Journal of World-Systems Research, IX:1, Winter 2003 67 Forging a New Hegemony? The Role of Transnational Policy Groups in the Network and Discourses of Global Corporate Governance* William K. Carroll Colin Carson William K. Carroll Colin Carson Department of Sociology University of Victoria Box 3050 Victoria, BC V8W 3P5 CANADA wcarroll@uvic.ca ccarson@uvic.ca http://web.uvic.ca/soci/ journal of world-systems research, ix, 1, winter 2003, 67–102 http://jwsr.ucr.edu issn 1076–156x © 2003 William K. Carroll & Colin Carson I N T RODUC TION In recent decades the development of a transnational phase of capitalism, said to include the global integration of national economies, the mobility of capital and global reach of accumulation circuits, and the growing role of organizations like the World Economic Forum (WEF) and the World Trade Organization (WTO), has claimed the attention of legions of social scientists. With this interest has come a concern to theorize the segment of the world bourgeoisie purported to represent transnational capital and the ideology, neoliberalism, which seems to underwrite its expansion. Th ese issues have gained additional salience as scholars such as Robinson and Harris (2000) and Sklair (2001) have discerned the formation of a fully transnational capitalist class (TCC). A range of theoretical perspectives on transnational capitalist class forma- tion now exists. In the early 1970s, dramatic increases in direct foreign invest- ment through multinational corporations led Hymer (1979) to observe that “an international capitalist class is emerging whose interests lie in the world economy as a whole system of international private property which allows free movement of capital between countries” (262). In the 1980s, the Gramscian turn in IPE, advocating a “historically grounded conception of the dialectic totality of structure and agency” in processes of class formation and world order (Overbeek Th is study situates fi ve top transnational policy-planning groups within the larger struc- ture of corporate power that is constituted through interlocking directorates among the world’s largest companies. Each group makes a distinct contribution toward transnational cap- italist hegemony both by building consensus within the global corporate elite and by edu- cating publics and states on the virtues of one or another variant of the neoliberal paradigm. Analysis of corporate-policy interlocks reveals that a few dozen cosmopolitans—primarily men based in Europe and North America and actively engaged in corporate management— knit the network together via participation in transnational interlocking and/or multiple policy groups. As a structure underwriting transnational business activism, the network is highly centralized, yet from its core it extends unevenly to corporations and individuals posi- tioned on its fringes. Th e policy groups pull the directorates of the world’s major corporations together, and collaterally integrate the lifeworld of the global corporate elite, but they do so selectively, reproducing regional diff erences in participation. Th ese fi ndings support the claim that a well-integrated global corporate elite has formed, and that global policy groups have contributed to its formation. Whether this elite confi rms the arrival of a transnational capitalist class is a matter partly of semantics and partly of substance. abstract * An earlier version of this article fi rst appeared in Global Networks 3:1, 29–57 ( January 2003). mailto:wcarroll@uvic.ca mailto:ccarson@uvic.ca http://web.uvic.ca/soci/ http://jwsr.ucr.edu/ William K. Carroll & Colin Carson68 Forging a New Hegemony? 69 2000), demonstrated that while the mechanisms of international trade and investment furnished structural conditions for global capitalist expansion, they could not provide the long-term vision needed for capitalist class formation. Van der Pijl (1998) and Overbeek and van der Pijl (1993) situate transnational class formation in the context of restructuring and stabilizing capitalist fractions (bank, commercial, industrial capital) under the global economic hegemony of neoliberalism. Of specifi c interest is the development of strategic vision in the social networks of the directors of corporations, banks, and planning groups of various sorts (van der Pijl 1998, 5). Cox (1987), Gill (1990, 1992), and Robinson and Harris (2000), describing similar practices in relation to transnational state apparatuses, view the TCC as both an embodiment of transnational capital and an expression of political power manifest by transnational (or interstate) institu- tions such as the International Monetary Fund (IMF), and the WTO. “World hegemony,” as such, “is describable as a social structure, an economic structure, and a political structure; and it cannot be simply one of these things but must be all three” (Cox 1983, in Overbeek 2000, 176). In a somewhat separate vein, Sklair (2001) places signifi cant emphasis on the ideological awareness of transnational executives and views the dissemination of a culture-ideology of consumerism as integral to transnational capitalist class formation. Robinson and Harris (2000) draw on many of these perspectives to announce the emergence of a fully transnational capitalist class whose “organic composition, objective position and subjective constitution…[is] no longer tied to the nation state” (14). As might be expected, the claim of such an epochal shift has forced a closer assessment of how the TCC is identifi ed. Indeed, the col- lection of critiques that followed the article’s publication brings to light several unresolved issues and questions, including the extent of the TCC’s geopolitical scale—particular emphasis is placed on the recalcitrance of a North/South divide—and its alleged autonomy from national contexts.1 From all sides of the current debate it is agreed that more direct evidence is needed. In fact, precious little systematic empirical data have been marshalled to date. However, a recent longitudinal study of the social structure of the international business community by Carroll and Fennema (2002) does speak to several of the key issues raised in the aftermath of Robinson and Harris’s (2000) intervention. While network analysis has long contributed to an empirical understanding of elite integration (and, by extension, class formation) at the national level (Useem 1984; Domhoff 1998), it was only with Fennema’s (1982) study of international networks of banks and industry that this analysis took transnational scale. Carroll and Fennema’s research builds from Fennema’s earlier work to examine changes in the network of interlocking directorates between 1976 and 1996, a period associated with the most recent surge in economic globalization. Among their key fi ndings were, on the one hand, moderate increases in transnational integration via weak ties that transect national borders, but on the other, recal- citrant national patterns of organization—thus their characterization of the transnational network as “a kind of superstructure that rests on rather resilient national bases” (2002:414). Carroll and Fennema conclude that while corporate interlocks within countries are often associated with the strategic control of capital, “transnational corporate interlocking is less about intercorporate control than it is about the construction of an international business community” (2002: 415). Such a community would be a rather pallid aff air if it were confi ned to the corporate boardrooms. In fact, given the persistence of national corporate networks, we might say that the articulation of a transnational capitalist interest requires sites beyond the boardrooms—places where business leaders can come together to discuss issues of shared concern, to fi nd common ground and to devise strategies for action. Business activism of this sort would seem an integral aspect of community development at the higher reaches of corporate power. Th e signifi cance of such arrangements is only enhanced by processes of globalization and the search for new forms of governance. In recent years these conditions have indeed prompted a range of scholarly attention on institutions of private author- ity and their self-regulatory potential (Ronit 2001:562; Keck and Sikkink 1998). Building on the concept of an international business community, and assert- ing the basic premise that those who direct the largest corporations are the leading edge of a capitalist class, this article situates fi ve global organizations of elite consensus-building within a larger structure of corporate power that is constituted through interlocking directorates. Th e elite policy-planning groups operate within an incipient “global civil society” (Shaw 2000) that is distinct from both state power and economic power yet intimately linked to both. It is from these sites that the strategic and moral visions and policy frameworks informing a transnational capitalist interest have been forged. By mapping the corporate-policy network we hope to shed light on the role that global policy groups are playing in the formation of a transnational capitalist class. POLIC Y GROU PS A S SI T E S FOR CONST RUC TI NG T R A NSNAT IONA L N EOLI BE R A L H E GE MON Y In the years since World War Two we can trace the development of a neo- liberal tendency within a diff erentiating global fi eld of elite consensus formation. 1. Th e Transnational Ruling Class Formation Th esis: A Symposium. Science & Society, 65, 4, Winter 2001-2002, 464-469. William K. Carroll & Colin Carson70 Forging a New Hegemony? 