Printer Friendly

The performance effects of headquarters' involvement in lateral innovation transfers in multinational corporations.

Abstract:

* Developing an analytical framework to help examine the impact of HQs direct involvement in lateral innovation transfers on efficiency and effectiveness of

* Develop hypotheses which are tested by multiple regression analyses on data relating to 129 lateral innovation transfers in multinational corporations

* Three key findings: dyadic willingness positively affects transfer efficiency and effectiveness willingness HQ involvement negatively effects transfer efficiency; HQ involvement negatively moderates the effect of willingness on transfer effectiveness

* HQ face a dilemma in managing lateral transfers: on the one hand, the network structure that typifies many MNCs has beneficial impacts on the performance of the transfer process; on the other hand the HQ need to ensure that lateral transfers initiated by subsidiaries ('organic' transfers) are in line with MNC strategy. However the findings from our study strongly suggest that HQ direct involvement in the transfer process is at best a very blunt instrument for achieving a resolution of this dilemma.

Keywords: Multinational corporations * HQ involvement- Innovation transfer efficiency. Effectiveness * Transfer performance

Introduction

Although the management of knowledge transfer and integration is regarded as a key component of organizational capability in multinational firms (Tsai and Ghoshal 1998; Tsai 2000, 2002; Schultz 2003; Gupta and Govindarajan 2000; Foss and Pedersen 2002; Minbaeva et al. 2003; Bjorkman et al. 2004; Persson 2006a; Zhou and Luo 2005; Adenfelt and Lagerstrom 2006; Monteiro et al. 2008) relatively little attention has been paid to the role of headquarters (HQ) in the process and hence to the possible consequences of HQ direct involvement in intra-MNC knowledge transfers. While the development of the MNC's knowledge assets has always been of strategic importance to the HQ; recent studies suggest a heightening of strategic assertiveness by the HQ vis-a-vis subsidiaries (Nolan et al. 2002; Yamin and Ghauri 2010; Buckley 2010), particularly in relation to knowledge and competence development in MNCs. HQ direct involvement in knowledge transfer (entailing instructing a particular transfer and participating in the transfer process) within the MNC would be a facet and a reflection of such enhanced assertiveness. HQ would naturally be actively involved in vertical knowledge transfers as they are either the sender or the recipient (in the case of 'reverse' transfers; Subramaniam et al. (1998) and Subramaniam and Venktraman (2001)). However where lateral, inter-subsidiary, knowledge transfers are concerned, current research has examined only HQ policies that may influence knowledge transfer by subsidiaries. Thus the few existing studies on HQ management of lateral knowledge transfers have focused on issues such as compensation policies; the employment of expatriates and the application of socialization and shared value approaches (Tsai and Ghoshal 1998; Gupta and Govindrajan 2000; Tsai 2000; Bjorkman et al. 2004; Persson 2006a). None of these studies consider the direct involvement of HQ in lateral transfers.

In this study we examine the consequence of HQ direct involvement for the efficiency and effectiveness lateral knowledge transfers. Previous studies on knowledge transfer have not adopted a performance perspective, although it is acknowledged that knowledge transfer is a costly or sticky process (Teece 1977; Szulanski 1996; Jensen and Szulanski 2004; Gupta and Govindarajan 2000). The costs stem partly from the inherent characteristics of the knowledge to be transferred (tacitness and complexity); partly from the motivational stances of individual subsidiaries towards transferring and receiving knowledge and the relational aspects of particular transfer dyads.

However this literature does not consider (a) the efficiency of the transfer--that is, whether and to extent the costs may be above the minimum that may be expected in any transfer projects and (b)--effectiveness of the transfer--the completion of the transfer project and a satisfactory transfer outcome (Daft 1992). We note that the efficiency and effectiveness of lateral transfers within MNCs are outcomes that are organizationally variable, that is the particular way in which the MNC organizes and governs lateral transfers will likely have a (differential) effect on the efficiency and effectiveness of the transfer. Thus whether the MNC relies on autonomous subsidiary decisions to organize lateral transfers (from 'below') or through HQ direct involvement (from 'above') should affect the amount of resources used and the cost of the transfer and whether the transfer is perceived as satisfactorily completed at the dyad level. That transfer outcomes are likely to be organizationally variable is implicit in the general discussion of knowledge (or 'best practice') transfers (Szulanski 1996; Jensen and Szulanski 2004; Gupta and Govindararjan 2000; Bjorkman et al. 2004). These studies put much emphasis on the motivational aspects and on whether the organizational environment within which the transfer takes place is 'arduous'. Subsequent literature (see below) does suggest that prior relational ties between subsidiaries have an important bearing on lateral transfer outcomes.

While a focus on the performance aspects of inter-subsidiary knowledge transfer is clearly useful for enriching understanding of knowledge transfer; it is crucial in relation to the direct involvement of HQ in the transfer process. HQ' goals vis-a-vis lateral innovation transfers are presumably derived from their strategic objectives. These could range from strengthening firm specific assets (Birkinshaw et al. 1998) to directing the development of technological and knowledge assets in its 'core' areas of activity (Nolan et al. 2002; Morgan 2001) or developing internal capabilities for diversification in particular directions (Andersson and Holm 2010). All of these objectives entail building up knowledge assets in order to enhance MNC competitive advantage: the MNC HQ will aim to build up knowledge assets more efficiently and effectively than competitors. However, if HQ direct involvement in lateral transfers impairs the efficiency and/or effectiveness of the transfer process, then the strategic objective may not be achieved.

In this study, knowledge transfer is viewed as a targeted, deliberate and a unidirectional process in which knowledge flows from a source to a recipient subsidiary. Knowledge transfer is thus distinctive from knowledge sharing or knowledge flows which may be multi-directional (Zhou and Luo 2005). We focus on the transfer of subsidiary generated innovations; the term 'innovation' referring specifically to the development of inventive ideas or technologies into marketable new products and processes (Teece 1986). Innovation transfer is thus regarded as a specific form of knowledge transfer where the knowledge relates to the know-how and other features of an innovation that would need to be transferred to the recipient subsidiary adopting the innovation in its host environment. Our research approach focuses on 'organic' lateral transfers--that is innovation transfers undertaken between subsidiaries without HQ being one of the two parties in the transfer relationship or an intermediate recipient/sender putting the characteristics of particular organic transfer dyads in the centre of the analysis. However, we examine the consequences of HQ' involvement in the lateral transfer in terms of its direct and moderating effects vis-a-vis the organic transfer dyad. The involvement of HQ transforms the dyad into a 'triad', where the direct effect concerns the link between the efforts of HQ to complete the transfer process and the efficiency and effectiveness of that process. The moderating effect concerns how the influence of the dyadic absorptive capacity and dyadic willingness on transfer outcomes may be modified or become redundant as a result of HQ direct involvement.

The paper is organized as follows. In the next section we discuss the organizational context that provides a rationale for the direct involvement of HQ in lateral transfers. We then develop hypotheses on the direct and moderating effects of HQ involvement on the efficiency and effectiveness of lateral transfers. In the subsequent section we explain the data collection process and the operationalisation of the theoretical constructs as well as the empirical analyses. The final section provides a conclusion and a discussion of the findings and their implications, particularly focusing on the control dilemmas faced by HQ.

Analytical Framework and Hypothesis Development

Current research indicates that lateral innovation transfers mostly take place in an organizational context with three inter-related features. First, some subsidiaries generate significant innovations and thus control considerable knowledge assets (Pearce 1999; Frost 2001; Frost et al. 2002; Yamin and Otto 2004; Almeida and Phene 2004; Mudambi and Navarra 2004; Phene and Almeida 2008; Cantwell and Mudambi 2005). Second, extensive knowledge sharing is associated with the emergence of extensive networks of ties and collaborative links between some subsidiaries (Tsai 2000; Tsai and Ghoshal 1998; Gnyawali et al. 2009; Monteiro et al. 2008). Finally, there is also competition between subsidiaries for markets as well as internal organizational resources (thus the phenomenon of 'coopetition', see e.g. Tsai 2002; Birkinshaw and Hood 1998; Birkinshw and Lingbald 2005; Bouquet and Birkinshaw 2008).

