The emergence of corporate international networks for the accumulation of dispersed technological competences.
Recent research using the framework of the competence-based approach to the multinational firm has attempted to trace out the technological evolution of large multinational corporations (MNCs) over time as a path-dependent learning process (Cantwell 1989, 1993, 1994). In the course of this process, MNCs move into new technological fields (they become technologically diversified), and they establish innovative activities in multiple geographical sites (their technological activity is internationalised). Some studies have now suggested that in an internationally integrated or 'globalised' MNC, the geographical dispersion of innovation may come to facilitate the technological development of the firm, since the MNC can tap into alternative streams of innovation in different centres, and establish favourable cross-border interactions between them (Cantwell 1995, Zander 1997, Dunning 1996). In this event, the internationalisation of technological activity will go more substantially beyond the role of facilitating the local exploitation of the MNC's core capabilities, and will instead become a more central part of the process by which its capabilities are developed and extended.
Elsewhere, we have shown that in fact the causality between the internationalisation and diversification of corporate technology runs in both directions (Cantwell/Piscitello 1997a). That is, the extent to which a firm has diversified its portfolio of technological capabilities influences (as well as being influenced by) the extent to which those capabilities are internationalised, and the linkage running in this other direction has been discussed especially by Granstrand (1998). However, the nature of this relationship has changed historically over time. Until quite recently, diversification and internationalisation could be seen as potential substitutes for one another. Up to the 1960s, increases in corporate competence were associated with either the greater diversification, or the greater internationalisation of technology, but not necessarily both (Cantwell/Piscitello 1997b). The individual firm at any point in time was confronted to some degree with a choice between these two possible strategies for growth and internal development, given that its resources were limited. Today, instead, at least for some corporate technology leaders, these strategies have become essentially complementary to one another, given that the firm's international network has matured sufficiently to provide it with the capacity needed to draw upon the new opportunities in the current international environment for the global or regional integration of activity. Increases in competence, together with its greater dispersion across fields of activity and across geographical sites, are likelier now to be combined, and to mutually reinforce one another.
In this paper, we focus on how the emergence of corporate international networks for technological development has led to the diversification or evolution of firm-specific competences into new fields of activity, through a restructuring of innovation across geographical sites to better access the local capabilities of each host country. Thus, we examine how the internationalisation and diversification of corporate technology have currently become more often positively related parts of a common process, rather than alternative ways in which competences might be developed (as it seems that they were in general in the past). While in our earlier studies we relied on a cross-firm statistical investigation over time, using panel data techniques, in this paper we explore the underlying data more closely, to examine at the level of individual firms the variety of associations between internationalisation and diversification that have emerged, today and in the past. We are particularly keen to identify those firms which are in some sense representative of the wider change in average behaviour, and in which a geographical dispersion of activity is now positively associated with a shift into new technological fields. Since one of our earlier studies (Cantwell/Piscitello 1997b) suggested that the emergence of a positive influence of internationalisation on diversification first occurred clearly in the 1980s, we compare the paths in the 1930s and in the 1980s of an historically consistent group of large firms, which have each held a continuous presence throughout that period. That is, we focus upon changes in the spread of corporate technology in 1930-1940, by comparison with the equivalent in 1980-1990.
In the next section, we discuss in a little more detail the explanation for the newly positive effect of an international span of corporate activity on the technological diversification of the firm, given the emergence of internationally integrated MNC networks. In the following section, we describe the data that we use in our study, in which corporate technological profiles are proxied by the US patenting activity of the largest US and European industrial companies historically. Our measures of internationalisation and diversification are then described. After this we present our findings, and discuss the implications. We begin by examining the shift towards a positive linkage between internationalisation and corporate technological diversification, and then deal in greater detail with the exact geographical and sectoral profile of the restructuring of the international MNC networks of technological development which had emerged by the 1980s. Finally, we draw some brief conclusions.
The Recent Effect of International Network Organisation on Corporate Technological Diversity Formerly local market oriented affiliates have been increasingly integrated into international networks within their respective MNCs, such networks coming to resemble 'heterarchies' more than hierarchies (Doz 1986, Hedlund 1986, Porter 1986, Bartlett/Ghoshal 1989, Dunning 1992). In technological activity, too, the location-specific capabilities of internationally dispersed MNC affiliates may have become more closely integrated than in the past, linked to a strategy for technology creation in the MNC as a whole, and not only with separate reference to each of the geographical parts of the company's business. This can be termed a new globalisation of technological innovation.
