Foreign Direct Investment Changes Ownership, not Location

Overall, inward and outward FDI stocks and flows tend to coincide; nations that invest abroad extensively are usually major recipients of FDI.

In Interpreting Developed Countries' Foreign Direct Investment (NBER Working Paper No. 7810), NBER Research Associate Robert Lipsey suggests that flows of foreign direct investment (FDI) among developed countries, where most FDI occurs, have little to do with the location of production. To a large extent, they are changes in ownership of specific productive assets, presumably from less efficient to more efficient owners and managers. There may be no change in the geographical location of aggregate production or production in a particular industry.

An example of such shifts in ownership with little change in the location of production took place in the U.S. manufacturing sector. The share of U.S. parent firms in domestic manufacturing output fell from 65 to 55 percent between 1977 and 1997, while the share of their overseas affiliates in their firms' output grew substantially. That might appear to be a change in the location of production attributable to FDI. However, the share of foreign-owned manufacturing affiliates in U.S. manufacturing production rose by 9 percentage points at the same time, almost completely offsetting the decline in the U.S. parent share. Thus, in terms of total manufacturing production, the outward and inward FDI movements practically cancelled each other out.

Lipsey finds that inward and outward FDI stocks and flows tend to coexist in a two-way and often offsetting manner across countries and in individual years for individual countries. That is especially true for large countries that are open to trade and investment. Overall, inward and outward FDI stocks and flows tend to coincide; nations that invest abroad extensively are usually major recipients of FDI.

Analyzing data for individual countries from 1970 through 1995, Lipsey finds that the relationship between inflows and outflows of FDI relative to total output is positive and significant in most of the 19 countries. That is evidence that economic conditions influence the turnover of assets as much as, or more than, the net flow of capital.

Neither inflows nor outflows of FDI are crucial to determining the level of capital formation in a given country, though. Lipsey's data show that even gross FDI inflows have been small relative to gross fixed capital formation. In most countries, gross inflows of FDI averaged 5 percent or less of capital formation and net inflows were between minus 5 and plus 5 percent of capital formation.

He argues that if the major role of FDI is to shift the location of production, then we should observe flows from industries of home-country comparative disadvantage, or declining comparative advantage, to locations with comparative advantages in those industries. To the extent that FDI flows reflect the technological advantages of firms in the source country, outflows should take place in the industries of that country's past or present comparative advantages. However, the FDI does not necessarily go to countries with comparative advantages in those industries. Inflows of FDI based on technology differences should come to industries of host-country comparative disadvantage.

In the case of the United States, U.S. affiliate production abroad (resulting from outward FDI) was particularly large in some, but not all, of the industries of major U.S. export comparative advantage. Industries of comparative U.S. disadvantage, such as textiles and apparel or iron and steel, showed relatively large inward production by foreign-owned affiliates in the United States (inward FDI production). That is what might be expected if foreign firms were technologically ahead of U.S. firms in those industries, and the foreign-owned production reflected a change in the ownership of U.S. production rather than a movement of production to the United States.

Examining changes over time in U.S.-owned production abroad, Lipsey finds two distinct patterns. In foods and metals, U.S. outward FDI production in recent years moved toward countries with relative comparative advantages in each industry group, suggesting the influence of country factor endowments and shifts in the location of production. In nonelectrical machinery, however, indications are that U.S. affiliate production moved toward countries with comparative disadvantages in the industry. That movement suggests the influence of U.S. firms' technological advantages.

If FDI transfers assets and production from less efficient to more efficient owners and managers, Lipsey theorizes, then FDI can be viewed in recipient countries as freeing capital frozen in industries that owners (including governments) would like to leave. It permits the owners to use their capital in what they consider more appropriate ways, at home or abroad. In investing countries, outward FDI permits the country's firms to better exploit their skills and technological advantages.