71 emergence of new forms of civil resistance crystallized around opposition to the legal incursions of capitalist globalization, including the MAI, the WTO, and World Bank and IMF initiatives. In turn has come “increasing concern with how best to co-ordinate actions to promote and consolidate it on diff erent scales, with its social and environmental costs and their adverse political repercussions, and with identifying and pursuing fl anking measures that would help to re-embed the recently liberated market forces into a well-functioning market society” ( Jessop 2000). Indeed, by the mid-1990s neoliberal order was increasingly fragmented around the question of how best to assure long-term stability and reproduction of transnational capital. For Robinson and Harris it is precisely this new regulatory positioning within the neoliberal paradigm, and the tensions this creates among global- izing elite, that have given rise to a transnational capitalist class defi ned both by economic structure and strategic-political rule—a class both in-itself, and for-itself (21).3 Th eir analysis very usefully divides the globalist policy fi eld into three neoliberal fractions, which we will employ to help frame our discussion of the projects of transnational policy groups. Th e fi rst fraction is free-market conservative. Infl uenced by economist Milton Friedman, this fraction calls for a complete global laissez-faire, drawing on fundamental neoliberal tenets of mon- etarism, state deregulation, “spontaneous order” of market relations, and posses- sive individualism. Reigning as neoliberalism’s singular voice under the so-called Washington consensus, the project would be splintered and somewhat mar- ginalized amidst the global economic crises of the 1990s. Stemming from these actualities, the fraction that according to Robinson and Harris (2000) is now dominant, neoliberal structuralism, advocates a “global superstructure that could provide a modicum of stability to the volatile world fi nancial system…without interfering with the global economy” (43). Following progenitors Bill Clinton and Tony Blair, its politics are distinctly ‘Th ird Way’—“fi nding a synergy between private and public sectors” as Giddens put it (1998:99–100). Gill (1995), notably, has discerned a very similar policy shift in the ‘new constitutionalist’ discourse, launched during the g7 Summit in Halifax, Nova Scotia, in June 1995 (413). Responding to the Mexican crisis of 1994-95, g7 members opted to “strengthen [economic] surveillance mechanisms under the aegis of the IMF, World Bank, and the BIS” (413). Contrasting with the position of free-market conservatives the new perspective held that “ideology and market power are not enough to Set in motion with Friedrich Hayek’s convening of the Mont Pelerin Society in 1947, its austere market-monetarist orientation gained a distinct, yet still mar- ginal, voice in an organizational ecology dominated by corporate liberal tenden- cies—a regulatory strategy upheld at the time by the fi rst truly North Atlantic planning body, the Bilderberg Conferences (1952). Rising to dominance decades later under the regimes of Reagan and Th atcher, undiluted neoliberal doctrine responded to structural shifts that beleaguered the post-war Keynesian-Fordist state and accelerated the spread of transnational corporations, the expansion of foreign direct investment and the interpenetration of capital. Lending sanction to the distinctly global regime of accumulation that was taking shape were the policy imperatives of privatization, trade liberalization, deregulation, tax reform, and the introduction of market proxies and benchmarking into the public sector—a grouping of corrosive neoliberal initiatives that John Williamson (1990), World Bank Chief Economist for South Asia (1996–1999), termed the “Washington consensus.” Integral to the political and cultural reproduction of this new order has been a synthesis of public and private elements from the states and civil societies of the capitalist world in several new private international policy groups, most notably the World Economic Forum (1971), the Trilateral Commission (1973), and the World Business Council for Sustainable Development (1995). While such groups make distinct strategic contributions to the fi eld of transnational neoliberal policy, they share three critical attributes. Each inhabits a space within civil society as “embedded elements of a social network, within which neoliberal business activism [takes] shape and form” (Carroll and Shaw 2001, 196). Th ey also act as vehicles of international elite integration, linking capitalists to a political-cultural community where class extremes are mediated and a “collective will” thrashed out (van der Pijl 1998). Finally, all, to varying degrees, endeavor to “translate class interests into state action by defi ning and promoting lines of policy that ensure the stability and reproduction of a system shaped by capitalist social relations” (Peschek 1987:216). In these ways, neoliberal policy groups can be said to function as “collective intellectuals”—‘deputies’ or agents of the capital- ist class “entrusted with the activity of organizing the general system of relation- ships external to…business itself,” as Gramsci described (1971:6). Still, the struggle to spread the neoliberal economic project on a global scale has been far from straightforward, and has experienced several major setbacks over the course of the last decade, including global recession and crises, 2 and the ². Th ese include Mexico in 1995, Asia in 1997, and Russia and Brazil in 1998. 3. Sklair (2001), quite similarly, sees ‘proactive global corporate citizenship’ as a cornerstone of contemporary processes of transnational capitalist class formation. William K. Carroll & Colin Carson72 Forging a New Hegemony? 73 ensure the adequacy of neoliberal restructuring…[and must be] institutional- ized at the macro-level of power in the quasi-legal restructuring of the state and international political forms” (Gill 421). Th e third fraction is neoliberal regulation- ist. Th is current, exemplifi ed by Joseph Stiglitz (1998), calls for a “broader global regulatory apparatus that could stabilize the fi nancial system as well as attenuate some of the sharpest social contradictions of global capitalism” (Robinson and Harris 2000: 43). Although each globalist fraction is divided on the amount of structural interference that should occur in the new ‘global economy’, all three are neoliberal in that “none question the essential premises of world market liberal- ization and the freedom of transnational capital” (ibid). T H E FI V E I N T E R NAT IONA L POLIC Y GROU PS In this article we focus on fi ve organizations that have come to comprise a fi eld of transnational policy formation, two with longstanding histories, and three whose origins lie within the contemporary wave of economic globalization. Th at fi eld has taken an historically stratifi ed and pluralistic shape as the groups have developed around specifi c visions, issues and networks. Th e Paris-based International Chamber of Commerce (ICC), founded in 1919, is the oldest of the business policy groups discussed here and the only one to maintain a primarily free-market conservative strategic vision. It is also the largest, grouping some 7,000 member companies and associations from over 130 countries. As a forum for transnational capitalist consultation launched by investment bankers in the shadow of World War I, the ICC has historically functioned as the most comprehensive business forum committed to liberaliza- tion, and now even refers to itself as the ‘World Business Association’. Although it has made forays into humanitarian and environmental policy issues,4 the ICC remains a bastion of global neoliberal doctrine, and has “long been a triumphant lobbyist for global economic deregulation in fora such as the WTO, the G8 and the OECD” (Balanyá et al. 2000:166). Th e ICC’s primary function is to institutionalize an international business perspective by providing a forum where capitalists and related professionals (e.g., law fi rms and consultancies, national professional and sectoral associations) can assemble to forge a common international policy framework in arenas ranging from investment to specifi c technical and sectoral subjects. Since the mid-1990s its eff orts to institutionalize an agenda of corporate self-regulation have fostered close working relationships with international institutions such as the WTO, UN Conference on Trade and Development (UNCTD), and the UN General Secretariat (ibid, 166–174). Th e ICC’s secondary function is to knit national chambers throughout the world into a single global network through its World Chambers Federation (WCF). Th e WCF also provides a vertical organizational link between the network of transnational capitalist interests carried by the ICC membership and the untold numbers of small- and medium-sized businesses which comprise the ranks of local national chambers of commerce. It is, however, a combination of the group’s free-market conservative vision, its institutionaliza- tion of transnational business practices, and its incorporation of local-level busi- ness into a global capitalist perspective, that gives the ICC a unique niche within the organizational ecology of global policy groups. Off ering a counterpoint to the austere, free-market conservative vision of the ICC, the Bilderberg Conferences have provided a context for more com- prehensive international capitalist coordination and planning. Founded in 1952, the Bilderberg, named for the Hotel de Bilderberg of Oosterbeek, Holland, “assembled, in the spirit of corporate liberalism, representatives of Right and Left, capital and organized labor” (van der Pijl 1998:121). Activities have typically revolved around issues of long-term planning and international order, and to this end Bilderberg Conferences have furnished a confi dential platform for corpo- rate, political, intellectual, military, and even trade union elites from the North- Atlantic heartland to reach mutual understanding. With its unique capacity to draw on these elite interests, the Bilderberg has helped to ensure that consensual policies were adopted by the transnational system of the West in general, and signatories to the NATO Alliance in particular (Mendes 1994). Th e group is run by a chairman, and a small, permanent steering committee, which invites approximately 115 participants to the yearly Conference. Compared to the ICC, Bilderberg’s lack of guaranteed membership, the breadth of its elite constituency, and its historically less doctrinaire political agenda have made it a more fl exible vehicle for transnational class formation. A good indication of this is the group’s migration from a predominantly cor- porate-liberal strategy, to one that in recent years appears more aligned with neoliberal structuralism. Indeed, by the mid-1990s organized labor was all but excluded—the single invited delegate being John Monks, General Secretary of the British-based Trades Union Congress. While labour was eff ectively shut out, neoliberal intellectuals—including Timothy Garton Ash of the Hoover Institute, Michael H. Armacost of the Brookings Institution, and William W. Lewis of the McKinsey Global Institute—have attended in numbers.5 4. For instance, see the Building Cooperation in Africa Report (December 2001) and the ICC Business Charter for Sustainable Development (April 1991). 5. Th e Spotlight Special Bilderberg Issue 1995–1996. William K. Carroll & Colin Carson74 Forging a New Hegemony? 