The theoretical rationale for the recent international integration of productive activity is that the economic benefits attributable to a more refined locational division of labour within the MNC have often come to outweigh the costs of being less nationally responsive in each market, costs associated with adverse political repurcussions and the continued national differentiation of demand (Doz 1986). In an integrated MNC network each affiliate specialises in accordance with the specific characteristics of local production conditions, technological capabilities and user requirements. The network benefits from economies of scale through the local concentration of particular lines of activity (increasing returns from local research in a specialised field as opposed to research in general), economies of locational agglomeration through an interchange with others operating in the same vicinity in technologically allied fields as suggested earlier, and economies of scope through the international intra-firm coordination of related but geographically separated activities. The experience acquired in a specialised activity in one location creates technological spillovers that can be passed on to other parts of the MNC network elsewhere. It has been shown that since the 1970s, in industries in which such net advantages to multinational integration were available, multinationality has been a source of competitive success and faster growth (Cantwell/Sanna-Randaccio 1993).
Other recent evidence suggests that this type of internationally integrated or globalised strategy for innovation is particularly characteristic of corporate technology leaders today (Cantwell 1995). The extent to which the affiliates of MNCs specialise within their industry across national boundaries in accordance with the comparative advantage of local expertise tends to rise as the technological strength of MNCs increases, and in particular tends to be greater for the leading MNCs that originate from the major locational centres of excellence for their industry (Cantwell/Sanna-Randaccio 1992, Cantwell/Janne 1999). The particular role of the largest and best known MNCs in the formation of internationally integrated networks for technological development is attributable partly to their capability to devote the resources needed to organise a complex organisational network, and partly due to their having a wide enough range of existing absorptive capacity (an established diversity of competences) to be able to effectively utilise and bring together a variety of new streams of innovation, each to some extent specific to their own particular local institutional setting or environment. For this reason we focus below on the largest firms (in terms of the volume of technological activity), and those which have established a sufficient degree of internationalisation in their strategies for technological development.
Recently, in at least the technologically leading firms, the nature of the competence creation process seems to have entered a new phase, based upon their ability to create an internal intra-firm but cross-border network for the development of both geographically and sectorally dispersed competences (Cantwell/Piscitello 1997b). What previously had been a dispersed set of loosely connected efforts for the consolidation and adaptation of competence within the firm (achieved through some combination of diversification and internationalisation), has been transformed in some companies into a more complex integrated and interactive network for the generation of new competence. This new system for corporate development relies on the interrelatedness between specialised activities conducted in particular locations, each of which takes advantage of spatially-specific resources or capabilities. In this event internationalisation, diversification and competence creation become for the first time necessarily interconnected and thus mutually positively related parts of a common process. Related to the shift towards this emergent diversity of competence creation in MNCs, others have commented upon the increasingly asset-acquiring role of foreign direct investment (Kogut/Chang 1991, Dunning 1992, 1995, 1996; Pugel/Kragas/Kimura 1996), and the shift in the composition of the foreign-located research laboratories of MNCs from the 'home-base exploiting' kind and instead towards the 'home-base augmenting' type (Kuemmerle 1996).
Our study here is based upon a database on the patenting activity in the US of the largest US and European companies over the period 1901-1990(1), developed at the University of Reading (see Cantwell 1995). The firms included in the database were identified in one of three ways. The first group consisted of those firms which have accounted for the highest level of US patenting after 1969; the second group comprised other US, German or British firms which were historically among the largest 200 industrial corporations in each of these countries (Chandler 1990); and the third group was made up of other companies which featured prominently in the US patent records of earlier years. In each case, patents were counted as belonging to a common corporate group where they were assigned to affiliates of a parent company.(2) The location of the original research facility that gave rise to each patent (the country of residence of the original inventor) is recorded in the data. The location of the parent company is another important dimension of the analysis, as this is treated as the home country or the country of origin of the corporate group. By consolidating patents attributable to international corporate groups, it is then feasible to examine the geographical distribution [TABULAR DATA FOR TABLE 1 OMITTED] of the technological activity of these firms (Cantwell 1995). In addition, the primary field of technological activity of each patent can be derived from the US patent class system, which provides a measure of corporate technological diversification.
