NBER Reporter: Winter 2000/2001
Robert E. Lipsey *
Foreign direct investment (FDI) flows and the operations of multinational firms have attracted increased research attention in recent years. This is partly because FDI has grown in importance as a form of capital flow and partly because FDI seems a more reliable form of finance for developing countries than portfolio investment or short-term lending in light of the recent Asian experience. Perhaps the most important reason is the emergence of a popular view that multinational firms control much of the world's economy.
In fact, the share of direct investment in the world's capital outflows has grown substantially since the early 1970s to reach about 25 percent in the early 1990s.(1) That share dropped in 1996 and again in 1997, with a surge in portfolio capital and short-term lending. But as portfolio and short-term capital flows declined after the start of the Asian crisis, FDI flows rebounded to 30 percent of the total world capital flows.
Despite this growth in FDI flows, the resulting production is still a small part of the world's total output: about 7 or 8 percent in 1995, compared with about 4.5 percent in 1970.(2) The petroleum sector was the most internationalized in 1970, but the internationalized share fell sharply afterwards, especially in developing countries where important operations were nationalized. Manufacturing is now the most internationalized sector, at over 16 percent of output, but apart from these two sectors, internationalized production remains under 4 percent of the world total.
In the past, FDI flows to individual countries were less volatile than other international capital flows: they changed direction less frequently and the range of fluctuations around their mean was smaller. That characteristic of FDI flows was demonstrated in the Latin American crises of the early 1980s. It was confirmed in the Mexican crisis of 1994, when direct investment inflows quickly regained their previous level, while other forms of capital inflow remained far below their peaks. And the pattern was further confirmed in the Asian crises of 1997, when direct investment inflows into developing Asia as a whole hardly paused in their rapid growth, while portfolio and other forms of investment dried up or turned negative on net balance.(3)
Over a longer horizon, the economies of East Asian countries have been transformed, and FDI has played an important role in most of the transformations. The industry structure of production and of exports has changed drastically. In 1977 almost two-thirds of East Asian manufactured exports were in foods, textiles and apparel, and miscellaneous manufactures, but in 1995 this percentage had dropped to one-third.(4) Over the same period, the exports of East Asian machinery industries grew from 17 to 44 percent of the total.
Much of that shift in export composition was propelled by direct investment in these countries, mainly from Japan and the United States. Foreign firms supplied technology and links to other parts of their production and trade networks. These links, added to the local resources, fueled rapid export growth and changes in export composition. Frequently, affiliates were established mainly for export production, but over time their output shifted toward production for local markets. That change was accompanied by growth in production by nonaffiliated host-country firms in the same or related industries.
The developing countries are almost all net recipients of direct investment. It is clear that foreign firms bring not only scarce capital, but also superior technologies and new industries to these countries. They sometimes also bring about drastic changes in the variety of industries and the composition of production.
Among developed countries, though, FDI flows seem to play a different role. While they are much larger than flows to developing countries, they typically do not radically change the composition of host-country production. Inflows often are matched by outflows in the same country in the same period.(5) The flows among the developed countries mainly seem to reshuffle the ownership of productive assets, moving them to owners who want them more than their current owners and who are willing to pay the most for them. Presumably, capital flows move assets from less efficient to more efficient owners, or from owners who are technologically or commercially backward in their industries to firms that are technological leaders. Or, capital flows may bring relief to owners who have found more attractive uses for their capital than their current industries. In none of these cases do such flows necessarily change the location of the production, assets, or employment of these industries, though.
One possible indication of the advantages of multinationals relative to other firms in their host countries is that, in every host country, multinationals pay higher wages than their locally owned counterparts. That is the case even in the United States, often thought of as the world's technological leader. Much of the foreign-firm wage premium in the United States is related to the larger size of foreign-owned establishments and to their choice of industries and locations. A study based on the recently published matched data from the Census Bureau and the Bureau of Economic Analysis found that all of the foreign establishment premium in manufacturing, but not in other industry groups, could be explained by these characteristics.(6) A larger foreign presence in an industry in a state had no effect on average wages in manufacturing once establishment size, industry, and location were taken into account, but it did raise average nonmanufacturing wages, and even wages in domestically-owned nonmanufacturing establishments.
Spillovers from foreign-owned to domestically-owned establishments are one possible source of host-country gains from inward FDI beyond the gains from higher productivity of the foreign-owned establishments themselves. One study using individual establishment data for Indonesian manufacturing(7) found higher productivity not only in foreign-owned establishments but also in domestically-owned establishments as the foreign participation in the industry increased. The extent of such spillovers was not affected by the degree of foreign ownership of individual plants.
