology transfers across borders, we examined the latent variable construct by AMOS 5.0 software. We used the increment fit index (IFI) and the comparative fit index (CFI) as key indicators of overall model fit. Figure 1 shows the construct measurement information.

Fig. 1
figure 1

Latent variable Measurement Information (a) Country Institutional Profile Construct; RI, Regulatory Institution; CI, Cognitive Institution; N, Normative Institution; (b) TPTT construct, TPTT , Technological Performance of Technology Transferring

Full size image

From Fig. 1a, we found that the correlations among the three subscales are significant, which means the convergent validity of the multidimensional CIP construct is accepted.

Confirmatory factor analysis (CFA) of the two latent variables yields fit indices as shown in Table 2. From Table 2, we found the country institutional profile (CIP) model and the technological performance of technology transferring (TPTT) model showed an acceptable fit on a wide range of goodness-of-fit measures (CFI = 0.887, NFI = 0.845, GFI = 0.916, IFI = 0.890, RMSEA = 0.111 for CIP; CFI = 0.945, NFI = 0.910, GFI = 0.876, IFI = 0.946, and RMSEA = 0.091 for TPTT), and no problems are found in residuals or standard errors. From Table 2 and Fig. 1, we can conclude the resulting factor structure showed clean three factor structures with all items loading significantly for CIP and a unique factor structure with all items loading significantly for TPTT.

Table 2 Latent variable CFA information
Full size table

If the CFA factor structure yields an acceptable fit, it indicates the presence of three distinguishable dimensions of the country institutional profile in China. We next sought to establish that the three-factor structure has a better fit than a factor structure which would indicate that participants can distinguish the three dimensions of CIP measured. To accomplish this, we compare the relative fit of one-factor and three-factor models of the CIP for each of the samples and collapsed them into one factor. Table 3 outlines the compared information.

Table 3 One-factor and three-factor of CIP comparison
Full size table

From Table 3, we clearly find that the three-factor model had a significant better fit than the one-factor models (Δχ 2 (Δdf) = 64.336). Thus we found that the CIP has three distinguishable dimensions in our Chinese samples. That means the multidimensional construct of country institutional profile (CIP) is more appropriate than the single dimensional construct. Also, according to the confirmatory factor analysis conducted in the Chinese fields, we can conclude that the construct originating in the western societies has an acceptable validity level.

To test Hypotheses 1, 2, and 3, we used structural equation modeling techniques. After deleting cases that have missing data for any of the indicators, we retained a total of 142 samples, which we used for subsequent model testing. Our first test included not only the hypothesized paths, but also all possible additional paths. The purpose of this estimation is to determine whether or not additional direct paths should be included in our proposed model to capture possible partial mediation effects. However, all of these paths were not significant; therefore, none of them were retained in the model. The final structural model and its resulting standardized parameter estimates are shown in Fig. 2. Fit indices resulting for this model suggest that it fits the data modestly well with χ 2 = 200.068, df = 100, RMSEA = 0.178, CFI = 0.856.

Fig. 2
figure 2

Final structural models of the present study, RI, Regulatory Institution; CI, Cognitive Institution; NI, Normative Institution; TPTT, Technological Performance of Technology Transferring. ➝ Indicates the significant relationship line;

figure a
Indicates the nonsignificant relationship line (the lines are disappeared in the final structure model)

Full size image

From Fig. 2, we find that the host country institutional profile (CIP) does affect fully the technological performance of technology transferring (TPTT). Specifically, the normative component of the host country institution is significantly positively related to its technological performance (β = 0.329, p < 0.001). Contrary to our hypothesis 1, the regulatory component of host country institution is significantly negatively related to technological performance (β = −0.230, p < 0.001), while there is no significant relationship between cognitive component and TPTT (β = −0.159, no significance). So hypothesis 3 is confirmed, while hypotheses 1 and 2 are not supported.

5 Discussion, conclusions, implication, and future research directions

There are many barriers or challenges during the process of technology transfers across borders. The present study specifically focused on the effects of host country institutional profiles, which are a means to conceptualize and measure country-level characteristics that affect organizations’ effectiveness when transferring technology across borders. Below, we discuss our main findings, then suggest several theoretical and practical implications. Finally, we address some of the limitations of this study and directions for future research.