Given the above context, researchers have generally concluded that, compared to hierarchical and centralized methods of coordinating inter-subsidiary interactions, socialization and relational network approaches are more likely to encourage inter subsidiary innovation transfers (Andersson and Holm 2002; Tsai 2002; O'Donnell 2000; Bjorkman et al. 2004). This line of analysis implies that 'organically' formed transfer dayds will generally achieve transfer outcomes without the need for direction or involvement from 'above'. It is also acknowledged that HQ can reinforce the organic process by structural and incentive policies that encourage subsidiary innovation and its transfer.

On the other hand, recent changes in the organization of the MNC indicate that, while it remains a differentiated organization with subsidiaries having varying roles and capabilities, HQ are prone to be more assertive in controlling the strategic direction of the MNC, implying a dilution of federal structures (Nolan et al. 2002; Yamin and Forsgren 2006; Yamin and Sinkovics 2007, 2010; Yamin and Ghauri 2010; Buckley and Ghauri 2004; Rugman and D'cruz 2000; Buckley 2010). Consistent with this literature, the perspective we adopt is that HQ' direct involvement in lateral innovation transfer is related to an imperative to assert strategic authority in relation to the development of the MNC's knowledge assets. We acknowledge the roles of HQ in MNCs are diverse and cannot be conceptualized in a singular manner (Anderson and Holm 2010). In particular HQ 'involvement' in lateral transfers could be motivated by different organizational rationales and is also likely to take different forms. In particular HQ involvement may occur to resolve conflict within an existing transfer dyad (Argyres 1995). In other words, HQ gets involved to 'turnaround' a lateral transfer project that is beset with inefficiencies. However, HQ involvement may go no further than a "threat of intervention", which may be sufficient to persuade managers in the dyad to reach an agreement (Andersson et al. 2006). More specifically, HQ involvement in such a case only entails participation in an already ongoing transfer project. It does not entail initiating a new lateral transfer.

In this paper we limit our focus specifically to the implications of HQ involvement motivated by an imperative to assert strategic authority. We have chosen this focus as, while there is a general acknowledgement of the 'demise' of the federative MNE, there is no detail empirical investigation of its implications for MNE management with respect to the critical issue of knowledge and competence development in MNEs. Consistent with our focus, we conceptualize HQ direct involvement to entail (a) both initiating (instructing) a lateral transfer project and remaining involved during the various stages of the transfer process and (b) that the involvement takes the form of participation in the transfer project. Strategic imperatives as perceived by HQ and subsidiaries are often quite different (Regner 2003). More specifically, HQ' views with regard to the strategically appropriate pattern and direction of lateral innovation transfers may not coincide with that of subsidiaries. For example, there is not necessarily a consensus between HQ and subsidiaries what practices and innovations are regarded as 'best' and should be diffused to the MNC as a whole (Birkinshaw et al. 2000). This argument gains credence if 'organic' lateral transfers occur more frequently amongst an 'in-crowd' group of subsidiaries with a history of mutual knowledge flows while other subsidiaries may remain 'isolated' (Monteiro et al. 2008). Given the existence of strong ties amongst the 'in-crowd', the pattern of 'organic' lateral transfers may reflect more the relational harmony amongst a group of subsidiaries rather than overall MNC strategic imperatives, thus necessitating the direct involvement of HQ involvement in lateral innovation transfers. Hymer's analysis of the organization of the MNC anticipated that this sorts of 'alliances and interactions' between a group of subsidiaries are a threat to the strategic authority of HQ 'as they lead to actions that run counter to those prescribed by higher management' (Cohen et al. 1979, p. 155).

While, as noted above, HQ are able to influence the motivation of individual subsidiaries to engage in lateral transfers, through linking performance evaluation to transfer activity (Bjorkman et al. 2004), such incentives may be insufficient to persuade a subsidiary to transfer to specific recipients favoured by the HQ. In this context HQ may directly instruct particular inter-subsidiary transfers. Furthermore, the HQ would also wish to be involved by participating in the post initiation phase of the transfer process to ensure the process progresses to a transfer outcome and that the adoption by the recipient is not merely 'ceremonial' (Kostova and Roth 2002). Below we develop hypotheses on how HQ involvement affects the efficiency and effectiveness of the transfer dyad both directly and indirectly. The direct effects relate to resource (including knowledge) implications of HQ participation in the lateral transfer process compared to the organic transfer process. The indirect effects relate to how HQ involvement moderates the link between the relational attributes of the transfer dyad and the efficiency and effectiveness of the (organic) innovation transfer.

Transfer Dyad Characteristics and the Efficiency and Effectiveness of Lateral Innovation Transfers

Two variable characteristics of transfer dyads affect the efficiency and effectiveness of lateral innovation transfers: the degree of absorptive capacity (Cohen and Levinthal 1990), and dyadic willingness. Absorptive capacity has an important impact on innovation transfer (Gupta and Govindrajan 2000). Previous research has emphasized the importance of relative absorptive capacity--that is absorptive capcity in relation to a specific source unit from which knowledge may be acquired (Lane and Lubatkin 1998) this is clearly the relevant perspective from which to examine the effect of absorptive capacity on the efficiency and effectiveness of lateral innovation transfers. Relative absorptive capacity is largely a function of the degree of 'homophily' between the partners in a specific transfer dyad and is likely to vary across different transfer dyads. Homophily refers to the similarity between two (or more) interacting units in certain attributes, such as prior knowledge base, previous experience; technological similarity or cultural attributes (Szulanski 1996; Gupta and Govindrajan 2000; Cho and Lee 2004; Simonin 1999, 2004). When interacting units are similar and share a common knowledge base or share common meanings with a common sub-language, the transfer of new ideas, technologies, and product design specifications details are easier. Homophily is particularly helpful with respect to transferring the tacit aspects, such as know-how underlying subsidiary innovations. Similarly, homophily helps the transfer dyad to affect changes in the innovation to improve its fit with economic and intuitional imperatives of the recipient's host environment (Jensen and Szulanski 2004). Difficulties in conveying tacit know-how and implementing subtle changes in the innovation can be resolved when homophily is high as concerted dialogue and communication across the transfer dyad is facilitated. Thus with high homophily transfers costs are likely not to exceed greatly from the expected or minimum level. Transfer speeds are also likely to be relatively high. Likewise homophily is likely to enhance the prospect of the effectiveness of transfer across the dyad. By contrast, when homophily across the dyad is low, the partners may encounter difficulties in resolving unexpected problems arising during the transfer of the innovation. Similarly, low homophily also complicates the process of adapting the innovation to the recipients' environment due to difficulties in effective dialogue and information exchange across the dyad (Hennart and Larimo 1998).

However, the dissimilarity between two parties to a transfer can be alleviated to some extent if the two parties know each other well and thus have learned to work together. Prior collaboration experience improves absorptive capacity. Units who interact regularly, even if they start with little similarity, over time assimilate one another's knowledge and develop a common knowledge base and a common specialized set of terminologies. When two parties to a transfer have developed a strong relation prior to the transfer effort, they gradually develop a shared communication frame whereby each party has come to understand how the other party uses subtle phrases and ways of explaining difficult concepts (Uzzi 1997). Moreover, collaboration experience leads to a preference to work and share knowledge with each other as opposed to seeking knowledge resident in other subsidiaries with whom they do not share previous collaborative experience. Such strength in a dyadic transfer relation should therefore reduce transfer costs by reducing the time and effort required to understand and incorporate knowledge into the focal project. We therefore put forward the following hypotheses:

Hypothesis 1a: There is a positive relationship between dyadic similarity (homopily) and the efficiency of lateral innovation transfer.

