‘PLACING’ FIRMS – ‘FIRMING’ PLACES:

GROUNDING THE DEBATE ON THE ‘GLOBAL’ CORPORATION

Peter Dicken

School of Geography

University of Manchester

Manchester M13 9PL

England

p.dicken@man.ac.uk

Paper presented at the Conference on ‘Responding to Globalization: Societies, Groups, and Individuals’, University of Colorado, Boulder, April 2002

Introduction[1]

In some ways, I am increasingly reluctant to use the term ‘global’ or ‘globalization’. These terms have become so pervasive that they have taken on a life of their own. They occupy their own etymological space and have, in much of their usage, become almost meaningless. The same words are used by different people to mean different things. Like all such popular terms, however, they are difficult to shake off. I guess we are stuck with them. But, if that is the case, we need to be careful in how we use them. In this paper, I focus only on the economic dimension of globalization, and on just one aspect of the economic: the business organization or firm. My aim is to explore the relationships between firms and geographical space or territory in the context of the ongoing debate over the organization of economic activity in the world economy. Specifically, I want to refute the assertion that we are seeing the inexorable displacement of a variety of organizational forms by convergence on a single dominant form - the so-called global corporation.

I will focus on two questions. First, I will review some of the evidence for the existence of ‘global’ corporations. Second, and more substantially, I will explore some of the complex, dialectical relationships between firms and places. This involves a number of issues. How is the nature of firms related to the specific places within which they are embedded? That is, do different places ‘produce’ different types of firms? How are place (and space) ‘used’ by firms as part of their competitive strategies? Is ‘placeness’ more than simply the fundamental need for all economic activities for spatial fixity (see Harvey 1995)? What are the implications for places (at different geographical scales) of geographically discriminatory behaviour?

Global corporations rule, OK?

One of the central claims of the ‘hyperglobalists’ is that international firms are inexorably, and inevitably, abandoning their ties to their country of origin and, by implication, converging towards a universal global organizational form. Kenichi Ohmae’s infamous exhortation to business managers is usually invoked as the exemplar of such a position:

Country of origin does not matter. Location of headquarters does not matter. The products for which you are responsible and the company you serve have become denationalized (Ohmae, 1990: 94)

Ohmae’s position reflects a pervasive view among many writers. It is not a new idea. The US Under-Secretary of State in the 1960s, George Ball, coined the label ‘Cosmocorp’ to denote what he saw as the then emerging global corporation (Ball 1967). Barnet and Muller’s 1974 book Global Reach: The Power of the Multinational Corporations was replete with anecdotal examples of the intentions of US corporate executives to transform their firms into ‘placeless’ global corporations. They quote the musings of a Chairman of the Dow Chemical Company:

I have long dreamed of buying an island owned by no nation and of establishing the World Headquarters of the Dow company on the truly neutral ground of such an island, beholden to no nation or society. If we were located on such truly neutral ground we could then really operate in the United States as US citizens, in Japan as Japanese citizens and in Brazil as Brazilians rather than being governed in prime by the laws of the United States…We could even pay any natives handsomely to move elsewhere (cited in Barnet and Muller, 1974:16).

More recent musings express rather vague ideas as to what is a ‘global’ corporation. For example, addressing the APEC CEO’s Summit Meeting in 1999, the then CEO of General Motors stated:

We need a general agreement on what constitutes a global corporation and what we mean when we say that globalization has increased. For purposes of today’s discussion, I suggest we forget the academic debate and agree that a corporation is global if it holds equity or debt in affiliated corporations located in a nation other than its home country. And, when we say that globalization has increased, we mean that the number of corporations that hold equity in firms outside their home nation has increased. Using those basic definitions, global corporations and globalization are not a new phenomenon

Not very helpful!