In all, the historical path was traced of the US patenting activity from the beginning of the century of 857 companies or affiliates that together comprise 283 corporate groups. For our current purposes, we considered a sample of 40 firms (out of the original 283) in existence in both the 1930s and the 1980s (see Table 1). The firms have been selected on the basis of some geographical, industrial and dimensional criteria. In particular, we considered six national groups - each of which essentially has representation across a number of the industries considered(3) - and from each of which we extracted our chosen companies through a sequence of selection steps related to (i) the firm's size in 1990 and its size historically (in order to include the largest firms), and (ii) in cases in which the highest ranked firms were of a similar size or in which no company in a national group was especially large historically (or, less frequently, is not very large today), we took into account the degree of internationalisation (preferring firms with a higher degree or rising trend of internationalisation in at least one of our two periods).
The Measures of Technological Diversification and Internationalisation
A number of proxies have been proposed to capture a firm's technological diversification, understood as the extent of the spread of its technological activity across fields. Some authors have considered the number of technological fields in which the firm is active (e.g. Patel/Pavitt 1997) or the use of the Herfindhal index and the entropy index to measure the degree of dispersion across fields (e.g. Oskarsson 1993). Our proxy is based on the consideration that technological diversification is inversely related to the extent of the concentration of the firm's technological specialisation in favoured sectors. The firm's profile of specialisation can be measured by an index of its revealed technological advantage (RTA), which for each particular sector of technological activity is defined by the firm's share in that sector of US patents granted to companies in the same industry, relative to the firm's overall share of all US patents assigned to firms in the industry in question. Specifically, denoting as [P.sub.ij] the number of US patents granted in sector or activity j to firm i in a particular industry, then the RTA index is defined as follows:
[RTA.sub.ij] = ([P.sub.ij]/[[Sigma].sub.i] [P.sub.ij])/([[Sigma].sub.j] [P.sub.ij]/[[Sigma].sub.ij] [P.sub.ij])
The index varies around unity, such that values greater than one suggest that a firm is comparatively advantaged in the sector of activity in question relative to other firms in the same industry, while values less than one are indicative of a position of comparative disadvantage. Importantly, the use of the RTA index allows us to control for inter-sectoral and inter-firm differences in the propensity to patent (Cantwell 1993, 1995).
The degree of technological diversification of the firm is measured by the coefficient of variation of the RTA index across all the relevant sectors for the firm.(4) Therefore, for firm i in each period considered, the proxy [DIV.sub.i] will be:
[DIV.sub.i] = [CV.sub.i] = [[Sigma].sub.[RTA.sub.i]]/[[Mu].sub.[RTA.sub.i]]
where [[Sigma].sub.[RTA.sub.i]] is the standard deviation and [[Mu].sub.[RTA.sub.i]] is the mean value of the RTA distribution for the firm i.
A remark on the interpretation of this measure may not be out of place here. When [CV.sub.i] is low, the cross-sectoral distribution of [RTA.sub.i] is widely dispersed, that is the profile of technological specialisation of firm i, is highly diversified across fields. On the other hand, when [CV.sub.i] is high, the RTA distribution is highly concentrated in certain fields and the degree of diversification of the firm will be low. Thus, [CV.sub.i] constitutes an inverse measure of technological diversification.(5) With regard to the firm's degree of internationalisation, we use the foreign share, which is the proxy almost universally used in other empirical studies (e.g. Pearce 1993). For firm i, the proxy [FS.sub.i] is defined as the share of patenting that is attributable to research located outside its home country in each period considered.(6) The Shift Towards a Positive Linkage between Internationalisation and Greater Corporate Technological Diversity
The historical change between the 1930s and the 1980s in the association between the internationalisation and the diversification of corporate technology is shown in Tables 2-5. More precisely, Tables 2 and 3 show the classifications of the leading fields of technological expansion abroad for internationalising firms in the 1930s and in the 1980s, respectively. The key to the sectoral codes is provided in Appendix 1. The sectors indicated are those which are both important to the industrial group in question, and show an increase in foreign share (FS) greater than 4% for the firm in the relevant period. The firms included are those in which there was an increase in foreign share of at least 4% in at least one important sector; on these grounds IG Farben, for example, was excluded from Table 2. Conversely, Tables 4 and 5 show the classifications of the major fields of decline in foreign [TABULAR DATA FOR TABLE 2 OMITTED] [TABULAR DATA FOR TABLE 3 OMITTED] [TABULAR DATA FOR TABLE 4 OMITTED] [TABULAR DATA FOR TABLE 5 OMITTED] technological activity for de-internationalising firms in the 1930s and in the 1980s.
