مشارکت ستاد فرماندهی و بهره وری توسعه نوآوری و انتقال در شرکت های چند ملیتی: موضوع بی توجهی محض؟
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|4570||2012||15 صفحه PDF||سفارش دهید||10710 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : International Business Review, Volume 21, Issue 2, April 2012, Pages 130–144
We compare two theoretical perspectives on the knowledge situation of headquarters (HQ) in multinational corporations: the bounded rationality perspective and the sheer ignorance perspective. We claim that these perspectives lead to different expectations when it comes to HQ's role and the effects of HQ involvement in innovation processes at the subsidiary level. More specifically, we examine the impact of HQ involvement on the efficiency of 71 subsidiary innovation projects. The findings show that HQ involvement in innovation development and transfer has a negative rather than a positive impact on efficiency in both processes. We contribute by showing and suggesting that the sheer ignorance perspective might provide a more accurate portrayal of the HQ knowledge situation than the bounded rationality perspective does. This study furthers understanding of the role of HQ in the contemporary, knowledge-creating multinational.
Innovation processes, i.e., the development and transfer of innovations within the organization, are considered crucial activities for the contemporary multinational corporation (MNC). Thus, the corporate capabilities required for the efficient development and internal transfer of innovations leading to new products, production and marketing processes constitute the basis of competitive strength for the MNC (Cantwell, 1989, Forsgren et al., 2005, Ghoshal and Bartlett, 1988, Holm and Pedersen, 2000, Zander, 1991 and Zander and Kogut, 1995) and thus a major concern for managers and researchers. It has also been pointed out, however, that, due to the specific characteristics of the MNC as a geographically and functionally dispersed organization (Andersson et al., 2002, Birkinshaw and Hood, 2001, Cantwell, 1989, Mudambi and Navarra, 2004 and Rugman and Verbeke, 2001), innovation processes are largely carried out at subsidiary rather than HQ level. Much of the relevant expertise needed for innovating is therefore assumed to be more rooted in the subsidiaries’ local knowledge than in HQ's knowledge. At the same time, corporate HQ plays an important role as a resource contributor (e.g., Ambos and Schlegelmilch, 2007, Chandler, 1991, Ghoshal and Bartlett, 1988 and Poppo, 2003), coordinator, and decision maker (Foss, 1997 and Ghoshal and Bartlett, 1988). A closer look at the literature on the MNC as a knowledge-creating entity reveals divergent views on the role of HQ in MNC innovation processes. One dominant perspective, here called ‘bounded rationality’, views corporate HQ as having a reasonable possibility of monitoring the innovation process at the subsidiary level, albeit from a distance (e.g., Buckley and Hashai, 2009, Doz and Prahalad, 1981, Egelhoff, 1988, Ghoshal and Bartlett, 1988, Ghoshal and Nohria, 1997 and Hennart, 1993). Through its direct involvement in the process and through different control mechanisms, HQ has a decisive influence in shaping innovation processes. The fact that HQ may lack crucial knowledge of innovation processes at the local level and may therefore have limited possibility to contribute to the processes – let alone control them – is basically treated as a problem that can be solved through appropriate organization. Therefore, this view assumes that HQ can efficiently design and (at least indirectly) control innovation processes. Another perspective, which we call ‘sheer ignorance’, is less optimistic about the possibilities open to HQ. Scholars have pointed out that the ability of HQ to control processes at the subsidiary level and/or contribute to them with its own expertise is far from clear (Andersson et al., 2007, Barner-Rasmussen et al., 2010, Birkinshaw et al., 2000 and Goodall and Roberts, 2003). They stress that serious problems relating to HQ's knowledge and power limit, and sometimes totally inhibit, its possibility to influence innovation processes at the subsidiary level. Basically, the discrepancy between the two perspectives mirrors different views on what can be called the HQ knowledge situation. The ability of HQ to play a positive role and contribute to the efficiency of innovation processes at the subsidiary level is highly dependent on its level of access to relevant knowledge. If we assume that HQ either (i) actually possesses knowledge of crucial importance for a particular innovation process, or (ii) does not possess such knowledge but has a fair understanding of what kind of knowledge the process requires but HQ lacks, or (iii), even worse, not only lacks the knowledge but also any inkling of what knowledge it lacks, it has a profound impact on HQ's potential role. The issue obviously calls for further inquiry. The purpose of this paper is