75 Emerging at the watershed of recent economic globalization in 1973, Th e Trilateral Commission (TC) was launched from within the Bilderberg meetings by David Rockefeller as a forum to foster eff ective collaborative leadership in the international system and closer cooperation among the core capitalist regions of northern Europe, North America and Japan—the “triad.” It maintains a consul- tative ruling class tradition, bringing together transnationalized fractions of the business, political, and intellectual elite during several yearly meetings, which it convenes at the national, regional, and plenary levels. Unlike the secretive Bilderberg, however, the TC “sought to develop a profi le with greater transpar- ency, public activities and sophisticated publications, responding to the greater sensitivity towards public relations” (van der Pijl 1998:124). Consistent with this strategy, its magazine, Trialogue (fi rst published in October 1973), pioneered what has become a mainstay in the cultural arsenals of transnational business policy groups: the widespread dissemination of neoliberal opinion and analy- sis, as in the World Economic Forum’s World Link magazine. A director, three regional chairmen, and three regional executive committees guide the TC; its 350 members are chosen on a national basis. In marked contrast to the ICC, the TC’s attempts to enshrine the discipline of capital have generally favored elements of regulation. In this regard, its infl u- ential 1975 report, Th e Crisis of Democracy, called for stronger economic planning measures, including job training and active intervention in the area of work, all in the service of “sustained expansion of the economy” (quoted in Wolfe 1980, 298). Deeply motivated by the 1970s energy crisis, the TC has also lobbied for integrating capitalism’s (semi-)periphery into contexts of international regula- tion, including “allowing the neocolonies a symbolically greater voice in organi- zations like the IMF, [and] tying neocolonial economies even closer to Western fi nance” (see Frieden 1980:72). An infl uential series of ‘ Task Force Reports’ (or Triangle Papers) on this issue have been delivered over its three-decade history (e.g. Watanabe et al. 1983). Overall, the TC’s project is to institutionalize elite economic, political, and intellectual/cultural bonds between the North-Atlantic heartland and the Asia-Pacifi c and to expand the regulatory sphere of capitalist discipline to incorporate metropolitan labour and (more recently) peripheral states. Th ese aims draw it in line with Robinson and Harris’ (2000) neoliberal structuralist formulation. Founded two years earlier, Th e World Economic Forum (WEF) convened Europe’s CEOs to an informal gathering in Davos, Switzerland to discuss European strategy in an international marketplace. Organized by renowned business policy expert, Klaus Schwab, the meetings were intended to secure the patronage of the Commission of the European Communities, as well as the encouragement of Europe’s industry associations. By 1982 the fi rst informal gathering of ‘World Economic Leaders’ took place on the occasion of the Annual Meeting in Davos, bringing cabinet members of major countries and heads of international organizations (including Th e World Bank, IMF, GATT) together with a burgeoning core membership of top international capitalists. Th e WEF moved beyond the TC to establish “global initiatives” that distin- guish it as the most paradigmatic example of neoliberal structuralism. Initially, the Forum promoted a free-market conservative agenda, but by the mid-1990s persistent capitalist crisis forced it to adopt a more regulatory tack, a shift sig- naled when, “on the eve of the 1997 meeting, fi nancier George Soros unexpectedly denounced ‘the destruction of those values which do not produce commercial return’ and ‘the totalitarian tendency of unregulated market capitalism’” (quoted in van der Pijl 1998, 134). By early 1997 the new mood was expressed in a project on ‘human social responsibility’, followed by a litany of ‘social issue’ task forces culminating with the Global Health Initiative (2001) and the Global Governance Initiative (2001). Th ese initiatives crosscut with the widespread practices of Corporate Social Responsibility (CSR) among TNCs and the rise of a culture of ‘global corporate citizenship’ Sklair (2001) considers integral to transnational capitalist class formation. Unlike the ICC, Bilderberg, and TC, the WEF is organized around a highly elite core of transnational capitalists (the ‘Foundation Membership’)—which it currently limits to ‘1,000 of the foremost global enterprises.’ Invited ‘constituents,’ however, represent a variegated range of globalist elites, including members of the scientifi c community, academics, media leaders, public fi gures, and various NGOs. Constituents populate a hodgepodge of policy work groups and forums, including the InterAcademy Council, the Business Consultative Group and the Global Leaders of Tomorrow. Like the ICC, however, the WEF actively extends its geopolitical reach and infl uence. It has done so primarily through yearly meet- ings apart from Davos and beyond the triad, as in the 1996 meetings in Turkey, China, and India (Annual Report 1995 6:6). To formalize this expansionist agenda—or to, in its own words, “advance regional development and coopera- tion in the global economy”6—the WEF recently established a distinct operating body called the Centre for Regional Strategies (CRS). Accordingly, Robinson and Harris (2000) have described the WEF as the “most comprehensive trans- national planning body…and a quintessential example of a truly global network binding together the [TCC] in a transnational civil society” (30). 6. https://members.weforum.org/site/homepublic.nsf/Content/Our+Organizat ion5CForum+Centres (site unavailable 4/1/03). https://members.weforum.org/site/homepublic.nsf/Content/Our+Organization%5CForum+Centres William K. Carroll & Colin Carson76 Forging a New Hegemony? 77 Th e last group to have taken up a niche within the fi eld of global elite policy-planning is the World Business Council on for Sustainable Development (WBCSD), founded in 1995. It is also the only group that can be characterized within Robinson and Harris’ (2000) typology as neoliberal regulationist. Formed in a merger of the Geneva-based Business Council for Sustainable Development and the Paris-based World Industry Council for the Environment (a branch of the ICC), it instantly became the preeminent business voice on the environment. By 1997, WBCSD membership comprised 123 top-TNC chief executives. A child of the UN’s Conference on Environment and Development (UNCED) 1992 Rio Earth Summit, the WBCSD refl ects a maturing elite awareness that entrenchment and expansion of transnational enterprise must be coupled with consensus over environmental regulation. Drawing primarily on the expertise and prestige of senior transnational executives, it articulated a critical connection between neoliberalism and regulatory struggles over the envi- ronment, especially those associated with the UN Environmental Programme (UNEP) and the UN Conference on Trade and Development (UNCTAD). What makes the WBCSD unique in the global policy fi eld are its eff orts to sur- pass the prevailing dualism of “business versus the environment” by forwarding a more comprehensive vision of capitalist social and moral progress—anchored by its central axiom of ‘eco-effi ciency.’ 7 Within this retooled version of sustainable development business, governments and environmental activists make conces- sions around a general interest in sustaining both the health of the natural world and the “health” of the global economy. Th e discourses and strategies of the WBCSD work to advance a global regulatory perspective (Robinson and Harris 2000) which moves beyond neo- liberal structuralism. Th e WBCSD’s refl exive discursive and organizational frameworks endeavor to draw realms free-market conservatives call ‘externali- ties’—from employee relations to the health and safety of consumers—into an inclusive regulatory regime. Th e practices and discourses of corporate environ- mentalism—now employed by TNCs from Procter & Gamble and Mitsubishi to Monsanto and Broken Hill Proprietary—are vital in this regard, and have in their own right contributed to a persuasive globalizing capitalist ideology (Sklair 2001). What the WBCSD furnishes is a refl exive orchestration of these corpo- rate initiatives into a class-wide hegemonic project. With these fi ve policy groups we can see how variants of transnational neo- liberalism have found organizational bases in the policy-formation fi eld. Only the International Chamber of Commerce functions from the perspective of free-market conservatism and speaks for and to a strictly business-centered con- stituency. Th e Bilderberg group, Trilateral Commission, and World Economic Forum in their own ways incorporate broadly neoliberal structuralist perspec- tives. Th e most recent addition to the fi eld, the World Business Council for Sustainable Development, orients itself primarily in terms of neoliberal regula- tionism. Taken as a whole, these global policy groups can be regarded as agencies of transnational capitalist class formation. Th ey provide intellectual leadership that is indispensable in the ongoing eff ort to transform transnational capital from an economically dominant class to a class whose interests take on a sense of universalism. Th e empirical questions to which we turn now concern the social relations that embed these groups within a structure of global corporate power. T H E GLOBA L COR POR AT EPOLIC Y N ET WOR K Our empirical analysis maps the social structure of the global corporate elite, the collection of leading corporate directors who participate in the network of major corporations and transnational policy groups. Th is elite is not coexten- sive with Sklair’s “transnational capitalist class.” His conception of the TCC goes further and includes transnational executives, a globalizing state fraction, a globalizing technical fraction, and a globalizing consumerist fraction (2000: 17). Yet our global corporate elite does include the major capitalists who exercise the investment and top-level management functions within the world’s largest corporations, as well as the organic intellectuals whose advice, as outside direc- tors, is sought by the same companies. We are interested in this corporate elite’s participation on the directorates of global policy groups, and in the contribution that such participation makes to overall elite integration. To date there has been no systematic study of the network of leading corporations and policy groups at the global level. Research carried out in national contexts suggests that cor- porate-policy interlocks contribute substantially to elite integration and to the hegemony of corporate capital (Useem 1984; Domhoff , 1998; Carroll and Shaw, 2001). Case studies such as Gill’s (1990) and anecdotal analyses such as van der Pijl’s (1998) suggest that much the same applies in the global fi eld. To explore this issue, our analysis poses three research questions: 1. At the level of individuals, who are the corporate directors at the centre of the corporate-policy network and how do their group affiliations create an inner circle of corporate governance and policy planning? 