Hypothesis 1b: There is a positive relationship between dyadic similarity and the effectiveness of lateral innovation transfer.

Hypothesis 2a: There is a positive relationship between previous collaboration experience of the dyad and the efficiency of lateral innovation transfer.

Hypothesis 2b: There is a positive relationship between previous collaboration of the dyad and the effectiveness of lateral innovation transfer.

Dyadic willingness also affects the performance of the process (Andersson et al. 2006). As a concept transfer willingness is defined at the level of dyad and is distinct from the source subsidiary's motivation to initiate transfer. There are a variety of reasons why transfer willingness may vary across different transfer dyads. Transfer willingness may be a consequence of high relative absorptive capacity; the closer the knowledge and technical expertise of the dyad partners, the easier it will be to come to a common understanding of the key obstacles in the specific transfer process and how these could be dealt with. This will enhance their willingness to provide resources to progress the transfer in a timely fashion. However whereas absorptive capacity as a concept mostly captures the cognitive aspects of the dyadic relationship, transfer willingness is more focused on the motivational dimension. Also it is possible that high absorptive capacity may be associated with low willingness if the two parties are competitors for resources within the organization or in the market (Birkinshaw and Lingblad 2005; Tsai 2002). Furthermore, dyadic willingness will be related to to the 'business logics' and interdependence of resources between the partners (Stahl 2004), independent of their level of absorptive capacity.

In any case the higher the transfer willingness, the more efficient and effective the transfer process is likely to be. This is because when both parties are willing partners in the dyads there is likely to be a greater attentiveness to transfer performance; to jointly anticipate problems needing resolutions and engage in timely collaboration and provide resources that will facilitate the progress transfer process. Thus we put forward the following hypotheses:

Hypothesis 3a: There is a positive relationship between transfer willingness in the dyad and the efficiency of lateral innovation transfer.

Hypothesis 3b: There is a positive relation between transfer willingness in the dyad and the effectiveness of lateral innovation transfer.

The Involvement of Headquarters in Lateral Innovation Transfers

Transfer projects often have third party inputs, but usually this is of an advisory nature-the source and the recipient unit may individually or together employ the services of technical experts or consultants (Szulanski 1996). However, these third parties do not stand in a superior hierarchical position to the transfer dyad. The HQ's deployment of its hierarchical authority to instruct particular lateral transfers and its subsequent participation in the transfer process has implications for the efficiency and effectiveness of the transfer process. We envisage that the HQ involvement in the transfer process affects the efficiency and effectiveness of lateral transfers both directly and indirectly, through moderating the effects of dyadic characteristics on transfer performance.

As a consequence of the direct involvement of HQ, the dyad partners can no longer solely utilize information exchange between themselves, or rely on shared tacit understandings they have gained through previous collaboration to speed up decision-making. All pertinent information needs to be shared with the HQ, who must approve all decisions. Furthermore, even though the HQ is not the recipient of the innovation, it may nevertheless be necessary for the source and the recipient subsidiaries to convey, individually, their own tacit knowledge to the parent, as without such tacit knowledge the HQ may be unable to participate effectively in the transfer process. Thus the source subsidiary will need to convey tacit knowledge regarding the innovation to the HQ; the HQ may have a general view of the strategic significance of an innovation but does not necessarily understand the subtle and context specific features that affect the transfer process. Likewise, the recipient's host environment may require subtle changes in innovation to affect the transfer (Jensen and Szulanski 2004) that HQ do not necessary appreciate and which the recipient will need to convey to. The participation of HQ will thus substantially increase the amount of information exchange rendering decision making more complicated and less timely compared to the situation where only the dyad is involved in the transfer. From the perspective of the dyad the time and resources devoted to affect the transfer are likely to be significantly higher than would be expected to be the case without the participation of HQ. Thus we put forward the following hypothesis:

Hypothesis 4a: There is a negative relationship between the direct involvement of HQ in a lateral innovation transfer and the efficiency of that transfer.

With regard to transfer effectiveness, the impact of HQ involvement is rather ambiguous. On the one hand, given that that the tacit and context specific nature of both the innovation and the required adaptations to be made, the source and recipient subsidiaries may perceive that the HQ do not fully appreciate subtle but important details of the innovation and the recipient's environmental imperatives and thus that compared to the dyad-only transfer outcome, the actual outcome suffers from compromises forced on the dyad by the HQ. This will indicate that the HQ involvement detracts from the effectiveness of lateral transfers. On the other hand, given that the HQ presumably attaches some importance to the transfer (that it has instructed), it will be anxious to ensure that the transfer progresses to a completion. This is reinforced by the fact that given its hierarchical authority, HQ can channel required resources to affect required adaptations and thus ensure the transfer's effective implementation. We thus put forward the following hypotheses:

Hypothesis 4b1: There is a negative relationship between the direct involvement of HQ in lateral innovation transfer and the effectiveness of that transfer.

Hypothesis 4b2: There is a positive relationship between the direct involvement of HQ in lateral innovation transfer and the effectiveness of that transfer.

With regard to the moderating effect of HQ involvement, one important point is that the HQ instruct lateral transfers that, presumably, the source and recipient subsidiaries will not themselves have initiated. Therefore the likelihood is that when HQ instruct particular lateral transfers, the dyad that is formed will not necessarily benefit from relational attributes beneficial to 'good' transfer outcomes. That is, compared to organically formed transfer dyads, instructed dyads will display lower dyadic absorptive capacity, collaborative experience and transfers willingness. HQ participation in the transfer process may also negatively affect the dyadic willingness as neither party will necessary view the innovation it has to adopt as necessarily being suitable or 'best' in its host market.

In addition, even if the level of dyadic absorptive capacity and willingness is high their effect on transfer efficiency and effectiveness may be reduced as a result of HQ direct involvement. This could happen if, for example, although the dyadic absorptive capacity may be high, due to the HQ involvement the dyadic parties may have lower interaction with each other and thus do not take advantage of the problem solving potential that absorptive capacity provides. Similarly, the impact of high willingness may be diluted if the subsidiaries have to devote more of their managerial focus and attention to the HQ at the expense of their dyadic partner. We thus put forward the following hypotheses:

[FIGURE 1 OMITTED]

Hypothesis 5a: The effect of relationship attributes in the transfer dyad (dyadic similarity, previous cooperation and transfer willingness) on the efficiency of lateral innovation transfer is negatively moderated by the direct involvement of HQ.

Hypothesis 5b: The effect of relationship attributes in the transfer dyad (dyadic similarity, previous cooperation and transfer willingness) on the effectiveness of lateral innovation transfer is negatively moderated by the direct involvement of HQ.

Figure 1 summarizes the hypothesized model, showing the expected effects of the dyadic relationship between the source and recipient subsidiaries on the transfer performance of the innovation, and the direct and moderating effect of HQ involvement on transfer performance.

Methods

Data Collection

This study is part of a larger project on the development and transfer of innovations in MNCs (the TIME-project) conducted at Uppsala University (e.g. Ciabuschi 2004; Stahl 2004; Persson 2006b; Hallin 2008; Tsai 2009). The unit of analysis in this particular paper is lateral transfer 'projects' (or the transfer dyad). Each project is the transfer of a specific innovation of a source subsidiary to a recipient subsidiary within the same multinational company. A source subsidiary's innovation may be transferred a number of times each time to a different recipient, yielding a distinctive transfer 'project'. The data for the empirical analyses is from a sample population of European- and US-based MNCs that are heavily involved in technology development and active in the manufacturing industry.