The first three columns of Tables 2 and 3 report the initial level of the overall foreign share of the firm, the change in the foreign share during the period, and the change in the CV. The analysis of the initial level of RTA for the firm in each sector considered, and the change in RTA over the period in question, gives us the three most relevant columns (the fourth, fifth and sixth), which enable us to define sectors for which the internationalisation and diversification of corporate technology are most closely connected to one another. In particular, whenever the firm shows a positive increase in its RTA starting from a value greater than 1 at the beginning of the period, the firm is undertaking an extension abroad of an established field of technological specialisation at home (we term this Extension); instead, if the value of the RTA index rises from less than 1 at the beginning of the period, we have the emergence abroad of a field in which the firm is not specialised at home (Emerging Fields); while if the firm experiences a decrease in the RTA, starting from a value greater than 1 at the beginning of the period, the expansion abroad only partially offsets the decline in an established field of specialisation at home (Substitution). Now Extension and Substitution represent cases in which the internationalisation of corporate technological activity is non-diversifying, in that it builds upon areas of established strength. By contrast, the category of Emerging Fields implies that there is a simultaneous internationalisation and diversification in corporate technological competence, since expansion of these activities abroad increases specialisation in an area of (previously) relative weakness. The remaining three columns list sectors in which (for one reason or another) internationalisation is of only limited significance for corporate technological diversification.
Tables 2 and 3 show that as hypothesised, across these leading firms there has been a shift away from internationalisation of a non-diversifying kind in the 1930s, towards internationalisation of a pro-diversifying kind in the 1980s. Of the largest firms with the most notable internationalisation of technological activity in the 1930s as depicted in Table 2, only a minority diversified - 13 concentrated their profiles of technological specialisation (their CV rose), while just 8 diversified (their CV fell). In contrast, among the leading internationalising companies in the 1980s listed in Table 3, the majority diversified - thus, 23 saw a diversification of their technological base (a fall in CV), and only 13 moved towards a more highly concentrated pattern of technological specialisation (their CV increased).
Alternatively, considering just those sectors which are listed under the three highlighted columns of Tables 2 and 3 described above, and taking firms which have at least three sectors under these headings, we can draw a further distinction between firms in which the internationalisation process itself directly lends support to technological diversification, and those in which it does not. For this purpose we define a pro-diversifying process of internationalisation as one in which, of the total number of sectors listed under the three columns indicated, over a third are in the Emerging Fields category, and a non-diversifying process as one in which a third or less come under this heading. Using this definition, for firms with three or more sectors recorded under the Extension, Substitution and Emerging Fields columns combined, in the 1930s (Table 2) a majority had a non-diversifying process of internationalisation (6 companies, as against 4 which were of the pro-diversifying kind), while in the 1980s (Table 3) a majority had instead a process of internationalisation that was pro-diversifying (17 firms, as against 12 whose strategies were non-diversifying).
However, it may be worthwhile to reflect a little further on the experience of firms that represented an exception to these general rules - that is, companies whose internationalisation seems to have been pro-diversifying already in the 1930s, or whose internationalisation remained of a non-diversifying kind in the 1980s. In the 1930s (Table 2), the four firms for which internationalisation was pro-diversifying (having over a third of at least three relevant sectors in the Emerging Fields column) are ITT, Bendix, and the British or European oil companies Shell and BP. Of these, we know that ITT was a special case, whose interwar expansion (and hence technological diversification) occurred through the rapid acquisition of the international business interests of other firms (including AT&T) outside the US (Wilkins 1974). Meanwhile, Shell and BP established their basic oil exploration and extraction facilities in the US and not in Europe, and so in the first instance located their oil-related research close to these sites. Hence, the foreign research shares of these companies were very high, and the 1930s was the period in which their corporate R&D took off and moved steadily into new areas from almost nothing (Cantwell/Barrera 1993).
Concerning Bendix, while it would seem that the internationalisation and diversification of corporate technology are positively related, it is worth noting that internationalisation was actually very low, with an overall foreign share close to no more than 4%. Thus, although Bendix did experience a substantial diversification of its technological base during the 1930s - its CV fell by about 94%, which is more than any other firm listed in Table 2 - this achievement rested mainly on diversification at home in the US, and was supported only secondarily and in a quite minor way through its internationalisation strategy - with an increase in total foreign share of just 0.3%.