2. At the level of organizations, what is the basic shape and form of the 7. “Eco-effi ciency” was fi rst coined by the BCSD in 1992. In its 1997 Annual Review, the WBCSD defi ned eco-effi ciency as “a management approach ... that allows compa- nies to improve their environmental performance while meeting the demands of the market…[by increasing] economic and ecological effi ciency” (8). William K. Carroll & Colin Carson78 Forging a New Hegemony? 79 interlocking directorates among the policy groups and between them and the world’s largest corporations? 3. What contribution do the global policy groups make to transnational corporate-elite integration? Th e point of the analysis is to investigate one dimension of transnational capitalist class formation—the corporate-policy network—with an eye toward what it tells us about the structural sources of elite integration as well as tension and possible fi ssure. SA M PLE A N D DATA Our research design obliged us to achieve a substantively representative sample of the leading corporations and global policy groups in the world capital- ist system, and to depict the interlocks between these organizations at a single moment (circa year end 1996). Th e fi ve organizations discussed above provide a reasonable instantiation of the leading policy groups. In selecting corporations, we needed to achieve an adequate representa- tion of the various sites in the world economy where corporate head offi ces are domiciled. Th is meant combining quantitative selection criteria with qualitative considerations around regional and sectoral representation. As Fennema (1982) found when he carried out the fi rst study of the transnational corporate network, selecting fi rms purely according to size can yield a sample that is entirely skewed toward the dominant economies in the world system. In our research, a strict size criterion would have selected into our sample a fairly balanced range of corpora- tions based in the triad—Europe, North America, and Japan—but a paucity of corporations from the rest of the world. Since we wanted to explore the network positioning of leading companies based in diff erent national domiciles, it was important that the sample include suffi cient numbers from countries beyond the triad. Th e sample of corporations was constructed on the basis of a number of sources, beginning with the Global 500 published in the June 1997 issue of Fortune, which ranks fi rms by sales or revenue.8 A drawback in the Fortune listing is that revenue is not a particularly good measure of the size of fi nancial institutions, some of which have relatively small revenue streams (often equivalent to net income) compared to their asset size. Th e latter is a far more appropriate measure of the concentration of capital within fi nancial institutions; thus we divided the sample into two strata and selected from the sources at our disposal all fi nancial institutions with assets of $100 billion or higher and all non-fi nancial corporations with revenues of $14 billion or higher. All values were taken for the time closest to year end 1996 and denominated in US dollars. Th is yielded a sample of 300 corporations, summa- rized in the fi rst two columns of Table 1. To this initial sample we added 50 companies, in an attempt to represent domiciles and sectors that by the size criterion alone were thinly represented: 1. The most serious deficiency had to do with the semi-periphery. Only 17 companies were selected into the sample according to size, and 10 of these were industrial companies based in South Korea. To achieve a broader representation, we relaxed the quantitative criteria and selected the next-biggest 14 non-financials (the smallest of which had 1996 revenues of $5.7 billion) and the nine biggest financials based in the semi-periphery (whose assets ranged from $80 billion down to $30 billion), for a total of 40 firms based outside the centre. 8. Other sources consulted were Standard & Poor’s Register of Corporations, Directors and Executives (1997), the top 1000 banks listed in Th e Banker ( July, 1997), the Global 1000 listed in Business Week, the Times 1000, and individual corporate listings in Moody’s International Manual. Table 1 – Composition of the Global Top 350 Sample INITIAL SAMPLE ADDED TO INITIAL SAMPLE TOTAL Biggest Non- Financials Biggest Financials Smaller Non- Financials Smaller Financials European Non-Financials 71 0 17 0 88 North American Non-Financials 71 0 7 0 78 Asia-Pacific Non-Financials 51 0 2 0 53 Semiperipheral Non-Financials 17 0 14 0 31 European Financial 0 44 0 0 44 North American Financial 0 24 0 0 24 Asia-Pacific Financial 0 22 0 1 23 Semiperipheral Financial 0 0 0 9 9 210 90 40 10 350TOTAL William K. Carroll & Colin Carson80 Forging a New Hegemony? 81 multiple directorships later verifi ed by cross-checking sources.12 Our analysis is restricted to what we consider the global corporate elite— those directing at least one corporation and one other organization in our sample (whether corporation or policy group). Th ese 622 individuals are a subset of the 6,751 directors of the world’s major corporations, 92 of whom held only one directorship in a sample organization at year end 1996. Many of the individu- als holding single directorships may be important business leaders in national contexts, and may thereby extend the reach of the global network into diverse national settings. However, they do not contribute to the global network’s coher- ence, which is our prime concern. Similarly, although some of the 53 corporations lacking any interlocks with other organizations in our sample may be ultimately connected to the global network via interlocks with other companies not included in this study, the focus here is on the fi rms that contribute one or more directors to the global network. Th us, our analysis features the 622 corporate directors and the 302 organizations (fi ve of them policy groups) that constituted the global corporate-policy network at year end 1996. R E SU LTS 1: A GLOBA L COR POR AT E ELI T E ? Our fi rst research question directs attention to the individuals who carry the transnational network: who are they and how do they create social structure through their group-affi liations? Table 2 categorizes the global corporate elite into eight types, based on (a) position in the network of corporate affi liations and (b) position in the network of policy-group affi liations. Th e fi rst fi ve types (a–e) depict the network’s inner circle of cosmopoli- tan linkers (cf. Useem 1984)—105 corporate directors whose affi liations span national borders, or link global policy boards to each other. Th ese directors are cosmopolitans much in the sense originally employed by Gouldner (1957): they are oriented not toward particular national fi rms and networks but toward a 2. The low value of the Canadian dollar at the time of the study meant that only one non-financial company based in Canada qualified for the sample in terms of its size. To enable a global network analysis that would include the largest Canadian non-financials, we selected seven additional such companies, bringing the Canadian total to eight non- financials and four financials. 3. To achieve the same number of British-based companies as in Fennema’s (1982) study,9 we selected an additional 17 non-financials, bringing the British total to 26 non-financials and eight financials. 4. To round out the portion of the sample domiciled in Japan we added two non-financials and one financial institution, whose sizes very nearly met our initial criteria, bringing the Japanese total to 51 non-financials and 22 financials. Our sample, then, combines quantitative and qualitative considerations in representing the leading corporations worldwide. According to a strict quantita- tive criterion one could say we have over-represented non-fi nancial corporations based in the UK, Canada and the semi-periphery, as well as fi nancial institutions based in the semi-periphery. However, given our primary concern with the social structure of the corporate elite, including the patterning of interlocking across regions, this over-representation is a methodological necessity. We next identifi ed the corporate directors who as of year end 1996 held posi- tions in two or more organizations (whether corporations or policy groups), thus carrying interlocks between the organizations. Th is meant fi rst recording the full names of all directors of our 350 corporations, including, in the case of two- board systems such as Germany’s, both the supervisory and the management board.10 Th e directors of the fi ve policy groups11 were recorded into the same fi le, and the names were then alphabetically sorted, with apparent or suspected 9. Since research related to this one (Carroll and Fennema, 2002) endeavored to replicate Fennema’s study of the global corporate network, it was also important to include suffi cient numbers of fi rms from the countries featured in Fennema’s sample, which purposively over-represented certain domiciles, in particular Britain. 10. Sources included Standard & Poor’s Register of Corporations, Directors and Executives (CD-rom version, 1997), Moody’s International Manual (1997) and corporate annual reports, whether hard copy or downloaded from websites. 11. In the case of the Trilateral Commission, which functions as a large board, we included all members as of October, 1996. For the ICC we included the Executive Board. For the WEF we included both the Foundation Board and the Members’ Council. For the WBCSD we included the Executive Committee as well as the individuals designated as representatives of the member companies. In the case of the Bilderberg Conference (which has no membership), only people who attended in both 1996 and 1997 were included. Information on policy-board membership was obtained either directly from the group, from the group’s annual report, or (in the case of Bilderberg) from published listings of attendees, as in Reuters’s listing for 1997. 12. For Japanese and South Korean names, whose translated versions in English sometimes vary, colleagues from these countries kindly assisted with the verifi cation pro- cedure. We are grateful to Professors Koji Morioka (Kansai University), Unno Yahiro (Kanazawa University), and Bokhyun Cho (Taejon National University of Technology, Seoul) for their indispensable assistance. William K. Carroll & Colin Carson82 Forging a New Hegemony? 