After several pretests of the questionnaire on MNC managers, convenient sampling was adopted owing to the lack of a comprehensive list of innovations in MNCs. The sampling of innovation was done initially in cooperation with HQ managers (e.g. CEOs, heads of divisions, and R&D directors). MNC managers were provided with the following selection criteria for innovation transfer cases: i) the innovation(s) was (were) of significance to the subsidiary; ii) they had the potential to be transferred; and iii) they had been completed no earlier than ten years ago and no later than one year prior to the interview. After the innovations were identified, we focused on those innovations that were developed at subsidiary level and later transferred to other fellow subsidiaries. However, reverse transfer from subsidiaries to HQ were excluded from the research sample. Also those innovations that were firstly spilled-over from HQ were not included in the research as the source subsidiary only serves as an intermediary in the innovation flow chain and has limited influence in the transfer process. During first stage interviews, MNC HQ or divisional HQ provided the researcher with access to subsidiary-level managers after a list of transferred innovations was decided.

At the second stage of data collection, source subsidiary managers, rather than HQ, were the respondents. Face-to-face interviews were conducted at source subsidiary site to administer a standardized questionnaire to key personnel who had been involved in the processes of development and transfer of the investigated innovations. Face-to-face interviews enhanced the researcher's understanding of the respondents' answers by the ability to discuss questions with the respondent and consequently they developed a deeper knowledge about the phenomenon in question (Sapsford and Jupp 2006). Another advantage of face-to-face interviews is that the researcher can ensure that the interviewee is a knowledgeable person about the transfer project. This cannot be achieved by mail survey. Beside, face-to-face interview minimizes respondents' difficulties in understanding the exact meaning of a question, quantifying their answers, or agreeing on the terminology used. It also allows interviewers to supply immediate support and explanation of the questions. It could reduce the possibility of misunderstanding and misinterpretation of research indicators. Also the presence of the researcher while respondents are answering the questions is expected to decrease the number of items omitted (Webster 1997). Thus, the data collected is assumed to yield better quality than through mail survey.

A total of 83 innovations developed by subsidiaries involving a total of 162 innovation transfer cases were collected from 2003 to 2006. Amongst 162 transfers, 33 cases are reverse transfers which were sourced from subsidiaries back to home countries thus are eliminated in the further analysis. In total, the sample is composed of 60 different innovations from 35 locations of 19 MNCs yielding 129 studied transfer projects.

Construct Measurement

Transfer Efficiency and Effectiveness

Whereas the efficiency of innovation transfer reflects expected costs of transfer, the effectiveness reflects the completion of the transfer project. Two measurements of efficiency have been adopted: we asked the subsidiary respondent about i) the extent to which the actual costs of the innovation transfer were higher than expected. As a second measurement we asked about the ii) how quickly the innovation was adopted by the counterpart. Both measures used a 7-point scale (where 1 = totally disagree and 7 = totally agree). The average value of the indicators was used for the variable 'transfer efficiency' (Cronbach's alpha was 0.74).

Two measures were also used to indicate the effectiveness of innovation transfer: we asked the respondent to evaluate to what extent i) the innovation transfer had been completed and ii) if the performance of the innovation transfer was satisfactory. Both measurements were answered on a 7-point Likert scale where 1 = not at all and 7 = very high. Cronbach's alpha was somewhat moderate (0.63). However, whereas the alpha value is partially dependent on the number of items (George and Mallery 2003), the level of 0.6 has been used as threshold level when the number of items are low (Robinson et al. 1991). We therefore chose to include the construct in the analysis and used the average value of the indicators for the variable 'transfer effectiveness'.

With regard to the measurement of transfer efficiency and effectiveness of innovation transfer, it would have been better to have respondents from both source and recipient subsidiaries. Cost and access limitations made this impractical. However as this study is focused on product and process innovations the biases resulting from only having respondents at the source subsidiary are somewhat ameliorated. Because the innovation is generated by the source subsidiary, it will have a close familiarity with technical and organizational steps that will assist efficient transfer. Also the adoption of the innovation by the recipient will have transparency to the source subsidiary. The innovations are mostly new products and processes and the source subsidiary will know whether they were adopted by recipients or not; compared to the transfer of an 'organizational practices' (Kostova and Roth 2002), the recipient will have less scope to adopt the innovation only 'ceremonially'; although it may choose not to market the innovation forcefully in its host country. It is also relevant that the innovations were less than 10 years old (and transfers are likely to be more recent still); also on average subsidiary innovations were transferred less than 3 times. Therefore the subsidiary respondent would not be significantly affected by memory or recall constraints (Huber and power 1985) and their assessment of the performance of innovations transfers are likely to be reliable.

The Attributes of the Dyadic Relationship

The key attributes of the (dyadic) transfer relationship are dyadic absorptive capacity (Cohen and Levinthal 1990; Lane and Labutkin 1998; Gupta and Govindarajan 2000) and dyadic willingness (Tsai 2000, 2002; Birkinshaw and Lingblad 2005). Dyadic absroptive capcity is indicated by two variables in the analysis: partner similarity and partner collaborative experience. First, "Partner similarity" included two measurements: to what extent i) technical differences and ii) organizational differences between the transfer partners have made the transfer problematic. Both items were measured on a 7-point Likert scale, where 1 = very little and 7 = very much. The average value of the indicators was used (Cronbach's alpha = 0.77). Second, "Dyadic experience of cooperation" includes two measurements: to what extent the transfer partners i) had previously cooperated and ii) previously shared knowledge. The same type of Likert scale was used and Cronbach's alpha was 0.76. The variable "Dyadic transfer willingness" is measured by the flowing three measurement scales: to what extent lack of resources within the innovation sending and receiving subsidiaries, respectively, had made the transfer difficult (the two measurements were reverse coded). The third measurement of 'willingness' captured iii) to what extent the innovation was difficult to use in the counterpart's business (also reverse coded). A 7-point Likert scale was used for the three measurements and the average value was used (Cronbach's alpha = 0.74).

Headquarters Involvement

As it is conceptualized in this study, HQ involvement in lateral innovation transfers is intended to capture the extent to which HQ both initiate transfers and participate in the post initiation phase of the transfer process. For the construct of HQ involvement, only one previous study exits (Ciabuschi and Martin 2010). Four items have been used to measure various forms of involvement. We asked to what extent: i) the HQ has formally instructed the subsidiary to share the innovation with the counterpart, ii) the HQ had themselves been involved in conducting the actual transfer, iii) the HQ had taken complete responsibility for the transfer, and, iv) that the transfer was driven by requirement from HQ. The four items were measured on a 7-point Likert-type scale (1 = not at all, 7 = very much). The average value of the indicators was used for the variable 'HQ involvement'.

Controls

The efficiency and effectiveness of innovation transfer in lateral relationships may depend on characteristics of the MNC organization structure. We therefore included a control for the "Autonomy" of the subsidiary, measured as the degree to which the subsidiary is able to decide about its investments in i) R&D and in ii) acquisitions, as well as the introduction of new products in its iii) domestic and iv) international markets. The four items were measured on a 7-point Likert scale, where 1 equals to 'HQ decides alone' and 7 equals to 'subsidiary decides alone', and averaged to create the variable 'autonomy' (Cronbach's alpha = 0.75). For the same reason we also controlled for the degree to which knowledge transfer is part of the evaluation criteria of HQ and thus the "Incentives" affecting transfer efficiency and effectiveness. Five items were used and measured on a 7-point Likert scale: i) the importance of knowledge sharing as a criterion used by divisional/business area HQ when measuring the subsidiary's performance, ii) the existence of financial bonuses for knowledge sharing, iii) the favoring of individual promotion for knowledge sharing, iv) the relatedness of this particular transfer to the evaluation system of HQ, and, v) the awareness among employees of the existence of bonuses for knowledge sharing. The average value of the indicators created the variable 'HQ incentives' (Chronbach's alpha = 0.71).