As for the future, we seem to have little more than the - often bizarre - prognostications of the business and management gurus, most of whom are quite clearly in the snake-oil business. Airport bookstalls are full of such ‘literature’. One example drawn from the pages of Business Week gives a flavour of this kind of prediction:

The global corporation becomes the leaderless corporation…Success will belong to companies that are leaderless – or, to be more precise, companies whose leadership is so widely shared that they resemble beehives, ant colonies, or schools of fish…You have only to look at biological systems to see that there are no big hierarchical stacks…Everything is low and flat, very adaptable and very cruel (Byrne 1999)

The argument underlying all these positions is, essentially, that technological and regulatory developments in the world economy have created a ‘global surface’ on which a dominant organizational form will develop and inexorably wipe out less efficient competitors who are no longer protected by national or local barriers. Such an organization is, it is argued ‘placeless’ and ‘boundary-less’. The claim that the placeless corporation is, or is becoming, the norm amongst international business firms received a substantial boost in the 1990s with the persistence of the Japanese financial crisis, following the collapse of the bubble economy at the beginning of the decade, and the equally unexpected East Asian financial crisis of 1997-1998. In effect, the model of the global corporation is now, to all intents and purposes, that of the US corporation. US market capitalism has seemingly triumphed; the neo-liberal IMF/Washington/Wall Street consensus sets the rules. It now seems a very long time ago that the US business and political communities were obsessed with the apparent failure of the US corporate model in the face of the Japanese variety of capitalism (and, indeed, of the newer forms emanating from Korea, Taiwan, and other parts of East Asia). Shareholder value is now the hot ticket. For some, indeed, it is the only show in town. The US-style corporation is seen as being the most effective way of maximizing such value. All other models of business organization are not only less efficient but will, inevitably, be vanquished in a neo-Darwinian struggle.

The notion that the world is now dominated by massive global corporations is not solely confined to the world of business. Many writers on the political left tend to take a similar view (although, of course, with a very different interpretation of the economic, social, and political implications). One of the oldest devices is to compare the size of TNCs with nation-states in order to demonstrate that TNCs have become more powerful than states. For example, a recent survey published by the Institute for Policy Studies (Anderson and Cavanagh (2000) makes the following claims, amongst others:

• ‘Of the 100 largest economies in the world. 51 are corporations; only 49 are countries (based on a comparison of corporate sales and country GDPs)…To put this in perspective, General Motors is now bigger than Denmark; DaimlerChrysler is bigger than Poland; Royal Dutch/Shell is bigger than Pakistan’ (3)
• ‘The 1999 sales of each of the top five corporations (General Motors, Wal-Mart, Exxon Mobil, Ford Motor, and DaimlerChrysler) are bigger than the GDP’s of 182 countries’ (3)

These are, of course, very striking comparisons. But it is far from clear what they really mean. In particular, as Wolf (2002) has pointed out the comparison is based on totally different criteria:

GDP is a measure of value-added, not sales. If one were to compute total sales in a country one would end up with a number far bigger than GDP. One would also be double-, triple-, or quadruple-counting…if corporations, too, are measured by value added as national economies are…they tend to shrink by between 70 and 80 per cent. In 2000, sales by General Motors were $185bn but value added was $42bn; sales by Ford were $170bn but value added was $47bn; and sales by Royal Dutch/Shell were $149bn but value added was only $36bn…Properly measured, Denmark’s economy is more than three times bigger than GM. Even impoverished Bangladesh has a bigger economy than GM.

So, in evaluating the scale of global corporations, such comparisons are not very helpful. They do, of course, perform the extremely valuable function of forcing us to think about the shifting relative power between private and public institutions, with all that this implies politically and economically and that, after all, is the purpose of the Institute for Policy Studies. Such comparisons, therefore, have real value as polemic but they do not tell us much about the ‘global-ness’ of corporations, or even the extent to which corporations are more, or less, oriented to domestic or foreign operations.

This begs the question of how we might define a ‘global’ corporation – a far from simple task. In my view, a global corporation would be a firm that not only has significant operations in a large number of countries (not necessarily all, of course) but also a firm whose geographically-dispersed operations are functionally-integrated, and not merely a diverse portfolio of activities.

Hu (1992) suggests a rather broader set of criteria to assess the extent to which TNCs are global:

• in which nation or nations is the bulk of the corporation’s assets and people located?
• by whom are the local subsidiaries owned and controlled, and in which nation is the parent company owned and controlled?
• what is the nationality of the senior positions (executive and board posts) at the parent company, and what is the nationality of the most important decision-makers at the subsidiaries in host nations?
• what is the legal nationality of the parent company?
• To whom would the group as a whole turn for diplomatic protection and political support in case of need?
• which is the nation where the tax authorities can, if they choose to do so, tax the group on its worldwide earnings rather than merely its local earnings?