Even if outward direct investment does not necessarily involve a reduction in a home country's production, or in home-country production by a country's multinationals, it may involve a reallocation of production or of certain aspects of production. These could affect home-country labor markets in various ways depending on the nature of the reallocation. In U.S. multinationals, there has not been any aggregate shift of production or employment to foreign locations, even in relative terms, since the late 1970s, except in manufacturing. There was some reduction in U.S. multinationals' share of manufacturing production and employment in the United States. However, that reduction was offset almost entirely by an increase in the shares of foreign multinationals, leaving the total production and employment share of multinationals virtually unchanged.(8)
In U.S. multinationals, a higher level of production in developing countries is associated with lower parent employment at home per unit of parent home-production. This implies that U.S. multinationals have allocated the more labor-intensive parts of their production to developing countries and retained their more capital-intensive and skill-intensive operations at home. There is some weak evidence that higher levels of foreign production also are associated with higher wages at home, but the higher wages do not seem to be related in particular to production in developing countries.
A comparison of home labor market effects in Swedish and U.S. multinationals(9) found that, in contrast to U.S. firms, Swedish multinationals with higher shares of employment abroad had higher employment, particularly blue-collar employment, per unit of output in the parent company at home. That suggests that Swedish firms allocate the more capital-intensive production to their foreign affiliates, most of which are located in advanced countries. A three-way comparison among U.S., Swedish, and Japanese multinationals found that Japanese parent firms behave more like Swedish than U.S. parent firms.(10) A Japanese parent's employment at home, given its level of home production, was found to be higher as its production abroad by the firm's foreign affiliates increased. Since Japanese firms probably would not be allocating labor-intensive processes to their high-wage home operations, their need for additional employment at home for supervision and other ancillary activities likely outweighs any allocation of labor-intensive operations to developing countries.
Japan is a smaller participant in outward direct investment than the United States, relative to its size, and is much smaller as a host to inward FDI.(11) However, Japanese outward FDI did help Japanese firms maintain market shares in some declining industries by shifting production to lower cost locations. Since the 1980s, Japan has been catching up, and the industry pattern of Japanese FDI has come to resemble that of the United States, especially in the importance of electrical machinery industries. Deregulation or liberalization in service sectors also may raise the level of inward FDI production, which is still unusually low among OECD countries.
Future work in this area will attempt to draw stronger inferences about the effects of FDI on host countries by using panel data for individual firms and establishments. Work is beginning on such studies for foreign investment in the United States, concentrating on takeovers of U.S.-owned firms by foreigners, and on foreign investment in Indonesia, with a focus on labor market issues.
1. R. E. Lipsey, "The Role of Direct Investment in International Capital Flows," in International Capital Flows, M. Feldstein, ed. Chicago: University of Chicago Press, 1999.
4. R. E. Lipsey, "Affiliates of U.S. and Japanese Multinationals in East Asian Production and Trade," in The Role of Foreign Direct Investment in East Asian Economic Development, EASE Volume 9, T. Ito and A. O. Krueger, eds. Chicago: University of Chicago Press, 2000.
7. M. Blomström and F. Sjöholm, "Technology Transfer and Spillovers: Does Local Participation with Multinationals Matter?," NBER Working Paper No. 6816, and European Economic Review, 43 (4-6) (April 1999), pp. 915-23.
8. R. E. Lipsey, "Foreign Production by U.S. Firms and Parent Firm Employment," NBER Working Paper No. 7357, September 1999; forthcoming in Multinational Firms and Impacts on Employment, Trade, and Technology, R. E. Lipsey and J. Mucchielli, eds. Harwood Academic Publishers.
9. M. Blomström, G. Fors, and R. E. Lipsey, "Foreign Direct Investment and Employment: Home Country Experience in the United States and Sweden," Economic Journal, 107 (445) (November 1997), pp. 1787-97.
11. M. Blomström, D. E. Konan, and R. E. Lipsey, "FDI in the Restructuring of the Japanese Economy," in Japan's New Economy: Continuity and Change in the 21st Century, M. Blomström, B. Gangnes, and S. La Croix, eds., forthcoming from Oxford University Press.
* Lipsey is a Research Associate in the NBER's Programs on International Finance and Macroeconomics, International Trade and Investment, Productivity, and Professor of Economics, Emeritus, at The City University of New York.