In the 1980s (Table 3), of the 29 firms with three or more sectors recorded under the relevant three headings, just seven had less than a third of the relevant fields in the middle column, that is the Emerging Fields category, which indicates a pro-diversifying form of internationalisation. The seven exceptions are Emhart, GM, Ford, Exxon, Du Pont, Asea and GE (UK). Nonetheless, it is worth noting that among these, three firms (i.e. Emhart, Ford and Asea) still diversified (that is, their CV fell), so that for them at least, the apparently non-diversifying character of internationalisation may create a misleading impression. For all three of these firms the Substitution abroad of established fields of strength at home might actually have facilitated the release of resources at home for technological diversification. So in these cases the relationship between the internationalisation and diversification of technology is unclear. More generally, taking into account that for this reason the third heading of Substitution does not tell us much about the relationship between internationalisation and diversification (the association is ambiguous since the firms may relocate abroad in order to diversify at home, or they may not(7)), we can say that only two firms seem to show internationalisation which is definitely of a non-diversifying kind, and which is based instead on the Extension of domestic strengths - Du Pont and GM. Yet for Du Pont, the increase in CV was very low, and (like with Bendix in the 1930s) the overall foreign share was no more than 2 or 3%, and so this leaves only GM as a company in which there is significant evidence of a direct inverse association between the internationalisation and diversification of technological activities in the 1980s.
In Tables 4 and 5 we consider firms whose overall foreign share fell by more than 1%, in the 1930s and in the 1980s respectively. In these cases we list sectors which represent important fields of development for the firm's industry, but in which the firm itself experienced a significant de-internationalisation in the period in question (a fall in foreign share greater than 4%). Since we are here examining de-internationalisation, the interpretation of the columns is different. In particular, whenever the firm shows a positive increase in its RTA in a sector starting from a value greater than 1 at the beginning of the period, the firm is undertaking a contraction abroad of an established field, which only partially offsets the consolidation of effort in that field at home (we term this Substitution); in contrast, if the value of the RTA rises from an initial position of less than 1, this implies that contraction abroad only partially offsets the expansion of an emergent field at home (Counteracting the Emergence of Fields); while finally, if the firm experiences a decrease in its RTA in a sector in which it starts from a value greater than 1, the contraction abroad facilitates the decline in specialisation of an established field (Contraction). From this alternative perspective of de-internationalisation, it is the latter Contraction strategy which has a pro-diversification effect, while Counteracting the Emergence of Fields at home has an anti-diversification effect.
Once again, with reference now to de-internationalising firms, there seems to be some evidence in Tables 4 and 5 of a shift towards international corporate restructuring in the 1980s, of a kind which is positively associated with technological diversification. As shown in Table 4, in the 1930s only one firm - Bosch - had three sectors in the highlighted columns, and this company had a non-diversifying strategy on the basis suggested above (since the pro-diversifying Contraction of established specialisation did not reach over a third of the total number of relevant sectors). Yet by the 1980s (Table 5) for the two firms for which de-internationalisation affected a reasonable spread of sectors - Thyssen and Volvo - it seems that the rationalisation of international technological activity had definitely assumed a pro-diversifying form (having a preponderance of sectors characterised by the Contraction abroad of established fields of specialisation). However, of the two only Volvo witnessed an overall increase in technological diversification (a fall in CV) during the 1980s. For Thyssen, the pro-diversifying effect of international network restructuring in the 1980s did no more than to ameliorate a tendency towards a more concentrated focus in the firm's technological profile.
The Geographical and Sectoral Profile of International Network Restructuring in the 1980s
Tables 6 and 7 depict the precise geographical composition of the expansion (or contraction) of technological activity abroad in the 1980s, for the most relevant sectors of internationalisation listed under Tables 3 and 5. The figures in brackets in these tables denote the percentage change in the share of corporate patenting accounted for by research in the host country in the sector in question. Tables 6 and 7 show respectively expanding and rationalising firms in the 1980s, as defined in relation to the firm's overall growth rate in patenting, and there are a few differences between these types of firm. In particular, for each of the four rationalising firms recorded under Table 7 (whose total patent stock fell in the 1980s), the international restructuring of technological activity seems to have spread across a number of sectors; while in Table 6 we find at least some firms (such as Bayer and BASF), which despite a growth in the overall level of their technological activity, did not expand abroad across a very wide range of relevant fields. Among the expanding firms (Table 6), we can also observe some cases which are characterised by a strongly geographically polarised international expansion (e.g. Hoescht which undertook the whole of its expansion abroad basically in the US) and which are therefore not particularly interesting in this context. By comparison, others such as SKF show a more complex pattern of restructuring, in that for example in Germany, while it moved into (what were for it) the newly emerging fields of motor vehicle and rubber and plastic product technologies (sectors 43 and 49 - see Table 1 for a full list of definitions), it also extended an established strength in metallurgical processes (sector 13). In contrast, it moved out of metallurgy in Italy, although this simply reinforces the impression that SKF's experience in Italy has been very much that of a pro-diversifying form of internationalisation.