to confront the predictions of the bounded rationality perspective with those of the sheer ignorance perspective. In line with Scott, we define perspective as a conceptual umbrella “with a number of varying approaches that bear a strong family resemblance” (Scott, 1981, p. 55). In this sense, bounded rationality and sheer ignorance are perspectives “with respect to a limited, approximate, simplified model of the real situation” (March & Simon, 1958, p. 153). These perspectives are therefore difficult to measure and compare directly. However, we can compare whether or not the two ‘simplified models’ of the HQ knowledge situation lead to different predictions for a specific empirical phenomenon. The empirical phenomenon is the relationship between HQ involvement in innovation processes and the performance of these processes. More specifically, we examine the impact of HQ involvement in subsidiary-level innovation development and transfer processes on the efficiency4 of these processes. We define efficiency in terms of time and cost, i.e., as the amount of resources required to develop and/or transfer a certain innovation (Szulanski, 1996 and Verona, 1999) and argue that each perspective leads to different expectations regarding the effects of this involvement on the efficiency both in the development of an innovation at a specific subsidiary and in the transfer of this innovation to other subsidiaries. By confronting these expectations with empirical data, we assess the relative merits of the two perspectives for increasing our understanding of the potential role of HQ in innovation processes undertaken within MNCs with highly distributed knowledge-creating processes.5 In the next section, we construct two contrasting models and formulate three pairs of competing hypotheses concerning the impact of HQ involvement on subsidiary innovation efficiency. These hypotheses are then tested on a dataset of subsidiary innovations, and the results are presented and discussed. The final section of the paper outlines the implications of the study, advances future research issues and identifies its limitations.
نتیجه گیری انگلیسی
Although the empirical results appear to provide more support for the sheer ignorance perspective than for the bounded rationality perspective, they must be interpreted with caution, as there might be other explanations for the findings. First, the negative impact on efficiency produced by the involvement of HQ may simply reflect a ‘bureaucratization’ effect. When HQ becomes involved at the subsidiary level, this implies that another party must be informed of the various aspects of the innovation process and take part in the decision process along the way. This will probably increase the need for formalized communication channels and information exchange between the parties involved. The consequences of this bureaucratization will be higher information costs, a longer process, and, thus, lower efficiency. If this reasoning is correct, our results may be due more to higher administrative costs, resulting from the involvement of a greater number of actors, than to HQ's ignorance of the process as such. However, a relevant question that remains is why the knowledge, authority, and other resources contributed by HQ to the process do not appear to counterbalance the negative impact of this ‘bureaucratization’, resulting in greater rather than lower efficiency. Related to this issue is the fact that our dependent variable is efficiency, not effectiveness. We have defined efficiency in terms of time and cost, i.e., amount of resources required to develop and/or transfer an innovation ( Szulanski, 1996 and Verona, 1999). Effectiveness has to do, instead, with the development of an innovation of appropriate quality and specifications for user/market needs (Verona, 1999) and for application of the transfer at the recipient unit (Kostova, 1999). HQ involvement in the innovation process may well increase the cost of the process but it may also lead to more profitable and extensive use of the innovation in the MNC: that is, lower efficiency could be coupled with higher effectiveness. However, the few studies of efficiency and effectiveness in innovation development and transfer processes, such as Brown and Eisenhardt (1995), Olson, Walker, and Ruekert (1995), Pérez-Nordtvedt, Kedia, Datta, and Rasheed (2008), and Verona (1999), have generally demonstrated that the examined factors influence the two dimensions in a similar way, i.e., if the effect is positive on one dimension it is also positive on the other. In particular, Pérez-Nordtvedt et al. (2008), in their study of knowledge transfer, demonstrate how closeness and trust in the relationships between involved parties positively influence both efficiency and effectiveness. This is an important area for future research, as effectiveness is another relevant criterion for assessing the consequences of HQ involvement. However, we can speculate that if HQ is in a situation of sheer