83 wider fi eld of action. Th e next three types (f–h) describe directors whose affi lia- tions position them on the periphery of the global network. From this typology we can note that: • Most of our global corporate elite have corporate affiliations only within one country. The contribution that the 400 purely national corporate linkers (type h) make to elite integration is indirect at best and contingent upon their serving on common boards with members of the inner circle. • At the other end of the spectrum, the most cosmopolitan directors are the six people who sit on multiple policy boards and whose corporate directorships cross national borders (type a). Indeed, it is clear that the international network is primarily a configuration of national corporate networks, integrated for the most part through the affiliations of a few dozen individuals who either hold transnational corporate directorships or serve on two or more policy boards. It seems that Carroll and Fennema’s (2002) characterization of the transnational corporate network as a superstructure resting upon resilient national bases applies also to the corporate-policy network. • There is an unsurprising relationship between participation in the transnational corporate network and participation on the policy boards.13 Individuals well-connected in the world of global corporate business also tend to be well-connected in the world of elite policy-planning and consensus formation, suggesting a highly centralized corporate-policy network. • Notwithstanding this relationship, the policy groups draw a number of “locals” into a field of global governance. Of the individuals who direct two or more companies that are all within a single country, five serve on multiple policy boards, and 34 direct one policy group. Finally, among those with only one corporate directorship, six direct multiple policy groups and 83 serve on single policy boards. As a transnational structure, the network is carried by a few dozen cosmo- politans—94 transnational corporate linkers (types a, d and e) and an additional 11 directors whose corporate affi liations are not transnational but who sit on mul- tiple global policy boards (types b and c). Th rough their networking, these 105 individuals make the most immediate structural contributions to transnational class formation. One way of displaying these contributions is to examine suc- cessive “layers” of the inner circle of cosmopolitan directors. In Figure 1 we show the confi guration of 18 corporations and four policy groups that is knit together by six individuals comprising type a of Table 2. We have drawn the lines from the persons (enclosed in rectangles) to the organizations (the policy groups are enclosed with curves): the arrows point to the organizations with which each person is affi liated. Th is most cosmopolitan nucleus is densely packed. Most of the six directors sit together on multiple policy boards. Indeed, Bertrand Collomb, (president of Lafarge, a French-based TNC not large enough to qualify for our sample, and 1997 “manager of the year” according to Le Nouvel Economiste) sits on all four policy boards and thus meets Minoru Murofushi, Chair of Itochu Corporation, on three of them. However, with the exception of Percy Barnevik (president of Swiss-based Asea Brown Boveri [ABB]) and Peter Sutherland (vice chair of BP and also a director of ABB), there are no interlock- ing corporate directorships among the six. Within this nucleus the integrative function of the policy boards is clear: without them these transnational linkers would be for the most part detached from each other; with them they comprise a tightly-knit social unit, with representation from the USA (Paul Allaire, president of Xerox), Britain (Sutherland), Japan (Murofushi), and continental Europe (Barnevik, Collomb, and Etienne Davignon [deputy chair of the man- agement board of Fortis]). In all, eleven European, six North American and one Japanese company are linked into this nucleus. In Figure 2, the network is extended to all corporate directors having two or more policy-group affi liations, including the 11 non-transnational linkers that comprise types b and c in Table 2. Th is addition brings in 14 more corpora- Table 2 – Positions in the Corporate-Policy Network Type Description N of Cases % a Transnational linker, 2 or more policy boards 6 1.0 b National linker, 2 or more policy boards 5 0.8 c 1 corporate board, 2 or more policy boards 6 1.0 d Transnational linker, 1 policy board 14 2.3 e Transnational linker, no policy board 74 11.9 f National linker, 1 policy board 34 5.5 g 1 corporate board, 1 policy board 83 13.3 h National linker, no policy board 400 64.3 Total 622 100.0 13. While 22 of the transnational corporate linkers have one or more policy-group affi liation, only 9 of national corporate linkers, and only 0.14 of all single-corporation directors do. Th is last percentage is calculated on the basis of all 6,129 single-corporation directors. William K. Carroll & Colin Carson84 Forging a New Hegemony? 85 tions—eight from the USA, three from Japan, two from Europe and one from Australia, and furnishes a picture of the corporate-policy network’s densely- connected core. Th e integrative role of four policy groups stands out: each is a meeting place for between eight and eleven of the 17 corporate directors, with the Trilateral Commission being the most centrally positioned. Th e TC and WBCSD serve as transnational meeting points for all three of the Japanese directors, but equally, these individuals serve as ambassadors between the fi elds of global policy work and of Japanese corporate governance, while also linking the TC with the WBCSD. Th e corporate-policy core includes the fi fth policy group—the ICC—on which sit Livio Desimone (chair of the WBCSD and CEO of 3M), and Helmut Maucher (vice president of the ICC, CEO of Nestle and director of the WEF). Maucher’s directorship with ABB further underscores the importance of this company in the global network: all three ABB directors in the network core also direct the Swiss-based WEF (in fact, all three sit on the 12-man Foundation Board at the heart of the WEF’s planning process), and two of them also are regular attendees at Bilderberg. Other centrally positioned corporations in the network core include US-based General Motors (itself interlocked with ABB via Barnevik) and Sara Lee (interlocked with GM via Sara Lee president John Bryan), and Toronto-based Canadian Imperial Bank of Commerce (CIBC). As a group, the 17 corporate directors whose group affi liations place them at the core of transnational class formation show an obvious Euro-North American bias. Corporations sited on the semi-periphery are entirely absent from the net- work, and only fi ve Asia-Pacifi c companies (four of them Japanese) are repre- Figure 1 – The Nucleus of Six Transnational Linkers and their Multiple Policy-Group Affiliations Figure 2 – The Core of 17 Corporate Directors with Multiple Policy-Group Affiliations William K. Carroll & Colin Carson86 Forging a New Hegemony? 87 ical details provided ample bases for categorizing each into a ‘national’ affi liation. For executives, we took the national domicile of their home fi rm to indicate their national base of operations; for outside directors we considered other biographi- cal information, including the locus of their careers and residence. Table 3 com- pares the national composition of our sample of corporations with the national composition of the inner circle. At the centre of things, Europeans and North Americans entirely predominate. Although our sample includes the 40 largest companies of the semi-periphery, corporate directors based outside the centre of the world system are completely absent from the group of 105. Th is implies that whatever interlocks link the network’s inner circle to its margins emanate from the centre, not the semi-periphery, of the world system. It is also noteworthy that certain national sites are over-represented among the cosmopolitans—especially such middle powers as Belgium, the Netherlands, Switzerland and Canada, along with three major EU powers, Britain, Germany and France. Conversely, three advanced capitalist states are under-represented— the USA (slightly), Italy (more so) and Japan (extremely). Th us, while Belgium and Japan each contribute six directors to the inner circle, Belgium’s proportion- ate contribution is far greater than Japan’s. A rudimentary analysis of class positions revealed that 39 cosmopolitans were executives in a Top 350 corporation and 26 were executives in other com- panies. Th e remaining 40 were corporate advisors, 12 of whom were retired corporate executives serving as outside directors to various fi rms. Th e inner circle is primarily a collection of leading corporate capitalists, directing some of the world’s largest companies as well as companies not in our sample. A further cross-classifi cation showed that proportionately more corporate capitalists than corporate advisors tend to be on the policy boards. Leadership in the policy domain has not been delegated to a separate stratum of organic intellectuals, or, put another way, top global capitalists serve also as organic intellectuals. Finally, a look at gender confi rmed that male dominance continues to be the order of the day at the very top of the global corporate world. Only six members of the inner circle were women, and four of these were advisors to corporations, not executives. R E SU LTS 2 : T H E N ET WOR K A S A N I N T E R ORGA N IZATIONA L FI ELD We now move to our second research question, and to a representation of the corporate-policy network as a set of inter-organizational relations. Such relations can be traced among the policy groups and between them and the cor- porations. We saw in Figure 3 that the group affi liations of just 17 cosmopolitans generate many interlocks among the policy groups. Th ese are displayed as an sented. At this point, we approach the limits of visually depicting the entire set of relations among persons and organizations. Yet, these top layers are part of a deeper structure that includes, in particular, the 88 transnational corporate link- ers who do not sit on multiple policy boards. Before moving to a structural analy- sis of the entire network, it is worth considering the national, class, and gender composition of the inner circle of 105 cosmopolitan corporate directors. Although the inner circle’s members are indeed cosmopolitans, this does not render them rootless. Analysis of their corporate affi liations and other biograph- Table 3 – Distributions of Companies and Individuals by National Domicile Domicile Percent of Firms Percent of Inner Circle Difference Canada 3.4 5.7 +2.3 USA 25.7 21.0 -4.7 Netherlands 2.6 7.6 +5.0 UK 9.7 16.2 +6.5 Germany 9.1 14.3 +5.2 France 7.1 11.4 +4.4 Italy 3.7 1.9 -1.8 Switzerland 2.0 3.8 +1.8 Sweden .9 2.9 +2.0 Belgium 1.1 5.7 +4.6 Spain 1.1 1.9 +0.8 Norway .3 0.0 -0.3 Australia .9 1.9 +1.0 Japan 20.9 5.7 -15.2 Brazil 1.4 0.0 -1.4 Mexico .6 0.0 -0.6 Venezuela .3 0.0 -0.3 Argentina .3 0.0 -0.3 Russia .9 0.0 -0.9 Turkey .3 0.0 -0.3 South Korea 3.7 0.0 -3.7 Hong Kong .9 0.0 -0.9 Taiwan .6 0.0 -0.6 Singapore .9 0.0 -0.9 Malaysia .3 0.0 -0.3 India .3 0.0 -0.3 South Africa 1.1 0.0 -1.1 Total 100.0 100.0 0.0 William K. Carroll & Colin Carson88 Forging a New Hegemony? 