On the basis of his empirical analysis of a sample of TNCs, Hu concludes that

these criteria usually produce an unambiguous answer: that it...[the TNC]...is a national corporation with international operations (i.e. foreign subsidiaries)

However, it does seem to be the case that most of the world’s largest business corporations think of themselves as either being ‘global’ or, at least, ‘globalizing’. In his analysis of the ‘transnational capitalist class’, Sklair (2001) interviewed senior figures in around 80 of the 500 largest corporations in the world in order to try to establish their use of, and the meaning they attached to, the term ‘globalization.’

The suggestion that multinationals were ‘national companies with units abroad’ was roundly rejected as old fashioned and not compatible with the demands of the contemporary global economy. Most Global 500 executives and managers in the sample considered their corporations to be in a transitional state between the multinational corporation and the global corporation, that is, they were to a greater or lesser extent globalizing. Also clear was the finding that, in order to fulfil a ‘shareholder-driven growth imperative’, most of the corporations considered that they had no alternative but to globalize…All of these findings demonstrate a move to globalization among the Global 500 and a certain level of success. Nevertheless, few of these corporations considered themselves entirely globalized (73).

A rather different approach is to analyse quantitatively the relative distribution of individual firms’ assets, sales, and employment between their home country and abroad. This is the approach adopted by Ruigrok and van Tulder (1995) and by recent volumes of the UNCTAD World Investment Report. Ruigrok and van Tulder (1995) analysed Fortune 100 data for the early 1990s and concluded that

of the largest one hundred core firms in the world, not one is truly ‘global’, ‘footloose’ or ‘borderless’. There is, however, a hierarchy in the internationalisation of functional areas of management: around forty firms generate at least half of their sales abroad; less than twenty maintain at least half of their production facilities abroad; with very few exceptions, executive boards and management styles remain solidly national in their outlook; with even fewer exceptions, R & D remains firmly under domestic control; and most companies appear to think of a globalization of corporate finances as too uncertain (159).

UNCTAD has taken this kind of approach a step further and developed a composite Trnansationality Index (TNI). This is simply a weighted average of three indicators: foreign sales as a percentage of total sales; foreign assets as a percentage of total assets; and foreign employment as a percentage of total employment. These data have now been published for seven successive years, from 1993 to 1999, allowing some comparison of trends over time. A number of important features can be identified:

1. The mean TNI for all 100 TNCs in 1999 was 52.6. In 1993, the mean was 51.6. In other words, for the top 100 TNCs as a group (although, of course, not a completely identical group because of entries and exits over the intervening period) there was no significant increase in their degree of transnationality. Hence, we cannot say that this elite group – what we could reasonably regard as the world’s global corporations – have, by this simple measure at least, become more global.
2. The fact that the TNI index for this group of firms is only a little over 50 suggests that, on average, these firms have roughly half their operations in their country of origin and half abroad. This does not suggest an especially high degree of globalization.
3. Only 57 of the 100 companies in 1999 had a TNI of greater than 50 and only 16 companies had a TNI greater than 75 (i.e. indicating more than three-quarters of their activities outside their country of origin).
4. There is no correlation between the size of a firm’s foreign assets and its TNI as Figure 1 indicates.

Figure 1: The relationship between firms ranked by TNI and by foreign assets, 1999

Source: calculated from UNCTAD World Investment Report, 2001

5. Only two companies – Nestlé and the Nippon Mitsubishi Oil Company – appear in the top 15 on the basis of both foreign assets and their TNI. In general, the largest TNCs in terms of total foreign assets all have relatively low TNIs as Table 1 shows. For example, largest TNC in terms of assets, General Electric ranks 75^th on the TNI list. GM, 4^th in terms of foreign assets is 83^rd on the TNI ranking. For Ford the numbers are 5 and 77: for Toyota, 6 and 82, for IBM 9 and 50.