As transfer is affected by the inherent characteristics of knowledge (Szulanski 1996), we also control for the innovation's degree of "explicitness" and "complexity". For example, the ability of transfer may be affected by the stickiness of the innovation. Complex innovations or innovations comprising high degree of tacitness may not only require costly transfer procedures and adaptation among receivers but also additional resources from HQ. Two items were used for "Innovation explicitness" and measured on a 7-point scale: i) the innovation technology/process know-how is more explicit than tacit, ii) the innovation technology/process know-how is easily codifiable. An average of the items created the variable (Cronbach's alpha = 0.66). Complexity was measured in the same way using three items: i) the innovation can satisfy a large number of functions, ii) the innovation comprises a high number of interacting sub-systems and components, iii) the innovation comprises a high variety of functional knowledge bases. The average value of the indicators was used for the variable (Cronbach's alpha = 0.84).

Harman's single-factor test has been applied to address the issue of common method when using single-source data (Jarvis et al. 2003). After loading all of the research variables into an exploratory factor analysis, the unrotated factor solution shows that 8 extracted factors accounted for 71% of the total variance. Since several factors were extracted, and as the first of these only accounted for a small proportion of the variance, common-method bias did not appear to be a problem Further, Cronbach's alpha (values have been provided above) and Dillon-Goldstein's Rho (Nunnally 1978) were used to evaluate the internal consistency reliability. With only two of them being slightly below the 0.7 threshold the internal consistency reliabilities of reflective measurement model are acceptable. In addition to the reliability assessment, convergent validity and discriminant validity are also tested. With all average variance extracted (AVE, i.e. the average variance shared between a construct and its indicators) greater than 0.6, it suggests a decent convergent validity of measurement models (Fornell and Larcker 1981). By further examining the cross-loadings of items, the study further confirm the discriminant validity of reflective measurements under study. Detail correlation matrix of research factors is presented in Table 1.

Empirical Analyses

Multiple regression analysis was employed to carry out the hypotheses testing. Table 2 shows the results of regression analysis on transfer efficiency and effectiveness respectively. First, transfer efficiency was tested in models 1, 2 and 3. Model 1, which only included the control variables, reveals no significant effects. The model has an insignificant F-value of 0.965 and a very low adjusted R2 of -0.001. In Model 2 the main variables of relationship attributes; collaboration experience, similarity and willingness were included in the test. This model reveals a significant F-value of 9.231 (p < 0.001) and an adjusted R2 of 0.370. Next, in addition to the variables in Model 2, Model 3 includes the moderating interaction effects of HQ involvement and the relational attributes of the transfer partners. This model shows an adjusted R2 of 0.383 and a significant F-value of 7.327 (p<0.001).

Regarding Transfer efficiency, the results of the empirical analysis indicate that neither 'similarity' nor 'collaboration experience' have significant effects, which is in contrast to Hypotheses la and 2a. The results further show that 'willingness' is positively related to the efficiency of innovation transfer, which supports Hypothesis 3a ([beta] = 0.531, p< 0.001). Thus, 'willingness' seems to be a stronger predictor of efficiency as compared to the relationship "homophily" and absorptive capacity arguments. Among the control variables we find that the 'explicitness' of the innovation, the subsidiary's 'autonomy' (Model 2) and the 'HQ incentives' for knowledge transfer (Model 2) are positively related to the transfer efficiency. However, the involvement of HQ is significantly and negatively associated to transfer efficiency, which is in support of Hypothesis 4a ([beta] = -0.253, p<0.01). Thus, HQ tends to be more associated to lateral transfer projects when they are costly and time-consuming (inefficient).

Model 3 then tests the moderating effect of HQ involvement. The model presents the consequences of the interaction between the HQ involvement and the effect of the dyadic relationship on transfer efficiency. In accordance with the procedures for testing interaction effects suggested by Aiken and West (1991), the independent variables were standardized before creating the multiplicative terms, thereby reducing possible distortion caused by strong correlations between the interaction term and its components. We find that the effect of 'dyadic similarity' on transfer efficiency is negatively moderated by the involvement of HQ in the innovation transfer process ([beta] = -0.164, p<0.1). This is in support of Hypothesis 5a. However, we get no support for a moderating effect on transfer efficiency as regards the effect of 'collaboration experience' or 'willingness' of the dyadic partners.

Next, Transfer effectiveness was tested in Model 4, 5 and 6. Model 4, which included the control variables only, reveals no significant effects: The insignificant F-value is 0.613 and the adjusted R2 is -0.014. In Model 5 the main variables of dyadic similarity, collaboration experience and willingness were included in the test. This model reveals a significant F-value of 12.645 (p<0.001) and an adjusted R2 of 0.454. In addition to the variables in Model 5, Model 6 includes the moderating interaction effects of HQ involvement and the relational attributes of the transfer partners. This model shows an adjusted R2 of 0.458 and a significant F-value of 9.587 (p<0.001).

Regarding Transfer effectiveness, the results show that 'collaboration experience' and 'willingness' have significant and positive effects ([beta] = 0.198, p<0.05 and [beta] = 0.609, p < 0.001), whereas 'similarity' is not significant (see Model 5). Thus, whilst Hypotheses 2b and 3b are supported we receive no support for Hypothesis lb. The direct involvement of HQ was not significant and, hence, hypotheses 4b1 or 4b2 were not supported. Among the control variables we find that the characteristics of the innovation; 'explicitness' and 'complexity' had significant effects on 'transfer effectiveness' ([beta]=0.203, p<0.01 and [beta] = 0.170, p < 0.05). Thus, when the innovation is tacit and/or complex the performance and completeness of the transfer are less likely to be satisfactory. We also found a positive effect of 'HQ incentives', through evaluation of subsidiary knowledge transfer, on 'transfer effectiveness' ([beta]=0.192, p <0.05, see Model 5).

The moderating effect of HQ involvement was tested in Model 6, investigating the consequences of the interaction between the HQ involvement and the effect of the dyadic relationship variables on transfer effectiveness (the same procedure of standardization was used as in the case of transfer efficiency above). The results show that the effect of 'willingness' on 'transfer effectiveness' is negatively moderated by the involvement of HQ in the innovation transfer process ([beta] = -0.153, p < 0.1). This is in support of Hypothesis 5b. However, HQ involvement did not moderate the effects of 'collaboration experience' and 'similarity' on 'transfer effectiveness' (which was not in support of Hypothesis 5b). Thus, 'dyadic willingness', which has substantial positive impact on transfer effectiveness, significantly declines in its importance to effectiveness as HQ's involvement increases. To put this findings into context, HQ's involvement in the transfer project where both counterparts show high willingness toward the transfer is unnecessary and damaging the completeness and satisfaction of transfer process. Hence, although HQ involvement does not significantly impair effectiveness directly, a negative impact is taking place via moderating effect on dyadic willingness.

In sum, 'similarity' had no effect on either efficiency or effectiveness of transfer. 'Collaboration experience' had no effect on 'efficiency', but was significant and positively related to 'effectiveness'. 'Willingness' was positively related to both 'efficiency' and 'effectiveness'. 'HQ involvement' was negatively related to 'efficiency', but had no direct effect on 'effectiveness'. Further, 'HQ involvement' had a negative moderating effect on the link between 'similarity' and 'efficiency', but was unrelated to the effects of 'collaboration experience' and 'willingness' on 'efficiency'. Concerning 'effectiveness', HQ involvement moderated the effect of 'willingness' negatively, whereas it was unrelated to the effects of 'collaboration experience' and 'similarity'. Thus, Hypotheses 2b, 3a, 3b, 4a, were supported whilst we had a partial support for the moderating negative effect of HQ involvement as expressed in Hypotheses 5a and 5b. Consequently, we received no support for Hypotheses 1a, 1b, 2a, 4b1, 4b2 and, partly, 5a and 5b.