89 inter-organizational network in Figure 4. TC emerges as a central meeting point for the transnational corporate elite, but the WBCSD also plays a highly inte- grative role. Th e Bilderberg conferences and the WEF each bring more than a dozen major corporate directors together,14 but the International Chamber of Commerce’s board has a lower profi le in the transnational corporate elite. Th is must be understood in terms of the ICC’s specifi c position in the organizational ecology of transnational neoliberalism. In contrast to the other four groups, the ICC’s distinctive contribution to transnational class formation is to integrate capitalism’s centre with its margins; hence the ICC board blends a smattering of the global corporate elite with various representatives of national and local capital.15 As for the inter-organizational relations, the 17 interlocking directors estab- lish close relations among four of the fi ve policy groups. For instance, four of the 15 leading corporate directors who sit on the WEF board are also regular attendees at Bilderberg. Extensive overlapping memberships among four of the fi ve groups suggest that inter-group diff erences in policy and strategy do not make for cleavages within the corporate elite. If direct interlocks among policy boards provide some basis for elite con- sensus-formation across organizations pursuing distinct yet complementary agendas, another source of elite integration can be found in the extent to which the social circles of the policy groups intersect. Do the same corporate boards that interlock with one policy group also interlock with another? Such overlaps would point to another source of corporate-policy group symbiosis—pulling the policy boards closer to each other via the mediation of corporate boards, and vice versa, and collaterally integrating the lifeworld of the global corporate elite. A key issue in this regard is which corporations—and which “national fractions” of capital—are centrally positioned in the structure of overlapping social circles. To highlight the main corporate participants in the policy-group social circles, Table 4 lists the 27 corporations maintaining at least three directorship interlocks with the policy groups. Heading the list is Zurich-based industrial conglomerate ABB, whose direc- tors serve on all fi ve policy boards. Interestingly, eight of the nine corporations that are most interlocked with the policy boards are industrials, the exception being the CIBC; and seven of those nine have one or more directors on each of the four most central policy groups. Although there is no one “nationality” that predominates in the policy-board social circles, the North Atlantic presence is striking. Th ere are also some interesting patterns in the group-affi liations. Th e board of the World Economic Forum is interlocked with a good many European fi rms; the Bilderberg Conference and World Business Council for Sustainable Development recruit from both European and American fi rms; and the TC includes directors from all regions of the world-system’s core—most remarkably, four directors from Osaka-based Kansai Energy. Th e TC is interlocked with all but one of the 27 corporate boards, and it maintains “thick” (multiple-director) interlocks with a number of fi nancial institutions, including Chase Manhattan. In contrast, the WBCSD tends to share directors with industrial companies such as Xerox, GM, Nestle, BP, Unilever, and ABB. Th ere thus appear to be elec- tive affi nities between large fi nancial institutions and the TC on the one hand, and certain industrials and the WBCSD on the other. Th e network of interlocking directorates among these 27 corporations and Figure 3 – Number of Interloks Among Five Global Policy Groups Note: Boxes are proportionate in size to the number of corporate-elite members affiliated with each group (indicateds within each box). Line thickness ref lects the number of shared elite members. 14. In addition to the 18 people who attended both the 1996 and 1997 Bilderberg meetings, 12 attended the Conference in 1996 but not 1997 and nine attended in 1997 but not 1996. 15. Specifi cally, its 1996 executive board of 27 members and international offi cers included 12 corporate directors based on the semi-periphery, 11 based in Europe, three in the USA or Canada, and one in Japan. William K. Carroll & Colin Carson90 Forging a New Hegemony? 91 fi ve policy groups is shown in Figure 4, where we can see the extensive ties that converge on four of the groups, each heavily interlocked with the others. Although some of the corporate boards interlock transnationally—e.g., Unilever, ABB, BP—most share directors only with companies based in the same country, and this is especially the case for thick ties such as those between Xerox and Sara Lee or Deutsche Bank and VW. Th is persistence of national networks in conjunction with a modest transnationalization of interlocking has already been reported by Carroll and Fennema; what our analysis adds is a sense of how, amid the persistence of national corporate power structures, the global policy groups fulfi ll an integrative function, bringing together corporate directors and capitalist interests from various quarters of the world-system’s centre. Extensive interlocking with policy boards is the prerogative of the relatively few corporations listed in Table 4, all but two of which interlock with multiple policy groups. Th e 27 corporations, barely eight per cent of our sample, account for 128 of the 305 directorship interlocks between all corporations and the fi ve global policy groups. Moreover, corporations whose boards overlap with the policy groups also tend to be central in the network of corporate interlocks: among our 350 corporations, the Pearson correlation between n of interlocks with policy groups and n of interlocks with other corporations is 0.434. Evidently, the corporate-policy network is quite centralized. Yet it also spreads out to include a great many companies whose directorates are only weakly linked to the policy Table 4 – Numbers of Corporate Directors on Five Global Policy Boards, 1996 Corporation Domicile ABB Swiss CIBC Canada GM USA Unilever Dutch/UK Sara Lee USA Xerox USA BP UK Aquitine France Nestle Swiss Hong Kong Savings Bank UK Fina Belgium Generale Bank Belgium Ericsson Sweden Kansai Energy Japan 3M USA American International USA Chase Manhattan USA Dayton Hudson USA Lucent USA Smithkline Beecham UK Deutsche Bank Germany Siemens Germany VW Germany Itochu Japan American Express USA Dupont USA Prudential USA TC 2 4 1 3 2 2 1 1 1 2 3 3 1 4 2 3 4 2 2 1 1 1 2 1 1 0 3 WBCDS 2 1 3 2 1 3 2 1 2 1 0 0 0 0 1 0 0 1 1 1 0 0 0 1 0 1 0 BLD 2 2 2 1 3 2 1 2 0 0 1 1 2 0 0 1 0 0 1 1 1 1 0 0 2 1 0 WEF 3 1 2 1 1 0 1 1 1 1 0 0 1 0 0 0 0 0 0 0 1 1 1 1 0 1 0 ICC 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 Total 10 8 8 7 7 7 5 5 5 4 4 4 4 4 4 4 4 4 4 3 3 3 3 3 3 3 3 Figure 4 – Interlocks Among Five Global Policy Groups and 27 Centrally- Positioned Corporations Note: National domiciles are indicated by closed curves. William K. Carroll & Colin Carson92 Forging a New Hegemony? 93 groups or to other major global corporations. Beyond the 27 centrally-positioned corporations are 52 companies with two directorship interlocks to the policy groups and 73 fi rms with one policy-group interlock. However, 198 of our 350 corporations, including nearly all companies domiciled in the semi-periphery, share no directors with the policy groups. Th is exclusion of the semi-periphery has been built into the declared constituency of the Bilderberg group and the TC (neither of which included in 1996 a single director of a semi-peripheral fi rm), but it largely applies to the other groups also.16 Th e only really salient regional fracture in the network is the massive divide between the worlds system’s centre and its (semi-)periphery. R E SU LTS 3 : T H E I N T EGR ATI V E CON T R I BU TION OF ELI T E POLIC Y GROU PS To test the integrative impact of policy-board affi liations we began with our entire corporate elite (622 individuals) and our entire sample of organizations (355). Since 53 of the 350 fi rms did not interlock with any organizations in our sample, this network has 924 points: 622 people, 297 corporations and fi ve policy groups. All but eight companies and fi ve people are ultimately connected, forming a dominant component that includes nearly all the persons and organizations that participate in the practice of interlocking directorships in one way or another. Th e network core that we mapped in our initial sociograms reaches to include most of the world’s major corporations. We next determined the size of the dominant component based purely on intercorporate ties. Leaving the policy-group affi liations aside, the dominant component shrinks only slightly, losing 18 companies and 23 people grouped into 11 small components. Although the n’s are small, these results show that some major corporate boards (like Toshiba, Sumitomo, NEC, Samsung, Montedison, Anglo-American Corp., Compaq, Cargill, and Koc Holdings), and some coun- tries (like South Korea, Turkey and South Africa) are brought into the transna- tional network purely through policy-group interlocks. Th e global policy boards have defi nite local impacts, as they establish portals into the transnational busi- ness community. Yet even without these mediations the intercorporate network extends to most of the world’s leading fi rms. As a more systemic test of the policy boards’ impact, we calculated the extent to which corporate interlocks with the fi ve groups reduce the distance between corporations in the global network. Given the centralized character of the corpo- rate-policy network, and the extensive interlocks and overlaps among the policy boards, we might plausibly expect the policy boards to function as “hubs,” pulling the corporations more closely together as corporate directors collaborate in the policy fi eld. With this working hypothesis, we examined the geodesic distances between points in the inter-corporate network, with and without the mediating ties pro- vided by policy-group affi liations. At this systemic level, the contribution of the policy groups to overall network integration is quite striking (see Figure 5). Th e density of interlocking for the entire dominant component of 271 corporations is .024; i.e., 2.4 of all pairs of companies are directly interlocked. However, only 10.1 of pairs of fi rms can reach each other via a third fi rm (distance=2). Yet when we include relations that are mediated by corporate ties to common 16. Th e WBCSD is interlocked with 51 corporate boards, but shares a director with only one semi-peripheral fi rm (Seoul-based Samsung Electronics); the WEF is interlocked with 27 corporations, but shares a director with only one semi-peripheral fi rm ( Johannesberg-based Anglo-American Corp.); the ICC is interlocked with seven corporations, but shares a director with only one semi-peripheral fi rm (Istanbul-based Koc Holdings, the investment vehicle of ICC President Rahmi Koc). Figure 5 – Distribution of Intercorporate Distances in the Dominant Component, With and Without Mediation of Policy-Group Ties William K. Carroll & Colin Carson94 Forging a New Hegemony? 