Table 1 The top 15 TNCs ranked by foreign assets, and transnationality index, 1999

Foreign assets

Rank	Company	Country	TNI rank
1	General Electric	United States	75
2	ExxonMobil	United States	22
3	Royal Dutch/Shell	United Kingdom	43
4	General Motors	United States	83
5	Ford	United States	77
6	Toyota	Japan	82
7	DaimlerChrysler	Germany	51
8	TotalFina	France	21
9	IBM	United States	50
10	BP	United Kingdom	18
11	Nestlé	Switzerland	2
12	Volkswagen	Germany	45
13	Nippon Oil Co	Japan	11
14	Siemens	Germany	41
15	Wal-Mart	United States	90

Transnationality Index

Rank	Company	Country	Foreign assets rank
1	Thomson Corporation	Canada	57
2	Nestlé	Switzerland	11
3	ABB	Switzerland	21
4	Electrolux	Sweden	80
5	Holcim	Switzerland	59
6	Roche Group	Switzerland	27
7	BAT	United Kingdom	35
8	Unilever	UK/Netherlands	24
9	Seagram	Canada	23
10	Akzo Nobel	Netherlands	75
11	Nippon Oil Co	Japan	13
12	Cadbury-Schweppes	United Kingdom	96
13	Diageo	United Kingdom	17
14	News Corporation	Australia	31
15	L’Air Liquide Grp	France	82

(Source: based on UNCTAD 2001: TableIII.1)

6. There are some substantial differences in the degree of transnationality between firms from different home countries. Not surprisingly, in general, firms from smaller countries (either geographically or in terms of their economic size) tend to have higher TNIs than firms from larger countries (this is very clear from Table 1). Figure 2 shows the pattern of TNIs for the leading home countries for both 1993 and 1999. A number of features are evident:

(a) Although the mean TNI for United States firms increased from 36.7 in 1993 to 41.5 in 1999, the overall degree of transnationality of US firms is lower than all the other major home countries other than Japan. In 1999, only one-third of the 25 United States firms in the list had a TNI greater than 50, with the highest value being 66.

(b) Similarly, Japanese firms have a low degree of transnationality, on average. The mean value increased from 33 in 1993 to 38.4 in 1999. Again, around one-third have a TNI less than 50

(c) Amongst European firms, those from the UK have, on average, the highest TNIs. In 1993, the UK mean TNI was 68.6; by 1999 it had risen to 75.5.

Figure 2: TNIs by country of origin, 1993 and 1999

Source: calculated from UNCTAD World Investment Report, 2001

Thus, the results of analysing the UNCTAD data are clear in one respect: the majority of the world’s leading TNCs still retain more than 50 per cent of their activities in their home country. In that sense they are, in Hu’s terms, ‘national firms with international operations’, notwithstanding the views expressed by corporate executives in Sklair’s survey. The major exceptions are firms from small home countries. But even if the figures did show a greater degree of ‘transnationality’ they would not tell us very much because the measure used is simply a dichotomous one: home versus abroad. It can tell us nothing about the real geographical extent of a firm’s activities. At least in principle, a firm could have a TNI of, say, 80 (meaning that 80 per cent of its activities were outside the firm’s home country) but all of those activities could be in one foreign country. An example would be the significant number of US firms that operate in Canada but not elsewhere. So, in fact, all the TNI and similar measures can do is to measure the extent to which a firm’s activities are domestic or foreign-based.

However, the UNCTAD data do throw light on some specific issues, particularly some of the conclusions of Anderson and Cavanagh’s comparison of TNCs and nation-states referred to earlier. In particular, the Anderson and Cavanagh list based on sales contains a much larger proportion of United States firms (41 per cent) than does the UNCTAD list based upon transnationality (25 per cent). In other words, United States firms are substantially less global than Anderson and Cavanagh’s study implies. Hence, the UNCTAD list is a much better measure of the global-ness of firms. One specific example can be used to illustrate this discrepancy. According to Anderson and Cavanagh, Wal-Mart is the biggest private employer in the world with 1,140,000 workers. The implication is that this has a major impact globally. However, if we look more closely at Wal-Mart, we find that the extent of its transnational operations is very limited indeed. Although Wal-Mart is the largest retailer in the world in terms of sales, its sales outside the United States represent only 14 per cent of its total sales. Likewise, Wal-Mart may well employ over one million workers, but around 85 per cent of these are located in the UnitedStates. Wal-Mart’s operations outside the Americas are very limited indeed. In Europe they are confined to the United Kingdom and Germany; in Asia to a small number of stores in China and Taiwan. It is hard, then, from a non-US perspective, to regard Wal-Mart as a ‘global’ corporation.