The evidence provided in Tables 6 and 7 may be summarised for each firm with reference to the two relevant issues of (i) the extent of restructuring; and (ii) [TABULAR DATA FOR TABLE 6 OMITTED] [TABULAR DATA FOR TABLE 7 OMITTED] the geographical spread of re-organisation. For this purpose, we define a 'medium' extent of international restructuring as one that spans across seven relevant technological sectors, and a 'medium' geographical spread of international reorganisation as one that involves activity across three countries [ILLUSTRATION FOR FIGURE 1 OMITTED]. Tracing two diagonal lines in Figure 1, it is possible to identify two groups of firms:
1) the first, characterised by greater geographical complexity, is constituted by: Bayer, Ciba Geigy, Shell, BP, SKF, Brown Boveri, Volvo and ITT;
2) the second, characterised by greater technological complexity is constituted by: GE (UK), Hoescht, ICI, Thyssen and Vickers.
The first point to note is that most firms lie either above the upper diagonal line, or below the lower one. That is, although international network restructuring in the 1980s as a general rule involved both internationalisation and a simultaneous positive impetus to corporate technological diversification, managers were still confronted with some degree of organisational trade-off between the management of greater geographical diversity, and the management of greater technological diversity. Only one firm (RTZ) was denoted on our measures as having both a 'high' extent of network restructuring across fields of activity, and a 'high' geographical spread of international reorganisation. However, since the bulk of RTZ's restructuring of activities was concentrated in the USA and Japan (see Table 7), we might argue that this firm belongs more with the second group whose organisational change has emphasised an increasingly complex network of technological fields.
Of the two groups we have described, there are a few more internationalising firms (seven, as against three) which appear to have favoured greater geographical rather than technological dispersion in their strategies for international restructuring. In this respect, the major chemical companies seem to have divergent strategies, with Bayer and Ciba-Geigy emphasising an increasingly complex geographical network, while Hoescht and ICI have organised a more complex technological network in a more geographically concentrated fashion.
It is worth noting that the firms in our sample which originate from small
countries - Sweden and Switzerland - all fall into first group, whose strategies involve a more geographically complex form of international network restructuring. This applies to SKF and Volvo, and to Ciba-Geigy and Brown Boveri. It might be argued that firms from small countries tend to place a higher value on a wider geographical spread of their activities, and that they have become more adept at managing such locational and cultural diversity than have the more monocentric companies which originate from larger countries.
Volvo and Thyssen should perhaps be mentioned separately as well, since these two companies are the ones which in the 1980s were de-internationalising (rather than internationalising) their technological activities. Thus, of the two, Volvo chose effectively to reduce the geographical scope of its operations, while Thyssen decided instead to reduce its technological scope.
Our findings support the notion that by at least the 1980s, MNCs had begun to utilise the internationalisation of technological development as a means of accessing locationally specialised branches of innovation across national boundaries. While historically as a rule, internationalisation had little direct connection to corporate technological diversification and the evolution of the firm into new fields of innovative endeavour, by the 1980s a much more direct and positive association between the two was largely established. In particular, our evidence suggests that the largest and technologically leading firms have witnessed the emergence of corporate international networks for the accumulation of both geographically and sectorally dispersed technological competences.