95 policy groups, fully 25.8 of pairs of fi rms are linked at one remove. And whereas through corporate boards alone, only 32.6 of all pairs of fi rms are connected at two removes or less, 72.2 can reach each other by no more than two intermedi- aries when paths running through the policy boards are included. At the outer reaches of the component, the diameter drops from 15 to nine when paths run- ning through the fi ve policy boards are included in the calculation of distances. A key issue is whether the broad pattern of participation in the policy groups draws corporate capital sited in particular locations in world system into the international business community. Do corporate-policy interlocks reduce the average distance among corporations based in diff erent countries? Figures 6 and 7 present aggregated sociograms in which each point represents the set of fi rms domiciled in a given country and line thickness represents the mean proximity between fi rms based in two countries. Th ese plots depict inter-national prox- imities based on a non-metric scaling of mean distances; hence within each two- dimensional space the distances correspond fairly well to actual mean distances between domiciles.17 When, in Figure 6, only corporate interlocks are considered in calculat- ing distances it is north-west continental Europe that is most transnationally integrated: mean distances among the German (GER), Dutch (NETH), Swiss (SWIS), Swedish (SWE) and Belgian (BEL) networks are often less than 3.0. For instance, the mean distance between Dutch and Swiss fi rms is 2.44—not much higher than the mean distance among Dutch fi rms (1.86) or among Swiss fi rms (1.60, the lowest mean distance in the network). France (FR) is also part of this continental bloc, with relatively close ties to corporate Belgium (mean dis- tance=2.57) and Germany (2.86). However, on average, fi rms sited in Britain and the USA are not as proximate, either to the European continent or to each other, perhaps refl ecting looser business systems organized more around “exit” options than around the exercise of “voice” within stable relations of corporate control, as on the continent (Carroll and Fennema 2002). Th e mean distances among Canadian-based fi rms (CAN, 1.91) resemble those of north-west Europe, and the ties linking Power Corporation and the CIBC to Belgian and French cor- porations make for a certain propinquity among these domiciles. Spanish (SP) and Italian (ITAL)–based fi rms and companies based in Australia (OZ) and Hong Kong (HK) are relatively peripheral. Mexican and Japanese corporations are very peripheral, the largest mean distances in the international network being 9.88 between Italian and Japanese fi rms and 9.33 between Mexican and Japanese fi rms. When the corporate-policy interlocks are included as indirect, mediating ties, in Figure 7, mean distances decrease sharply. Companies sited in the three Anglo-American countries—heavy participants on the policy boards—become fully integrated with the continental European block, whose own transnational proximities thicken. However, fi rms domiciled in southern Europe or outside the North-Atlantic heartland remain relatively peripheral. Th us, the basic pat- Figure 6 – Mean Inter-National Distances Among 271 Corporations, Based on Corporate Interlocks Only 17. Note that the thickest lines represent mean distances less than 2.5; the thinnest represent mean distances between 3.5 and 3.999. Mean distances greater than 4 are rep- resented as absent ties, although in fact all corporations in the component are, by defi ni- tion, ultimately connected. Th e full matrices of mean distances are available from the fi rst author. William K. Carroll & Colin Carson96 Forging a New Hegemony? 97 tern of diff erential regional participation in the network is maintained, even as the absolute distances drop. Th e largest mean distance, in an aggregated network that now includes Mexican and Japanese fi rms, is only 4.5 (between Mexico and Hong Kong and also between Australia and Mexico); the mean distance between corporate Japan and corporate Italy drops from 9.88 to 4.08. Indeed, it is clear that for corporate Japan the policy groups play an important bridging role into global management. By virtue of the participation of Japanese corporate direc- tors on the policy boards, the mean distances between corporate Japan and fi rms domiciled in the North Atlantic drop from a range of 6.15–8.00 to a range of 3.33–3.64. Th e one Australian based fi rm in the component can serve to illustrate how the mediation of corporate-policy interlocks pulls the transnational network together. J.B. Prescott, CEO of Broken Hill Proprietary, a major transna- tional mining company and Australia’s largest industrial fi rm, sits on both the WBCSD and WEF boards, thereby establishing a representation of Australian- based resource capital within the two Swiss-based policy groups. Th ese boards, as we have seen, are staff ed with directors of a great many corporations based primarily in Europe and North America, but also Japan. As we saw in Figure 2, the WEF board alone includes three major directors of three Swiss corpora- tions; hence Prescott’s participation in WEF governance establishes ties at one remove between BHP and these Swiss companies. Since the Swiss network is itself highly integrated Prescott’s WEF and WBCSD directorships have the eff ect of reducing the mean distance between BHP and the seven Swiss-based companies from 3.50 to 2.67. Th e impact with regard to Japan-based corpora- tions, several of whose directors sit with Prescott on the WBCSD, is even more dramatic. When the mediating impact of policy-group affi liations is included the mean distance between BHP and the Japan-based members of the component plummets from 8.40 to 3.75. DISC US SION In conclusion, let us fi rst revisit our three research questions and take stock of what we have learned. Th e fi rst question we posed concerned the role of key individuals at the centre of the network. We have found that a few dozen cos- mopolitans—primarily men based in Europe and North America and actively engaged in corporate management—knit the corporate-policy network together by participating in transnational interlocking and/or multiple policy groups. Th is inner circle creates the interlocks that make the network a transnational formation. A mere 17 corporate directors, some of whom serve on as many as four policy boards, create a plethora of relations among the policy groups. As a structure supporting transnational capitalist class formation, the network is highly centralized in terms of the individuals and organizations that participate in it. Yet from its core it extends unevenly to corporations and individuals posi- tioned on the fringes of the transnational network. Our second question focused on the organizational level, at which we found that the policy groups diff er markedly in the extent to which the directors of the world’s leading corporations participate on their boards. Th e International Chamber of Commerce (ICC), whose contribution to transnational capitalist class formation is focused around the integration of the centre with its margins within a discourse of free-market conservatism, is least involved at the core of the network. In contrast, the other four groups, which advocate more structuralist or regulationist variants of neoliberalism, are deeply enmeshed within the global corporate elite. Th ey are substantially interlocked with each other as well as with Figure 7 – Mean Inter-National Distances Among 271 Corporations, Including Paths Mediated by Five Policy Groups William K. Carroll & Colin Carson98 Forging a New Hegemony? 99 common corporate boards, a small number of which account for two-fi fths of all the corporate-policy links. Most signifi cantly, while the North Atlantic is especially well represented in the contingent of interlocked corporations, cor- porate capital domiciled outside the world system’s core states is almost entirely detached, suggesting that van der Pijl’s (1984) image of a North Atlantic ruling class has retained its cogency to the close of the twentieth century. Compared to this dominant pattern, other elements of possible fractionation—as in the elec- tive affi nities that appear to attract fi nancial capital to the Trilateral Commission (TC) and industrial capital to the World Business Council for Sustainable Development (WBCSD)—barely register. Finally, although the practice of interlocking corporate directorates already links most of the world’s leading corporations into a single network, corporate- policy interlocks make a dramatic contribution to global corporate-elite integra- tion. Th is additional layer of social structure, within which leading corporate capitalists step beyond their immediate economic interests to take up matters of global concern, pulls the directorates of the world’s major corporations much closer together, and collaterally integrates the lifeworld of the global corporate elite. But if the policy groups mediate and thereby strengthen inter-corporate relations they do so selectively, in a way that reproduces regional diff erences in participation. Th us, even as the presence of 27 Japanese corporate directors, distributed among three of the fi ve policy boards, pulls corporate Japan closer to the network’s North Atlantic centre-of-gravity, that centre becomes even more tightly bound through the heavy participation of North Americans and Europeans on the policy boards. Th ese fi ndings support the claim that a well-integrated global corporate elite or business community has formed, and that global policy groups have been instrumental in its formation. Whether this elite confi rms the arrival of a transnational capitalist class is partly a matter of semantics and partly a matter of substance. From one perspective, it is striking how selective participation in the corporate-policy network is, and how centralized its structure is. Within an already elite group of leading corporations and corporate directors, those who actually constitute the network comprise a small core of cosmopolitan individu- als and corporations, with a strongly Euro-North American bias. In contrast, most individuals who participate in the global network do not hold elite posi- tions beyond their home nation. As a mode of business activism, the network evokes the image of a vanguard more than a mass movement. Yet as we have seen, it comprises a single connected component, with considerable reach, and the policy boards eff ectively draw the national sub-networks into an integrated transnational structure. Moreover, claims about the formation