ignorance, it will be also when it comes to how innovations are used and marketed in different parts of the MNC, i.e., it is likely that HQ involvement will also generate less effectiveness. As a reflection of the managerial implications of our findings, we can put forward the following question: do our results mean that HQ should keep its hands off innovation processes at the subsidiary level? Not necessarily. Notwithstanding the efficiency versus effectiveness issue, authority exercised by headquarters can still be rational, even in circumstances of sheer ignorance. First, when several research projects are simultaneously in progress, a centralized authority, even if essentially ill-informed about the projects, may be needed to signal priorities. When the need for decisions and action is urgent, somebody has to decide (Foss, 2002). Second, a distinction between decisive and non-decisive knowledge can be made. In circumstances of sheer ignorance, if most of the knowledge that is hidden from HQ is non-hidden in character and wrong decisions are not too costly, an authority relationship may still be efficient (Casson, 1994). Finally, it can also be argued that, even when HQ does not understand the range of alternatives relating to research projects at the subsidiary level, it may still be able to estimate the ‘reasonable expected outcome’ and, in this sense, exercise its authority through monitoring (Foss, 2002). In any case, as a central research implication derived from our findings, we argue that it is time to question the dominant view of HQ as a central player designing and controlling innovation processes in the MNC under conditions of bounded rationality. This paper urges for further research on the relative merits of the two perspectives. For instance, it would be interesting to investigate to what extent HQ involves itself in innovation processes at the subsidiary level and why. The bounded rationality view would suggest that such involvement mirrors, more than anything else, HQ's actual knowledge about the local context in which the innovation process is carried out. Ceteris paribus, the more it knows, the more it interferes. Under the sheer ignorance perspective, in contrast, HQ's involvement would reflect role expectations rather than actual knowledge. The more other stakeholders expect it to interfere, the more it does so. Consequently, an empirical analysis of the impact of HQ's actual knowledge and actual role expectations on its involvement will shed more light on the relative strengths of the two perspectives. Future research on the sheer ignorance perspective could also be directed at HQ's role vis-à-vis other areas of subsidiary activities such as strategic behavior, internationalization and marketing. A reasonable hypothesis would be that the consequences of sheer ignorance at HQ level are different depending on the type of subsidiary activity under consideration. The consequences of genuine uncertainty, which is at the core of the sheer ignorance perspective, might be less serious for activities in which knowledge about the local context plays a minor role. Bounded rationality might be a more reasonable perspective from which to analyze the role of HQ in more standardized activities, while the sheer ignorance perspective might provide a more fruitful tool for understanding the same role vis-à-vis innovation processes. This is a fascinating area for future research. Finally, certain limitations suggest caution in the interpretation of our findings. First, due to difficulties in directly measuring and comparing perspectives – as emphasized at the beginning of this paper – our ‘indirect’ analysis does not necessarily imply that sheer ignorance is a more common knowledge situation and thus a ‘superior’ model of MNC HQ behavior. It does, however, represent a theoretical reflection and an empirical example in which it is difficult to explain a specific phenomenon from the bounded rationality perspective. Therefore, alternative perspectives within the rationality spectrum, such as the sheer ignorance approach, might be seen as more promising. Future research could consider the possibility of developing direct measures of the HQ knowledge situation and using them in different models. A second limitation is the cross-sectional nature of the study, which precludes causal inferences regarding the relationships between the various exogenous and dependent constructs tested in our model. It is time for a longitudinal approach, something still rarely found in international business research. A third is that, although one of the strengths of our study is the sample, in terms of its size and cross-country and cross-industry coverage, and its focus on the phenomenon of innovation development and transfer in business subsidiaries, we used non-probability sampling, as in most studies in the field (Yang, Wang, & Su, 2006). Studies using probability samples would therefore help to assess the potential generalizability of the findings.