However, it seems that the management of these networks remains a difficult and complex organisational task. For this reason, firms which have developed internationally integrated networks for technological development have tended to concentrate either on a wider geographical dispersion of competence generation, or alternatively instead on a wider sectoral dispersion of interrelated competences across technological fields but in a more confined set of geographical locations. It remains to explore these suggestions more fully through more detailed future research. Appendix 1. Sectoral groups
1 Food and tobacco products 2 Distillation processes 3 Inorganic chemicals 4 Agricultural chemicals 5 Chemical processes 6 Photographic chemistry 7 Cleaning agents and other compositions 8 Disinfecting and preserving 9 Synthetic resins and fibres 10 Bleaching and dyeing 11 Other organic compounds 12 Pharmaceuticals and biotechnology 13 Metallurgical processes 14 Miscellaneous metal products 15 Food, drink and tobacco equipment 16 Chemical and allied equipment 17 Metal working equipment 18 Paper making apparatus 19 Building material processing equipment 20 Assembly and material handling equipment 21 Agricultural equipment 22 Other construction and excavating equipment 23 Mining equipment 24 Electrical lamp manufacturing 25 Textile and clothing machinery 26 Printing and publishing machinery 27 Woodworking tools and machinery 28 Other specialised machinery 29 Other general industrial equipment 30 Mechanical calculators and typewriters 31 Power plants 32 Nuclear reactors 33 Telecommunications 34 Other electrical communication systems 35 Special radio systems 36 Image and sound equipment 37 Illumination devices 38 Electrical devices and systems 39 Other general electrical equipment 40 Semiconductors 41 Office equipment and data processing systems 42 Internal combustion engines 43 Motor vehicles 44 Aircraft 45 Ships and marine propulsion 46 Railways and railway equipment 47 Other transport equipment 48 Textiles, clothing and leather 49 Rubber and plastic products 50 Non-metallic mineral products 51 Coal and petroleum products 52 Photographic equipment 53 Other instruments and controls 54 Wood products 55 Explosive compositions and charges 56 Other manufacturing and non-industrial
1 The advantages and disadvantages of using US patents as indicator of technological activity are well known and quite widely discussed in the literature (e.g. Schmookler 1950, 1966, Pavitt 1985, 1988, Zander 1994).
2 Affiliate names were normally taken from individual company histories.
3 With the exception of the oil sector, in which we had only two national groups.
4 The choice of the relevant sectors is particularly important to define the range over which the RTA distribution is to be considered. We analysed two alternative measures. The first one is the coefficient of variation of the RTA distribution across all sectors in which firms in the relevant industry had a stock of at least 100 patents at some point during the period 1930-1990. In this case the number of sectors considered is the same for all firms in an industry, and it is fixed over time. The other one is the coefficient of variation of the RTA distribution across a variable number of sectors, according to the constraint that firms in an industry group collectively have an accumulated stock of at least 100 patents in each year considered separately. Although the two measures are strongly correlated, the latter one has been preferred, given that we wish to focus on just four years in particular.
5 This measure has often been used as well in the analysis of business concentration across firms within an industry, as opposed to concentration or dispersion across sectors within a firm (see Hart/Prais, 1956). It is worth noticing that for a given number of firms or in our case sectors (N), there is a strict relationship between the Herfindhal index (H) and the coefficient of variation (CV) (Hart, 1971). The relationship is: H = ([CV.sup.2] + 1)/N.
6 Using the same procedure applied to build the proxy DIV, we considered a second proxy defined as the coefficient of variation across national shares of patenting for the firm. Nevertheless, this is well correlated with the foreign share measure and therefore we preferred the proxy more commonly used in the literature.
7 This would become clearer only by looking at the exact composition of substitution abroad and diversification at home.
Bartlett, C. A./Ghoshal, S., Managing Across Borders: The Transnational Solution, Boston, MA: Harvard Business School Press 1989.
Cantwell, J. A., Technological Innovation and Multinational Corporations, Oxford: Basil Blackwell 1989.
Cantwell, J. A., Corporate Technological Specialisation in International Industries, in Casson, M. C./Creedy, J. (eds.), Industrial Concentration and Economic Inequality, Aldershot: Edward Elgar 1993.
Cantwell, J. A., Introduction, in Cantwell, J. A. (ed.), Transnational Corporations and Innovatory Activities, London: Routledge 1994.
Cantwell, J. A., The Globalisation of Technology: What Remains of the Product Cycle Model?, Cambridge Journal of Economics, 19, 1995, pp. 155-174.
Cantwell, J. A./Barrera, M. P., The Rise of Corporate R&D and the Technological Performance of the Largest European Firms from the Interwar Years Onwards, University of Reading Discussion Papers in Economics, no. 271, September 1993.
Cantwell, J. A./Janne, O., Technological Globalisation and Innovative Centres: The Role of Corporate Technological Leadership and Locational Hierarchy, Research Policy, 28, 1999 (forthcoming).
Cantwell, J. A./Piscitello, L., A Note on the Causality Between Technological Diversification and Internationalisation, University of Reading Discussion Papers in Quantitative Economics and Computing, no. 52, May 1997 a.
Cantwell, J. A./Piscitello, L., Accumulating Technological Competence - Its Changing Impact on Corporate Diversification and Internationalisation, University of Reading Discussion Papers in International Investment and Management, no. 232, June 1997b.