5.1 Main findings

The idea of using country institutional profiles as a means to characterize the national environment is consistent with the social embeddeddness perspective in the organizational field that suggests that individuals, organizations, and organizational routines are affected by the social environment in which they exist (Kostova 1997). That is, institutions play a vital role in organizational and individual behaviors. In order to import and assimilate the foreign technology, the developing counties try to set up laws and regulations and cultivate social climates to promote the technology transfer. Some studies have indicated that the host country institutional environment has a strong effect on the technology transfer process. But our empirical study, based on a Chinese context, did not fully confirm the previous cognitions and studies. Just as institutional theory had suggested, we find the normative component of the host country institution is significantly positively related to the technological performance of technology transfers. According to Kostova (1997), the normative institution reflects the social shared norms, values, beliefs, and assumptions about human nature and human behavior. Thus, a positive normative institution intends to cultivate social climate supportive of technology transfers across borders because some negative feelings, such as the Not-Invented-Here (NIH) syndrome (Katz and Allen 1982), would be restricted during technology transfer process across borders. At the same time, if the receiving parties recognize the value of technology transfers across borders, such as gaining advanced technology and management and skills, they would whole-heartedly support the programs and commit to the processes.

Contrary to our hypothesis 1, the regulatory component of the host country institution is significantly negatively related to technological performance of technology transfers. What are the reasons for this? Generally speaking, regulation would promote the effectiveness of technology transfers across borders. Why does the regulatory component of the host country institution significantly restrict technology transfers? We speculate a possible reason as follows. When the force of regulations or legislations is appropriate, the existing laws and rules in a particular national environment would promote certain types of behaviors. When regulations or legislations in host countries are too strong, the transferring parties would feel numerous forces or compulsions, which would create resistance, and then the effectiveness of technology transfers is restricted.

Our studies also show there is no significant relationship between the cognitive component of the host country institutional profiles and the technologic effectiveness of technology transfer. We doubted whether we could generalize institutional theory in a Chinese context. Because our study framework is based on institutional theories that emphasize the dependence between organizational activities and external environments, we guessed that perhaps the theories’ usefulness is limited in our Chinese context study; that is, the effectiveness of technology transfer has little relationship to external institutions.

Although many studies (e.g., Minbaeva et al. 2003; Robson et al. 2003; Liu and Wang 2005) have insisted that organizational characteristics such as size, industry, tenure and the entry mode would have a significant effect on the technological effectiveness of technology transfers, our studies do not provide supporting evidence of that. We think that sampling bias or error may be an issue.

5.2 Theoretical implications

Our studies can potentially contribute to a number of areas. The paper extends the institutional theory in international management. Kostova (1997) suggested two logical future extensions would be to develop institutional profiles for other countries and for other key managerial issues. Our studies simply try to expand these issues. The main contribution of the paper is to validate the country institutional profiles scale for technology transfers across borders in the mainland of China. Another major contribution of the paper is to international technological entrepreneurship and to technology transfer across borders, especially in the developing nations. International technological entrepreneurship is a key way to transfer technology across borders (Inkpen and Dinur 1998). In the perspective of the technology provider, maximizing the technological advantages is the ultimate goal; in the view of the technology accepter, acquiring the necessary technology and knowledge is the destination. But for both parties, technology transfer across borders is difficult and complex. From the institutional theory, this study provides strong evidences in the Chinese context.

5.3 Practical implications

Our empirical study indicates that technology transfers across borders would bring with it precious advanced technology, management and skills, but we also found that the transferring process is not easy. There are many factors that would affect the effectiveness of technology transfers across borders (Cummings and Teng 2003; Liu and Wang 2004). For the host government, especially for those governments who urgent want to achieve advanced technology, management and skills with foreign capital (just like China) (Inkpen and Diur 1998; Liu and Wang 2004), legislation and regulation are overemphasized for the expected goals; the reverse unexpected results would generate strong resistance. A proactive role on behalf of governments would be to develop favorable institutional climates for certain individual and organizational behaviors. Specifically, the harmonious social and normative climates for technology transferring across borders would be an incentive for achieving expected goals through technology transfers across borders.

Although our empirical studies do not support the effects of control variables, such as size, industry, tenure, and entry mode, we insist that the control variables should not be ignored. For example, the different entry modes encounter different risks or liability from foreignness (Zaheer 1995; Liu and Wang 2004), and industry does matter to organizational performance (Minbaeva et al. 2003; Robson et al. 2003). So in the view of management practices, faced by the question of how to improve the technological effectiveness of technology transfers across borders, key organizational characteristics that include size, tenure, industry, and entry mode should be considered.

5.4 Limitations and future research directions

There are some limits in the present study. Though the construct of host country institutional profile (CIP) is examined by 167 samples from the mainland of China, and the reliabilities of the three subscales of CIP are acceptable, the final confirmatory structural model fit indices are modest, indicating that a better model is needed. That means a better country institutional profile (CIP) model could exist. But what is the better CIP model? Future studies need to expand this issue. At the same time, because most of the samples in our studies are from the relatively developed regions on the mainland of China, sampling bias or error could affect the stability of the CIP construct. Further efforts should make up for this sampling bias. Also, our empirical results are based on a cross-sectional data design; longitudinal studies are needed urgently to consider the complex process of technology transfers across borders, because the technology transfer process is closely linked to organizational learning, which is often affected by the times.