What I forward here was sent by Jairus' friend Ulhas Joglekar to Michael Perelman's pen-l. The Keynesian David Yaffe (!) who has written an empirical analysis of globalization may be interested in this as well. rb Business Line Tuesday, May 14, 2002 The power of American MNCs C. P. Chandrasekhar Jayati Ghosh The recent release, by the Bureau of Economic Affairs of the US Ministry of Commerce, of the preliminary results of the 1999 benchmark survey of US direct investment abroad (Survey of Current Business April 2002), permits an assessment of the expansion and restructuring of the operations of US multinationals during the years of globalisation. In this edition of Macroscan, C. P. Chandrasekhar and Jayati Ghosh assess the degree to which the evidence supports conventional perceptions of the role of multinationals and the impact of their operations on developed and developing economies. AT THE CENTRE of the process of integration advanced by liberalisation and globalisation during the 1990s stands the multinational corporation. And in terms of visibility and importance, US multinationals such as General Motors, IBM, Microsoft and Coca-Cola, occupy a pre-eminent position. This makes the release by the US Government's Bureau of Economic Affairs (BEA) of the preliminary results of the 1999 benchmark survey of US direct investment abroad an event of some significance. A US MNC in the BEA data comprises a US parent company and its affiliates abroad. Much of the BEA data relate to majority-owned foreign affiliates (MOFAs), which accounted for 84 per cent of employment of all non-bank foreign affiliates. For some time now the BEA has been conducting benchmark surveys or censuses of these MNCs once in five years, and annual sample surveys for the intervening years. Comparable benchmark surveys are available for the years 1982, 1989, 1994 and 1999, permitting a preliminary assessment of the changing role and impact of multinational operations during the years of globalisation. The BEA's data do point to an acceleration of the expansion of US multinationals (Charts 1 and 2). Over the 17-year period 1982-99, the nominal gross product of and employment in US multinationals grew at an annual rate of 5.1 and 1.2 per cent respectively. This average trend conceals, however, a moderate acceleration in the rate of growth of gross product and employment from 4.3 and 0.1 per cent respectively during 1982 to 1989, to 5.7 and 1.9 per cent during 1989-99. Further, if we break down the 1989-99 period, the evidence seems to point to a substantial increase in rates of growth of gross product and employment from 4.7 and 0.3 per during 1989-94 to 6.6 and 3.5 per cent. These trends, Charts 1 and 2 indicate, separately hold for parents of US multinationals and their MOFAs as well. It is no doubt true that trends in nominal gross product figures may be vitiated by price movements. However, the facts that inflation has been subdued the world over during the 1990s and that employment figures follow gross product movements, make these trends representative of actual events. There are two sets of issues which the acceleration in the growth of US MNCs raise. First, are issues related to the impact of that acceleration on the structure of output and employment in the US economy. The second, are issues relating to the impact of these multinationals globally, and in particular on developing countries. The more conservative apprehensions regarding globalisation are based on the possibility that the growing importance of multinational corporations could reflect a process of hollowing out of economic activity in the developing countries and their relocation abroad, resulting in job and production losses in the metropolitan centres of global capitalism. Those who argue that liberalisation and globalisation are positive from the point of view of the developing countries also advance a less asymmetric version of this argument. In that view, globalisation, besides increasing the incomes of developed country corporations with attendant positive consequences in those countries, allows for an expansion of production, employment and exports in developing countries, driven by multinational corporations. In these views, the defining feature of foreign direct investment (FDI) during the years of globalisation is that each enterprise-level investment decision is a component of a larger process of relocation of whole industries from sites in more industrially developed economies to less developed ones, resulting in the global restructuring of industrial production. Relocation takes its `ideal' form, which involves the once-for-all or gradual closure by more developed country producers of capacity at home and the establishment by the same producers of equal or larger capacities in developing country sites. This trend towards relocation, if it occurs, reflects a change in the nature of global flows of FDI during the years of liberalisation and globalisation. With the wave of liberalisation that began in the early 1980s, it is argued, the relevance of foreign investment stimulated by import-substituting policies in the developing world has declined. The removal of non-tariff barriers to trade and the reduction of tariffs on most imports have done away with the need to jump barriers to control markets. Most developing country markets can be accessed as easily through imports from abroad of commodities either in their final form or ready for assembly. The corollary is that foreign investments aimed at catering to domestic markets must be competitive with imports accessible at relatively low tariff rates. That is, the segmentation of the world market that the import-substituting years implied and the consequent distinction between sites for local market-oriented production and world market-oriented production is disappearing. Investments aimed at catering to domestic markets should be capable, therefore, of catering to the world market. The resulting dissociation between sites of production and markets implies that a firm would now choose to invest in a particular location only if that site can serve as one of the production locations for its world market operations. Thus, when an enterprise chooses a new location for investment it is in essence `relocating' capacity that can service