Cantwell, J. A./Sanna-Randaccio, F., Intra-industry Direct Investment in the European Community: Oligopolistic Rivalry and Technological Competition, in Cantwell, J. A. (ed.), Multinational Investment in Modern Europe, Aldershot: Edward Elgar 1992.
Cantwell, J. A./Sanna-Randaccio, F., Multinationality and Firm Growth, Weltwirtschaftliches Archiv, 129, 1993, pp. 275-299.
Chandler, A.D., Scale and Scope: The Dynamics of Industrial Capitalism, Cambridge, MA: Harvard University Press 1990.
Doz, Y., Strategic Management in Multinational Companies, Oxford: Pergamon 1986.
Dunning, J. H., Multinational Enterprises and the Global Economy, Wokingham: Addison-Wesley 1992.
Dunning, J. H., Reappraising the Eclectic Paradigm in an Age of Alliance Capitalism, Journal of International Business Studies, 26, 1995, pp. 461-491.
Dunning, J. H., The Geographical Sources of the Competitiveness of Firms: Some Results of a New Survey, Transnational Corporations, 5, 1996, pp. 1-29.
Granstrand, O., Towards a Theory of the Technology-based Firm, Research Policy, 27, 5, 1998, pp. 465-490.
Kogut, B./Chang, S. J., Technological Capabilities and Japanese Foreign Direct Investment in the United States, Review of Economics and Statistics, 73, 1991, pp. 401-413. Kuemmerle, W., The Drivers of Foreign Direct Investment into Research and Development - An Empirical Investigation, Harvard Business School Working Paper, no. 96-062, 1996.
Hart, P. E., Entropy and Other Measures of Concentration, Journal of the Royal Statistic Society, Series A, 134, 1971, pp. 73-85.
Hart, P. E./Prais, S. J., The Analysis of Business Concentration: A Statistical Approach, Journal of the Royal Statistical Society, Series A, 119, 1956.
Hedlund, G., The Hypermodern MNC: A Heterarchy?, Human Resource Management, 25, 1986, pp. 9-25.
Oskarsson, C., Technology Diversification - The Phenomenon, its Causes and Effects. A Study of Swedish Industry, Department of Industrial Management and Economics, Goteborg: Chalmers University of Technology 1993.
Patel, P./Pavitt, K. L. R., The Technological Competencies of the World's Largest Firms: Complex and Path-dependent, but not Much Variety, Research Policy, vol. 26, 1997, pp. 141-156.
Pavitt, K. L. R., Patent Statistics as Indicators of Innovative Activities: Possibilities and Problems, Scientometrics, 7, 1-2, January 1985.
Pavitt, K. L. R., Uses and Abuses of Patent Statistics, in Van Raan, A. F. J. (ed.) Handbook of Quantitative Studies of Science and Technology, Elsevier Science Publishers 1988.
Pearce, R. D., Industrial Diversification Amongst the World's Leading Multinational Enterprises, in Casson, M. C. (ed.), The Growth of International Business, London: Allen and Unwin 1983.
Porter, M. E., Competition in Global Industries: A Conceptual Framework, in Porter, M. E. (ed.), Competition in Global Industries, Boston: Harvard Business School Press 1986.
Pugel, T. A./Kragas, E. S./Kimura, Y., Further Evidence on Japanese Direct Investment in US Manufacturing, Review of Economics and Statistics, 78, 1996, pp. 208-213.
Schmookler, J., The Interpretation of Patent Statistics, Journal of the Patent Office Society, vol. 32. no. 2, 1950, pp. 123-146.
Schmookler, J., Invention and Economic Growth, Cambridge, MA: Harvard University Press 1966.
Wilkins, M., The Maturing of Multinational Enterprise: American Business Abroad from 1914 to 1970, Cambridge, MA: Harvard University Press 1974.
Zander, I., The Tortoise Evolution of the Multinational Corporation, Stockholm: Institute of International Business, Stockholm School of Economics 1994.
Zander, I., Technological Diversification in the Multinational Corporation: Historical Evolution and Future Prospects, Research Policy, 26, 2, 1997, pp. 209-228.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||The Organisational Evolution of the Multinational Enterprise - A Tribute to Gunnar Hedlund|
|Author:||Cantwell, John; Piscitello, Lucia|
|Publication:||Management International Review|
|Date:||Apr 15, 1999|
|Previous Article:||The organisation and evolution of the multinational enterprise: an information cost approach.|
|Next Article:||Deviation from fit: an advantage when environments change.|