of a transnational capitalist class do not depend exclusively on the structure of elite networks. Sklair (2001), for example, points to cultural practices—the worldly assumption of social responsibility, the shared ideology of consumerism—as integral aspects of transnational capitalist class formation. As Gramsci understood, class forma- tion involves both structure and culture, and although network analysis gives some purchase on the former we have done no more than telegraph some of the discursive elements of neoliberal globalization as a hegemonic project. However one might assess the thesis of transnational class formation, conspicuously absent from the corporate-policy network are corporations and capitalists based on the periphery and semi-periphery of the world system, and in this sense the network seems to present one facet of a collective imperialism, organized so as to help manage global capitalism from the centre (see Steven 1994). In the blending of persuasion and coercion that such management entails, the policy groups clearly seek to persuade. Th ey operate at one remove from the structural adjustment programs, “poverty reduction strategies” and other enforcement mechanisms, including the capacity for military intervention, that are the province of statist bodies, whether national or international. Th ey foster discussion of global issues among members of the corporate elite, often in com- bination with other infl uential political and professional elites. Th ey facilitate the formation of a moving elite consensus that is framed within one or another variant of neoliberal discourse. Th ey educate publics and states on the virtues of the neoliberal paradigm. In short, they are agencies of political and cultural lead- ership, whose activities are integral to the formation of a transnational capitalist class. Th e network of interlocks between policy boards and the world’s major corporations forms an important communication structure in this process. All fi ve of the policy groups are embedded in the global network, and with exten- sive interlocking among four of them and a key elite-level connection between the most “regulationist” and most “free-market” group,18 there is no evidence of political fracture along the lines of Robinson and Harris’ (2000) typology. By the same token, each group has its own history and modus operandi, occupies a unique niche in the organizational ecology of transnational neoliberalism, and fi nds a distinctive location in the network. We have seen that the ICC is com- paratively marginal to the life of the global corporate elite as we have defi ned it, yet its policy work sustains a very broad network that links local capital from sites throughout the world system into the centre, in a hard-line project of free- 18. See Figure 2, which shows that the Chair of the regulationist WBCSD is also an ex offi cio director of the ICC, owing to the ICC’s founding sponsorship of the WBCSD. William K. Carroll & Colin Carson100 Forging a New Hegemony? 101 market conservatism. In contrast, the Bilderberg Conference is exclusively Euro- North American and well ensconced in the corporate network, and its gatherings bring business leaders together with political leaders in informal discussions that have tended to promote a neoliberalism that retains a managerial role for the state. Th e World Economic Forum (WEF) and Trilateral Commission (TC), both strongly integrated with the corporate network, champion a similar project, but they render it more tangible in the activities of various working groups and the issuance of extensive policy documents and other texts. Both groups bring together agents and interests beyond the Euro-North American core and beyond the corporate elite per se, in explicit attempts to articulate a global political-eco- nomic interest. Finally, the WBCSD extends the general interest to the peaceful coexistence of capitalism and nature, and like the WEF and TC, draws Japanese business leaders into the network. Instead of political fracture, we submit that neoliberalism’s own pluralism, as enunciated by the diff erent groups, ensures that the consensus is a loose and variegated one, not a monolithic doctrine. Although our systematic data refer to 1996/1997, subsequent developments suggest that the policy groups have continued their eff orts to articulate a general transnational interest within a broadly neoliberal paradigm, but not without sig- nifi cant contestation, beginning with the “Battle in Seattle” that raged around the World Trade Organization’s 1999 Ministerial meeting and continuing through the mass protests that have greeted meetings of the IMF and World Bank in Washington (April 2000) and in Prague (September 2000), and the World Economic Forum in Davos (2000) and New York (2002). Against these de- legitimating moves from below, we can note that in July 1999 the UN Secretary- General and the President of the ICC announced a “global compact” between the UN and the private sector “to spread the benefi ts of globalization;” that as of 2000 the Trilateral Commission was restructured to include representation of the Asia-Pacifi c semi-periphery (People’s Republic of China, Taiwan, Th ailand); and that in 2000-2001 the World Economic Forum began to include NGOs representing “civil society” in its annual deliberations and designated a Non- Governmental Organizations Council. For its part the World Business Council for Sustainable Development continued to expand its project of corporate envi- ronmental hegemony, forging a crucial regulatory alliance of transnational capi- talist development. Th e structural analysis we have presented here provides only a glimpse of a contentious formation that is very much under construction. R E F E R E NC E S Balanya, B., Anne Doherty, Olivier Hoedeman, and Erik Wesselius (2000) Europe Inc: Regional and Global Restructuring and the Rise of Corporate Power, London: Pluto Press in association with Corporate European Observatory (CEO). Carroll, William K. and Murray Shaw (2001) “Consolidating a Neoliberal Policy Bloc in Canada, 1976 to 1996,” Canadian Public Policy, 27, 195–216. Carroll, William K. and Meindert Fennema (2002) “Is Th ere a Transnational Business Community?” International Sociology, 17, 393–419. Cox, Robert (1987) Production, Power and World Order, New York: Columbia University. Domhoff , G. William (1998) Who Rules America? Th ird Edition. Mountain View, CA: Mayfi eld Publishing Company. Fennema, Meindert (1982) International Networks of Banks and Industry. Th e Hague: Martinus Nijhoff Publishers. Frieden, Jeff (1980) “Th e Trilateral Commission: Economics and Politics in the 1970s,” in Holly Sklar, (ed.) Th e Trilateral Commission and Elite Planning for World Management. Boston: South End Press. Giddens, Anthony (1998) Th ird Way: Th e Renewal of Social Democracy. Cambridge: Polity Press. Gill, Stephen (1990) American Hegemony and the Trilateral Commission. Cambridge: Cambridge University Press. Gill, Stephen (1992) “Economic Globalization and the Internationalization of Authority: Limits and Contradictions,” Geoforum, 23: 3, 269–283. Gill, Stephen (1995) “Globalisation, Market Civilization and Disciplinary Neoliberalism,” Millennium: Journal of International Studies, 24: 3, 399–423. Gouldner, Alvin W. (1957) “Cosmopolitans and Locals: Toward an Analysis of Latent Social Roles, I,” Administrative Science Quarterly, 2, 281–306. Gramsci, Antonio (1971) Selections from the Prison Notebooks of Antonio Gramsci. New York: International Publishers. Hymer, Stephen (1979) Th e Multinational Corporation: A Radical Approach. Cambridge, England: Cambridge University Press. Jessop, Bob (2000) “Good Governance and the Urban Question: On Managing the Contradictions of Neo-Liberalism,” Comment on Urban21, Published in German in MieterEcho June (2000). Quoted online: Department of Sociology, Lancaster University, http://www.comp.lancs.ac.uk/sociology/soc075rj.htm (link verifi ed 4/1/03). Keck, Margaret E., and Kathryn Sikkink (1998) Activists Beyond Borders: Advocacy Networks in International Politics. Ithaca, NY: Cornell University Press. Mendes, Alfred (1994) “Bosnia, Bohemia & Bilderberg: Th e Cold War Internationale,” Common Sense: Journal of the Edinburgh Conference, Issue 16, 5–15. Overbeek, Henk (2000) “ Transnational Historical Materialism: Th eories of Transnational Class Formation and World Order,” in Ronen Palan (ed.) Global Political Economy: Contemporary Th eories, London: Routledge, 168–83. http://www.comp.lancs.ac.uk/sociology/soc075rj.htm William K. Carroll & Colin Carson102 Overbeek, Henk and Kees van der Pijl (1993) “Restructuring Capital and Restructuring Hegemony: Neo-liberalism and the unmaking of the post-war order,” in Henk Overbeek (ed.) Restructuring Hegemony in the Global Political Economy: Th e rise of transnational neo-liberalism in the 1980s, London: Routledge, 1–27. Peschek, Joseph G. (1987) Policy-Planning Organizations: Elite Agenda and America’s Rightward Turn, Philadelphia. PA: Temple University Press. Robinson, William I., and Jerry Harris (2000) “ Towards a Global Ruling Class? Globalization and the Transnational Capitalist Class.” Science & Society, 64:1, 11-54. Ronit, Karsten (2001) “Institutions of Private Authority in Global Governance: Linking Territorial Forms of Self-Regulation,” Administration & Society, 33: 5, 555-578 (November) Shaw, Martin (2000) Th eory of the Global State: Globality as Unfi nished Revolution. Oxford: Cambridge University Press. Sklair, Leslie (2001) Th e Transnational Capitalist Class. Oxford, Blackwell. Steven, Rob (1994) “New World Order: A new imperialism.” Journal of Contemporary Asia, 24, 271–96. Stiglitz, Joseph (1998). “More Instruments and Broader Goals: Moving Toward the Post-Washington Consensus.” Th e 1998 WIDER Annual Lecture (Helsinki, Finland), Th e World Bank, January 7, 1998. Useem, Michael (1984) Th e Inner Circle. New York: Oxford. Van der Pijl, Kees (1984). Th e Making of an Atlantic Ruling Class. London: Verso. Van der Pijl, Kees (1998) Transnational Political Economy and International Relations. New York: Routledge. Watanabe, T., Jacques Lesourne and Robert S. McNamara (1983) “Facilitating Development in a Changing Th ird World: Trade, Finance, Aid,” Task Force Report 27, Th e Trilateral Commission. Williamson, John (1990) “What Washington Means by Policy Reform,” in J. Williamson, ed., Latin American Adjustment: How Much Has Happened?, Washington: Institute for International Economics. World Business Council for Sustainable Development (1997) Annual Review, Conches- Geneva: World Economic Forum. Wolfe, Alan (1980) “Capitalism Shows its Face: Giving up on Democracy,” in Holly Sklar (ed.) Th e Trilateral Commission and Elite Planning for World Management, Boston: South End Press. World Economic Forum (WEF). (1995/96). Annual Report, Conches-Geneva: World Economic Forum. Carroll & Carson