the local market, or its erstwhile `home'-country market, or its third country markets or some combination of those markets. This trend is seen as having been facilitated by changes in technology. To start with, the revolution in transport and communications has reduced costs and increased the ease of communication so substantially that: i) managerial control of internationally dispersed capacity has become easier; and ii) the share of transport costs (for inputs purchased and outputs sold) in total costs has fallen dramatically in the case of many commodities. Secondly, changes in technology have in many industries segmented production processes, so that individual components of the process can be undertaken independent of each other at diverse sites. This permits firms to relocate particular (say, labour-intensive) segments of even technology-intensive production processes to alternative sites depending on their characteristics. Add to all this the fact that over the last two decades or more there has been a rapid dismantling of protective regimes and relaxation of regulations on foreign investors across the globe, and the basis for a significant change in the character of foreign investment should be clear. After allowing for national peculiarities and variations in political structures, any production site worldwide is becoming a potential site for production for world markets. The individual firm is detached from dependence on home country resources and has the opportunity to locate itself in environments where it can overcome the disadvantages stemming from specific macroeconomic developments such as appreciating exchange rates or microeconomic features such as high wage levels, and substantially enhance its international competitiveness. The BEA evidence, however, does not tally with this view that the expansion of multinationals would be accompanied by a tendency where growth in the periphery would be at the expense of presence in the metropolis. As already mentioned, the acceleration in expansion of MNC operations has been as true of the parent firms as of their MOFAs. This is true of both gross product and of employment. Further, there has been only one short span of time, 1989-94, when employment growth in US MNC parents was marginally negative. And these were years when growth in the US and worldwide had slowed, suggesting that employment movements during those years were influenced more by macroeconomic trends than by firm-level strategies. Overall, the presence of parents in the global operations of MNCs still remains strong. As Chart 3 shows, the share of MNC parents in the worldwide gross product of US MNCs has remained more or less constant during the years 1989-99. This is true in the case of all industries in which MNCs participated, and of manufacturing, finance, insurance and real estate (excluding depository institutions). It is only in services that there has been a significant decline in parent share of gross product during the latter half of the 1990s. This persisting presence of parents in the economic activity of US MNCs, has also meant that the contribution of US parent firms to US GDP has also remained relatively constant during the period 1989-99 (Chart 4). Whether relocation occurred or not, the parents of US multinationals were making a significant and persisting contribution to overall US economic activity. However, there are two confusing aspects to these trends. Another way of assessing the impact of MNC operations is to examine the contribution of MNC parents and affiliates to parent and host country GDP. Chart 4, which examines the contribution of US parents to US gross product over a longer period of time suggests that there was a sharp reduction (6 percentage points in `all industries' and 8 percentage points in the case of manufacturing) in the contribution of US parents to US MNCs GDP between 1982 and 1989, followed by stability in that contribution thereafter. Thus, if at all the relocation argument is supported by the gross product figures, it appears to be during the early years of globalisation. It could of course be argued that this exercise, which requires comparison of figures from the BEA's benchmark surveys of US direct foreign investment abroad and its National Income and Product Accounts, could be fraught with problems. However, the BEA itself has in the past attempted to adjust aggregate gross product figures to make them more comparable with the FDI figures. For improved comparability with US-parent gross product, GDP of all private US businesses was adjusted to remove from the total categories not applicable to non-bank US parents - specifically, GDP of depository institutions; imputed rental income of owner-occupied farm and non-farm housing; and rental income of persons. The results yielded by this comparison of adjusted figures provided in Chart 5 tallies with the view supported by Chart 4 that any loss of production due to relocation occurred, if at all, during the early years of globalisation and not during its peak years in the 1990s. But the litmus test of relocation lies not in output but in employment trends, which provide a second cause for confusion. As Chart 6 show, the share of US parents in MNC employment worldwide remained more or less constant during 1982-89, but fell quite significantly (5 percentage points in `all industries' and manufacturing) during 1989-99. This could be interpreted to mean: i) that relocation by MNCs resulted not so much in a major loss of employment in the US, as in a faster growth of MNC employment in the periphery than in the core; and ii) that this process of relocation did not result in any loss in share of parents of US multinationals in gross product generated worldwide or in the contribution of these parents to US GDP. According to the BEAs own interpretation of trends based on its 1994 benchmark survey, the persistence, despite relocation of production, of a high share in manufacturing GDP of US parents partly reflects "the firm-specific intangible assets (such as patents or brand images) that allow these firms to earn profits that are sufficient to overcome the additional costs of producing in foreign markets." This refers, of course, only to that part of surplus that is repatriated in some form to the parent country, and