So, all the TNI (and similar measures) do is to measure the extent to which a firm’s activities are domestic or foreign-based. They tell us nothing about their geographical spread. This can only really be achieved from company-specific investigation. Examination of firms’ annual reports or websites, although revealing great inconsistency in their degree of geographical specificity, tend to show that only a relatively small number of firms have globally-spread operations. Rather more tend to have strong regional biases in their operations.

‘Placing’ firms; ‘firming’ places

The quantitative data discussed in the previous section are useful for my present purposes because they demonstrate the continuing real ties between firms and their home countries. But they do not tell us anything about the nature of those ties: about the qualitative nature of TNC activities and their relationship to place. Neither do they help us to establish whether or not TNCs of different national origins are becoming similar in their modes of operation. It is at least possible that TNCs may retain more of their assets and employment in their home country but still be converging organizationally and behaviourally towards a universal, global form. Equally, some firms may be becoming more global but in non-uniform ways. If so, what role does the firm’s country of origin play in this?

The response of geographers to assertions of the rise of the placeless global corporation has been to re-assert - somewhat reflexively - the importance of place. But we haven’t done very much more either to think through what this involves conceptually or to provide robust empirical evidence. I don’t claim in this paper to do either of these other than in a superficial manner, but what I hope to do is to point to some of the conceptual and empirical issues involved in what is, I believe, an important issue in both academic and policy terms.

The basis of my argument is that firms, just like all other forms of social organization, are fundamentally and intrinsically spatial and territorial. They are spatial in the sense that they are responsive to geographical distance and to spatial variations in the availability of necessary resources and of business opportunities. Such spatiality may have – indeed most often has – a territorial manifestation. Hence, firms are territorial as well as spatial in the sense that the ‘surface’ from which firms originate and on which they operate is most commonly made up of a tessellated structure of territorial entities arrayed along a continuum of variable and overlapping scales, including those of political governance. Firms themselves have territorial extent: the roughly bounded area over which they conduct their operations (e.g. their market area; their labour catchment area, their supply area, etc). For some functions of the firm the territory may be intensely local, for others it may approach the global. Such firm territories, however, are usually temporally volatile, spatially discontinuous, and not clearly bounded. Indeed, firms in competition inter-penetrate each other’s territories in highly complex and contested ways. More particularly from the perspective of this paper, firms are territorial in that they derive some of their characteristics from, and also directly influence, the characteristics of specific territories and places.

The theoretical basis for hypothesising that TNCs ‘produced’ in different places will continue to display a significant degree of organizational differentiation lies in the much used (and sometimes abused) concept of social embeddedness. The most widely-quoted proximate authority on this topic is undoubtedly the economic sociologist Mark Granovetter (see Granovetter 1985, Granovetter and Swedberg 1992, Smelser and Swedberg 1994) but its intellectual origins lie with Karl Polanyi’s (1957) seminal work, The Great Transformation. The intricacies of the embeddedness concept lie beyond the bounds of this paper. Suffice it to argue that all business firms are rooted within specific social, cultural, political and institutional contexts which help to influence the ways in which they develop.

At least in origin, TNCs are ‘locally grown’; they develop their roots in the soil in which they were planted. The deeper the roots the stronger will be the degree of local embeddedness, such that they should be expected to bear at least some traces of the economic, social and cultural characteristics of their home country. In other words, they continue to contain elements of the local within their modes of operation…the local social-cultural milieu is a major influence on how firms evolve and behave even when their operations are geographically very extensive. This is not to argue a case for cultural determinism or even to argue that all firms of a given nationality are identical. Clearly they are not. But they do tend to share some common features (Dicken, Forsgren, Malmberg 1994:34).

Although such embeddedness may occur at a variety of geographical scales the most significant scale would appear to be that of the national state, the major ‘container’ within which distinctive practices develop. . The term ‘container’ should not be taken too literally. It is used here as a fairly loose metaphor to capture the idea that nation-states are one of the major ways in which distinctive institutions and practices are ‘bundled together’. Of course, such containers are not (except in very rare cases) hermetically sealed off from the outside world. The container is permeable or leaky to varying degrees. It can be argued that the impact of the modern information technology and communications systems has made national containers even more permeable. That’s true – but it does not mean that the container no longer exists at all. Indeed, there is a good deal of compelling evidence to show the persistence of national distinctiveness - although not necessarily uniqueness - in structures and practices which help to shape local, national and international patterns of economic activity.

From an economic