Conclusion: Discussion of Findings, Limitations and Implications for Future Research

Discussion of Findings

Although a number of recent studies have indicated a significant departure from 'federal' structures in MNCs with increasing central control (Yamin and Forsgren 2006; Buckley and Ghauri 2004; Rugman and D'Cruz 2000), the focus has been more on HQ reconfiguration of production and operational aspects of MNC activity but not specifically on HQ stance with respect to knowledge creation and transfer aspects. Some degree of product development and associated R&D activity has been outsourced to 'first tier' suppliers (Nolan 2002) specifically as a result of advancements in information communication technologies (Jean et al. 2010a, b) but internal creation and transfer of knowledge is clearly remaining an important facet of the strategic development of MNCs. Most studies stress that product development and design capabilities are retained within the MNE and are less likely to be outsourced. In fact a frequent rationale for outsourcing is to enhance the focus of the MNE on product development and design capabilities as key core competencies (Buckley 2010). Thus departure from a federal structure implies a greater HQ engagement with issues relating to innovation and knowledge development.

From this perspective, the findings of this study suggest that the 'hands on' approach of managing lateral transfers confronts the HQ in the MNC with a dilemma. This is because the organic innovation transfers benefit from a background of inter-subsidiary networks that tends to enhance their efficiency and effectiveness. The strong impact of dyadic willingness on efficiency and effectiveness of lateral transfers is a particularly important finding in this context; it affirms suggestions in the literature that inter-subsidiary relational harmony and social capital are particularly important drivers in the transfer process (Tsai 2000; Tsai and Ghoshal 1998; Gnyawali et al. 2009). Our findings indicate that the dyadic willingness is considerably more important than even absorptive capacity in facilitating a smooth transfer process. Furthermore the positive effect of the (transfer focused) evaluation scheme on transfer effectiveness (and, partially, on transfer efficiency) is also testimony to the dilemmas faced by the HQ. On the one hand HQ attach importance to subsidiaries transferring their innovations while at the same time instructing and participating in transfer projects that, as we noted previously, subsidiaries would not have opted for.

Although HQ may directly involve themselves to achieve transfers that are perceived as compatible with their knowledge development strategy, it is also the case that HQ may be disadvantaged when managing the inter-subsidiary transfer processes from 'above'. Thus the direct involvement of HQ in knowledge transfer imposes some negative consequences in terms of incurring a loss of efficiency without any clear gains in effectiveness. The literature on subsidiary knowledge creation has strongly suggested that subsidiary innovation frequently bears context specific features that would affect its deployment in other units (Andersson et al. 2001). This is highly germane to understanding why HQ involvement is likely to have negative efficiency consequences on lateral transfers. Thus either the source subsidiary needs to spend resources in conveying highly tacit knowledge to the HQ or comply with 'short cuts' and compromises dictated by HQ, leading to a diminution in the efficiency of the transfer. Similar difficulties arise in the context of the recipient where the effective adaptation of the innovation to the host environment requires a close understanding (entailing much tacit knowledge) of institutional and other environmental conditions in the host environment that, again, the HQ is likely to lack. Thus the basic issue here is that while the HQ are able to exert hierarchical authority to ensure a particular transfer, they lack the necessary knowledge to effectively participate in the transfer process. The findings of our study in relation to the moderating effects of HQ involvement reinforce the above comments. Notably, HQ involvement reduces the effects of partner absorptive capacity ('similarity') and, in particular, dyadic willingness, in the transfer process. Thus, in one sense, the dilemma faced by the HQ is that in 'suppressing' the 'organic' transfer process, HQ deprive itself of a significant relational resource embodied in inter-subsidiary ties. Of course, the issue for the MNC is whether this relational asset is utilized in full compatibility with strategic imperatives for the MNC as a whole.

This paper has contributed to our knowledge of lateral innovation transfers in MNCs. It has made a significant contribution to understanding the factors that affect the efficiency and effectiveness of lateral transfers and, in particular to how HQ involvement affects the process. The paper points to significant dilemmas faced by HQ in their management of lateral transfers: on the one hand, the network structure that typifies many MNCs has beneficial impacts on the performance of the transfer process; on the other hand the HQ need to ensure that lateral transfers initiated by subsidiaries ('organic' transfers) are in line with MNC strategy. However the findings from our study strongly suggest that HQ direct involvement in the transfer process is at best a very blunt instrument for achieving a resolution of this dilemma.

Limitations and Future Research

This study set out to investigate the effect of HQ direct involvement on the efficiency and effectiveness of lateral innovation transfers in MNCs. We have deliberately adopted a narrow perspective of HQ involvement as a reflection of a heightening of strategic control by HQ. Our findings therefore should be treated with some caution: although they show a negative relationship between the involvement of HQ and transfer efficiency, we cannot rule out the possibility that such involvement is greater when lateral transfers are somewhat more difficult; and thus prone to greater inefficiency. Further research is needed to establish the direction of causation more unambiguously. A related point is that our paper has not empirically investigated the antecedents of HQ involvement, only its effects. So, clearly, there is ample scope for further research to disentangle the various antecedents of HQ, allowing for the main key roles of HQ; such as problem solving and as its role organizing the process of combining dispersed capabilities within the MNC (Kogut and Zander 1992). Another fruitful line of research is to further investigate the moderating effects of HQ involvement on the relational attributes of transfer dyads such as dyadic willingness. Our results indicate the moderating effects are likely to be negative. Future studies can fruitfully investigate whether and why this may be the case, possibly through case studies of HQ involvement in lateral transfers.

DOI 10.1007/s11575-011-0070-7

Received: 10.01.2010 / Revised: 23.05.2010 / Accepted: 05.07.2010 / Published online: 29.03.2011

References

Adenfelt, M., & Lagerstrom, K. (2006). Knowledge development and sharing in multinational corporations: The case of a centre of excellence and a transnational team. International Business Review, 15(4), 381-400.

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park: Sage Publications.

Almeida, P., & Phene, A. (2004). Subsidiaries and knowledge creation: The influence of the MNC and host country on innovation. Strategic Management Journal, 25(8-9), 847-864.

Andersson, U., & Holm, U. (2002). Managing integration of subsidiary knowledge in the multinational corporation: A note on the role of headquarters. In M. F. V. Havila & H. Hakansson (Eds.), Critical perspectives on internationalisation (pp. 359-385). Uppsala: Elsevier.

Andersson, U., & Holm, U. (2010). Introduction and overview. In U. Andersson & U. Holm (Eds.), Managing the contemporary multinational: The role of headquarters (pp. 1-29). Cheltenham: Elgar.

Andersson, U., Forsgren, M., et al. (2001). Subsidiary embeddedness and competence development in MNCs--a multi-level analysis. Organization Studies, 22(6), 1013-1034.

Andersson, U., Mudambi, R., et al. (2006). Unpacking inter-unit knowledge transfer in multinational corporation. AIB UK Annual Conference, Manchester.

Argyres, N. S. (1995). Technology strategy, governance structure and interdivisional coordination. Journal of Economic Behavior & Organization, 28(3), 337-358.

Birkinshaw, J., & Hood, N. (1998). Multinational subsidiary evolution: Capability and charter change in foreign-owned subsidiary companies. The Academy of Management Review, 23(4), 773-795.

Birkinshaw, J., & Lingblad, M. (2005). Intrafirm competition and charter evolution in the multi-business firm. Organization Science, 16(6), 674-686.