not to that which is used either to expand international assets or to increase the retained surpluses of the affiliates themselves. Clearly, relocation does not imply lower profits for US companies, whatever else it may mean for the US economy. Rather, as has been argued by many, international expansion has been a way of maintaining or enhancing the surpluses garnered by US parents from the world market. The value of that surplus also increases in real terms because of the fall in primary commodity prices associated with globalisation. An examination of the industrial distribution of US multinationals in 1999 (Table 1) helps clarify matters further. There are just two major industry groups which account for more than 10 per cent of MNC gross product worldwide: manufacturing and information. Within manufacturing, transportation equipment, chemicals and computers and electronic products are the main sectors of MNC presence. An interesting feature of MNC presence in manufacturing is that while parent manufacturing firms account for 50 per cent of aggregate gross product of MNC parents, MOFAs in manufacturing account for 72 per cent of aggregate gross product of US majority-owned affiliates. This larger share of MOFAs when compared with parents in manufacturing as opposed to all industries does suggest that MNC manufacturing presence relative to overall presence is greater abroad than at home. Information is a new category in the 1999 survey, which did not exist in earlier surveys. Its large share in aggregate gross product is true only of parent firms (13.3 per cent), while the share of MOFAs in this area in aggregate gross product is still small (3.6 per cent). The sub-sector is dominated by one industry, broadcasting and telecommunications. This is an area where "intangible assets", and the control they provide, may play a major role in explaining the high share of parent MNCs in gross product. That possibility is corroborated by the relative importance of MNCs in the US industry in this sector when assessed in terms of employment (Table 2). While US parents of MNCs account for less than 20 per cent of all non-bank private employment, and parents of manufacturing MNCs for 45 per cent of US manufacturing employment, the figure stands at 53 per cent in the case of the information industry. This has an important implication. Even as the share of manufacturing employment generated by US MNCs abroad is showing signs of growing faster than employment at home, new sectors of MNC activity in the services areas, particularly information, are helping strengthen employment in the US. This not only tallies with the growing role of services in US employment and GDP, but also explains in part the growing emphasis on liberalisation of services trade on the part of US trade negotiators at the WTO. MNC trade does account for a substantial share of US trade. MNC associated US exports accounted for 63 per cent of total US exports in 1999, having fallen from 77 per cent in 1982 (Table 4). A substantial chunk of those exports were to MNC affiliates abroad. Thus intra-firm trade accounted for 25 per cent of all exports. If the surveyed MNCs account for all of these intra-firm exports, then it follows that close to 40 per cent of MNC exports from the US is intra-firm. Intangibles embodied in these goods and those sold to other persons account for a substantial share of parent gross product. What is surprising is that MNCs have a much smaller role in US imports than in US exports. The share of MNC associated US imports has fallen from 50 per cent in 1982 to 37 per cent in 1999, and only 17 per cent of US imports are intra-firm. Thus the view that American firms are increasingly relocating abroad to cater to US markets appears to be far from the truth. US imports come from other sources. US multinational parents and MOFAs are still predominantly targeting local markets, and if at all US MNCs are targeting foreign markets based on US production rather than US markets based on global production. This, however, does not mean that MNC presence is not an important factor affecting developing countries. This is the view often gleaned from the fact that even now a large share of global FDI flows are to the developed countries. But developing countries are the ones in which US MNCs account for a significant share of host GDP (Table 3). In fact, 11 out of the top 20 countries ranked according to MNC share in host GDP are developing countries, including Singapore, Malaysia, Hong Kong, Indonesia, Chile, Mexico and the Philippines, which are known to have followed strategies aimed at attracting FDI. There are many more developing countries in the list of the top 40 in terms of MNC contribution to host GDP. Even this evidence should be treated with caution, since it does not include joint ventures in which US MNCs have a minority share. What needs to be noted is that the expansion of US capital abroad has increasingly taken the form of acquisitions of existing firms as opposed to investment in green-field projects. Such acquisitions, which are followed by the modernisation or even replacement of the acquired firms' assets, allows for US firms to capture market shares in which pre-existing brands are replaced by those of the acquiring firm. In 1999, for example, 577 of 1077 newly established affiliates of US MNCs were acquired rather than newly established. The consequent standardisation of brands sold worldwide has been widely noted. Underlying such standardisation is the growing command of US firms of the gross product of host countries. Combined with the emergence of new industries like the information sector where US MNCs dominate, this could mean that the next benchmark survey could reflect a qualitative shift in MNC presence in the developing world. But the evidence as of now indicates that that presence would not be so much a sign of relocation of US economic activity to low cost sites abroad, but the growing dominance of these MNCs over world markets through exports and local production. Copyright © 2002, The Hindu Business Line. Republication or redissemination of the contents of this screen are expressly prohibited without the written consent of The Hindu Business Line
This archive was generated by hypermail 2b30 : Sun Jun 02 2002 - 00:00:07 EDT