Birkinshaw, J., Hood, N., et al. (1998). Building firm-specific advantages in multinational corporations: The role of subsidiary initiative. Strategic Management Journal, 19(3), 221-241.

Birkinshaw, J., Holm, U., et al. (2000). Consequences of perception gaps in the headquarters-subsidiary relationship. International Business Review, 9(3), 321-344.

Bjorkman, I., Barner-Rasmussen, W., et al. (2004). Managing knowledge transfer in MNCs: The impact of headquarters control mechanisms. Journal of International Business Studies, 35(5), 443-455.

Bouquet, C., & Birkinshaw, J. (2008). Weight versus voice: How foreign subsidiaries gain attention from corporate headquarters. Academy of Management Journal, 51(3), 577-601.

Buckley, P. (2010). The role of headquarters in the global factory. In U. Andersson & U. Holm (Eds.), Managing the contemporary multinational: The role of headquarters (pp. 60-84). Cheltenham: Elgar.

Buckley, P. J., & Ghauri, P. N. (2004). Globalisation, economic geography and the strategy of multinational enterprises. Journal of International Business Studies, 35(2), 81-98.

Cantwell, J., & Mudambi, R. (2005). MNE competence-creating subsidiary mandates. Strategic Management Journal, 26(12), 1109-1128.

Cho, K. R., & Lee, J. (2004). Firm characteristics and MNC's intra-network knowledge sharing. Management International Review, 44(4), 435.

Ciabuschi, F. (2004). On the innovative MNC: Leveraging innovations and the role of it systems. Doctoral thesis, Department of Business Studies, Uppsala University, Uppsala.

Ciabuschi, F., & Martin, O. (2010). Determinants of HQ's involvement in innovation transfer. In U. Andersson & U. Holm (Eds.), Managing the contemporary multinational: The role of headquarters (pp. 182-210). Cheltenham: Elgar.

Cohen, R. B., & Felton, N. (Eds.). (1979). The multinational corporation: A radical approach--papers by Stephen Herbert Hymer. Cambridge: Cambridge University Press.

Cohen, W., & Levinthal, D. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35, 128-152.

Daft, R. L. (1992). Organization theory and design (4th ed.). St. Paul: West.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.

Foss, N. J., & Pedersen, T. (2002). Transferring knowledge in MNCs: The role of source of subsidiary knowledge and organizational context. Journal of International Management, 8(1), 49-67.

Frost, T. S. (2001). The geographic sources of foreign subsidiaries' innovations. Strategic Management Journal, 22(2), 101-123.

Frost, T. S., Birkinshaw, J. M., et al. (2002). Centers of excellence in multinational corporations. Strategic Management Journal, 23(11), 997-1018.

George, D., & Mallery, P. (2003). SPSS for windows step by step: A simple guide and reference, 11.0 update (4th ed.). London: Pearson Allyn & Bacon.

Gnyawali, R. D., et al. (2009). Knowledge ties among subsidiaries in MNCs: A multi-level conceptual model. Journal of International Management, 15(4), 387-400.

Gupta, A. K., & Govindarajan, V. (2000). Knowledge flows within multinational corporations. Strategic Management Journal, 21(4), 473-496.

Hallin, C. (2008). Subsidiaries as sources and recipients of innovations in the multinational corporation. Doctoral thesis, Department of Business Studies, Uppsala University, Uppsala.

Hennart, J.-F., & Larimo, J. (1998). The impact of culture on the strategy of multinational enterprises: Does national origin affect ownership decisions? Journal of International Business Studies, 29(3), 515-538.

Huber, G. P., & Power, D. J. (1985). Retrospective reports of strategic-level managers: Guidelines for increasing their accuracy. Strategic Management Journal, 6(2), 171-180.

Jarvis, C. B., Mackenzie, S. B., & Podsakoff, P. M. (2003). A critical review of construct indicators and measurement model misspecification in marketing and consumer research. Journal of Consumer Research, 30(2), 199-218.

Jean, R.-J. B., Sinkovics, R. R., & Cavusgil, S. T. (2010a). Enhancing international customer-supplier relationships through IT resources: A study of Taiwanese electronics suppliers. Journal of International Business Studies, 41(7), 1218-1239.

Jean, R.-J. B., Sinkovics, R. R., & Kim, D. (2010b). Drivers and performance outcomes of relationship learning for suppliers in cross-border customer-supplier relationships: The role of communication culture. Journal of International Marketing, 18(1), 63-85.

Jensen, R., & Szulanski, G. (2004). Stickiness and the adaptation of organizational practices in cross-border knowledge transfers. Journal of International Business Studies, 35(6), 508-523.

Kogut, B., & Zander, U. (1992). Knowledge of the firm, combinative capabilities, and the replication of technology. Organization Science, 3(3), 383-397.

Kostova, T., & Roth, K. (2002). Adoption of an organizational practice by subsidiaries of multinational corporations: Institutional and relational effects. Academy of Management Journal, 45(1), 215-233.

Lane, P. J., & Lubatkin, M. (1998). Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19(5), 461-477.

Minbaeva, D., Pedersen, T., et al. (2003). MNC knowledge transfer, subsidiary absorptive capacity, and hrm. Journal of international Business Studies. 34(6), 586-599.

Monteiro, L. F., Arvidsson, N., et al. (2008). Knowledge flows within multinational corporations: Explaining subsidiary isolation and its performance implications. Organization Science, 19(1), 90-107.

Morgan, G. (2001). The multinational firm: Organizing across institutional and national divides. In G. Morgan, P. H. Kristensen, & R. Whitley (Eds.), The multinational firm: Organizing across institutional and national divides (pp. 1-24). Oxford: Oxford University Press.

Mudambi, R., & Navarra, P. (2004). Is knowledge power? Knowledge flows, subsidiary power and rent-seeking within MNCs. Journal of International Business Studies, 35(5), 385-406.

Nolan, P., Sutherland, D., et al. (2002). The challenge of the global business revolution. Contributions to Political Economy, 21(1), 91-110.

Nunnally, J. C. (1978). Psychometric theory (2nd ed.). New York: McGraw-Hill.

O'donnell, S. W. (2000). Managing foreign subsidiaries: Agents of headquarters, or an interdependent network? Strategic Management Journal, 21(5), 525-548.

Pearce, R. (1999). The evolution of technology in multinational enterprises: The role of creative subsidiaries. International Business Review, 8(2), 125-148.

Persson, M. (2006a). The impact of operational structure, lateral integrative mechanisms and control mechanisms on intra-MNE knowledge transfer. International Business Review, 15(5), 547-569.

Persson, M. (2006b). Unpacking the flow: Knowledge transfer in MNCs. Doctoral thesis, Department of Business Studies, Uppsala University, Uppsala.

Phene, A., & Almeida, P. (2008). Innovation in multinational subsidiaries: The role of knowledge assimilation and subsidiary capabilities. Journal of International Business Studies, 39(5), 901-919.

Regner, P. (2003). Strategy creation in the periphery: Inductive versus deductive strategy making. The Journal of Management Studies, 40(1), 57-82.

Robinson, J. P., Shamver, P. R., et al. (1991). Criteria for scale selection and evaluation. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of social psychological attitudes. San Diego: Academic.

Rugman, A. M., & D'cruz, J. R. (2000). Multinationals as flagship firms: Regional business networks. Oxford: Oxford University Press.

Sapsford, R., & Jupp, V. (2006). Data collection and analysis (2nd ed.). London: SAGE.

Schulz, M. (2003). Pathways of relevance: Exploring inflows of knowledge into subunits of multinational corporations. Organization Science, 14(4), 440-459.

Simonin, B. L. (1999). Ambiguity and the process of knowledge transfer in strategic alliances. Strategic Management Journal, 20(7), 595-623.

Simonin, B. L. (2004). An empirical investigation of the process of knowledge transfer in international strategic alliances. Journal of International Business Studies, 35(5), 407-427.

Stahl, B. (2004). Innovation and evolution in the multinational enterprise. Doctoral thesis, Department of Business Studies, Uppsala University, Uppsala.

Subramaniam, M., & Venkatraman, N. (2001). Determinants of transnational new product development capability: Testing the influence of transferring and deploying tacit overseas knowledge. Strategic Management Journal, 22(4), 359-378.

Subramaniam, M., Rosenthal, S. R., et al. (1998). Global new product development processes: Preliminary findings and research propositions. The Journal of Management Studies, 35(6), 773-796.

Szulanski, G. (1996). Exploring internal stickiness: Impediments to the transfer of best practice within the firm. Strategic Management Journal, 17(Winter special issue), 27-43.

Teece, D. J. (1977). Technology transfer by multinational firms: The resource cost of transferring technological know-how. The Economic Journal, 87(346), 242-261.

Teece, D. J. (1986). Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy. Research Policy, 15(6), 285-305.

Tsai, W. (2000). Social capital, strategic relatedness and the formation of intraorganizational linkages. Strategic Management Journal, 29(2), 297-317.

Tsai, W. (2002). Social structure of "Coopetition" within a multiunit organization: Coordination, competition, and intraorganizational knowledge sharing. Organization Science, 13(2), 179-191.

Tsai, H.-J. S. (2009). Factors affecting the performance of inter-subsidiary innovation transfer in multinational corporations. Doctoral thesis, Manchester Business School, University of Manchester, Manchester.

Tsai, W., & Ghoshal, S. (1998). Social capital and value creation: The role of intrafirm networks. Academy of Management Journal, 41(4), 464-476.

Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embeddedness. Administrative Science Quarterly, 42(1), 35-67.

Webster, C. (1997). Effects of researcher presence and appeal on response quality in hand-delivered, self-administered surveys. Journal of Business Research, 38(2), 105-114.

Yamin, M., & Forsgren, M. (2006). Hymer's analysis of the multinational organization: Power retention and the demise of the federative MNE. International Business Review, 15(2), 166-179.

Yamin, M., & Ghauri, P. (2010). A critical assessment of the business network perspective on HQ control in multinational companies. In U. Andersson & U. Holm (Eds.), Managing the contemporary multinational: The role of headquarters (pp. 125-137). Cheltenham: Elgar.

Yamin, M., & Otto, J. (2004). Patterns of knowledge flows and innovative performance in MNEs. Journal of International Management, 10(2), 239-258.

Yamin, M., & Sinkovics, R. R. (2007). ICT and MNE reorganisation: The paradox of control. Critical Perspectives on International Business, 3(4), 322-366.

Yamin, M, & Sinkovics, R. R. (2010). ICT deployment and resource-based power in multinational enterprises. Futures, 42(9), 952-959.

Zhao, H., & Luo, Y. (2005). Antecedents of knowledge sharing with peer subsidiaries in other countries: A perspective from subsidiary managers in a foreign emerging market. Management International Review, 45(1), 71-97.

Prof. M. Yamin ([mail])

Department of Business Studies, University of Manchester, Manchester, UK

e-mail: mo.yamin@mbs.ac.uk

Prof. M. Yamin

Uppsala University, Uppsala, Sweden

Dr. H.-J. Tsai

Manchester Business School, University of Manchester, Manchester, UK

Prof. U. Holm

Department of Business Studies, Uppsala University, Uppsala, Sweden
Table 1: Correlation matrix of research factors

 Mean SD 1 2 3

1. Tr. efficiency 4.862 1.378 1.00
2. Tr. effectiveness 5.303 1.250 0.60 ** 1.00
3. Dyadic co-op. exp. 4.706 1.605 0.12 0.31 ** 1.00
4. Dyadic similarity 5.465 1.351 0.32 ** 0.34 ** 0.21 *
5. Dyadic willingness 5.415 1.351 0.44 ** 0.54 ** 0.20 *
6. Subsidiary autonomy 3.966 1.556 0.10 -0.04 0.05
7. HQ incentives 3.072 1.384 -0.01 0.04 -0.01
8. HQ involvement 3.110 1.853 -0.23 -0.07 0.04
9. Explicitness 2.438 1.514 0.06 0.01 -0.20 *
10. Complexity 4.425 1.567 -0.04 -0.10 -0.01

 4 5 6 7

1. Tr. efficiency
2. Tr. effectiveness
3. Dyadic co-op. exp.
4. Dyadic similarity 1.00
5. Dyadic willingness 0.44 ** 1.00
6. Subsidiary autonomy -0.08 * -0.16 1.00
7. HQ incentives -0.19 * -0.18 * -0.02 1.00
8. HQ involvement -0.22 * -0.02 -0.09 0.18 *
9. Explicitness 0.01 -0.20 * -0.05 -0.03
10. Complexity -0.20 * -0.07 0.14 0.41 **

 8 9

1. Tr. efficiency
2. Tr. effectiveness
3. Dyadic co-op. exp.
4. Dyadic similarity
5. Dyadic willingness
6. Subsidiary autonomy
7. HQ incentives
8. HQ involvement 1.00
9. Explicitness -0.03 1.00
10. Complexity 0.23 * 0.00

* Represents that correlation is significant at the 0.05 level
(2-tailed)

** Represents that correlation is significant at the 0.01 level
(2-tailed)

Table 2: Regression equations with efficiency and effectiveness
as dependent variables

 Dependent variables

 Efficiency

Model 1 2 3

Innovation 0.116 0.260 ** 0.242 **
explicitness
Innovation complexity -0.116 0.081 -0.060
Subsidiary autonomy 0.123 0.182 * 0.128
HQ incentives 0.051 0.163 ([dagger]) 0.124
Dyadic collaboration 0.062 0.024
experience
Dyadic similarity 0.037 0.131
Dyadic willingness 0.531 *** 0.508 ***
HQ involvement -0.253 ** -0.256 ***
HQ * collaboration 0.110
HQ * similarity -0.164 ([dagger])
HQ * willingness -0.076
R square 0.034 0.415 0.444
Adjusted R square -0.001 0.370 0.383
F-value 0.965 9.231 *** 7.327 ***

 Dependent variables

 Effectiveness

Model 4 5 6

Innovation 0.021 0.203 ** 0.161 *
explicitness
Innovation complexity -0.155 -0.170 * -0.187 *
Subsidiary autonomy -0.016 0.081 0.111
HQ incentives 0.081 0.192 * 0.171 *
Dyadic collaboration 0.198 * 0.219 **
experience
Dyadic similarity 0.041 0.045
Dyadic willingness 0.609 *** 0.610 **
HQ involvement -0.060 -0.025
HQ * collaboration 0.016
HQ * similarity 0.091
HQ * willingness -0.153 ([dagger])
R square 0.022 0.493 0.511
Adjusted R square -0.014 0.454 0.458
F-value 0.613 12.645 *** 9.587 ***

([dagger]) p < 0.10; * p < 0.05; ** p <0.01; *** p < 0.001
COPYRIGHT 2011 Gabler Verlag
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:RESEARCH ARTICLE
Author:Yamin, Mo; Tsai, Hsin-Ju "Stephie"; Holm, Ulf
Publication:Management International Review
Date:Mar 1, 2011
Words:10584
Previous Article:Exploring the effects of vertical and lateral mechanisms in international knowledge transfer projects.
Next Article:Effective global strategy implementation: structural and process choices facilitating global integration and coordination.
Topics:

Terms of use | Privacy policy | Copyright © 2022 Farlex, Inc. | Feedback | For webmasters |