From the July 1999 SURVEY OF CURRENT
BUSINESSThe combined domestic and foreign operations of nonbank U.S. multinational companies (MNC's) continued to grow at a strong pace in 1997, according to preliminary estimates from the annual survey of U.S. direct investment abroad conducted by the Bureau of Economic Analysis (BEA)./1/ Current-dollar estimates of the worldwide gross product of U.S. MNC's (U.S. parents and majority-owned foreign affiliates (MOFA's) combined) increased 5.6 percent in 1997, compared with an 8.1-percent increase in 1996 and an average annual increase of 5.5 percent from 1989 to 1996 (table 1).
Two other key measures of MNC operations grew much faster in 1997 than in 1989–96. MNC employment increased 6.1 percent, compared with a 1.5-percent increase in 1996 and negligible growth in 1989–96. Capital expenditures increased 13.7 percent, compared with a 5.2-percent increase in 1996 and an average annual increase of 3.9 percent in 1989–96.
The strong growth in MNC operations in 1997 reflected continued economic growth and a strong upturn in new investments. In 1997, real gross domestic product (GDP) increased 3.9 percent in the United States, up from a 3.4-percent increase in 1996; in major host countries, real GDP increased 3.2 percent, on average, up from a 2.4-percent increase in 1996./2/ These favorable economic conditions resulted in the expansion of existing MNC operations and the growth of MNC operations through acquiring or establishing other businesses, both in the United States and abroad. The upturn in new investment partly reflected the growing availability of investment opportunities abroad as a result of legal and economic changes—such as the privatization of electric utilities and telephone companies and the liberalization of direct investment policies in foreign host countries.
Additional highlights of U.S.-MNC operations in 1997 follow:
Revisions to the 1996 estimates.—The estimates of U.S.-MNC operations for 1996 were revised to incorporate the final results of the 1996 Annual Survey of U.S. Direct Investment Abroad./3/ The year-to-year percent changes for the three key measures were revised less than 1 percentage point from the changes shown in the preliminary estimates: The increase in gross product was revised up 0.8 percentage point to 8.1 percent; the increase in employment was revised down 0.2 percentage point to 1.6 percent; and the increase in capital expenditures was revised up 0.2 percentage point to 5.3 percent.
Organization of the article.—The first part of this article analyzes the worldwide operations of U.S. MNC's; the second part analyzes their domestic (U.S.-parent) operations; and the third part analyzes their foreign (foreign-affiliate) operations.
This section examines worldwide U.S.-MNC operations and compares the domestic and foreign aspects of these operations./4/
Gross product of all U.S. MNC's grew 5.6 percent in 1997, to $2,090 billion; the U.S.-parent component of U.S.-MNC gross product grew 6.1 percent, and the MOFA component grew 4.2 percent. Available evidence suggests that for both parents and MOFA's, these increases reflected increases in real gross product to a greater extent than they reflected increases in prices. The growth of U.S. parents' gross product, at 6.1 percent, substantially exceeded the 1.4-percent rate of U.S. price inflation (as measured by the GDP implicit price deflator for all private U.S. businesses, except depository institutions). Despite the dampening effect of the significant appreciation of the U.S. dollar, the gross product of MOFA's grew 4.2 percent in dollar value, exceeding the 2.8-percent average rate of price inflation in the host countries./5/
Worldwide production and the productive resources of U.S. MNC's remained concentrated in the United States: In 1997, U.S. parents accounted for about three-fourths of MNC gross product, capital expenditures, and employment and for over two-thirds of MNC profit-type return. From 1989 to 1997, however, the distribution of gross product and employment shifted slightly from the United States to abroad: The MOFA share of worldwide MNC gross product edged up from 23 percent to 25 percent, and the MOFA share of MNC employment rose from 21 percent to 25 percent (table 2). In contrast, the MOFA share of worldwide MNC profit-type return fell from 34 percent to 31 percent, probably in response to cyclical factors here and abroad that were relatively less favorable to MOFA's in 1997 than in 1989.
By industry, the most significant shift towards foreign operations was in manufacturing: The MOFA share of MNC gross product rose from 26 percent in 1989 to 29 percent in 1997; the MOFA share of MNC capital expenditures rose from 26 percent to 30 percent; and the MOFA share of MNC employment rose from 27 percent to 33 percent.
Production abroad by U.S. MNC's may have been stimulated by new market opportunities and improved business conditions abroad, such as the further integration of the European Union, the end of hyperinflation in Brazil and Argentina, and the economic liberalizations in Eastern Europe.
The output of U.S. MNC's (sales to unaffiliated customers plus inventory change) consists of both the gross product that originates within the MNC's themselves and the gross product that originates elsewhere and is incorporated in the intermediate inputs purchased by MNC's from unaffiliated suppliers. The gross product originating in U.S. MNC's consists of the gross product of both the U.S. parents and their foreign affiliates.
The origins of U.S.-MNC output in 1997 were essentially unchanged from 1989: The share of the output of U.S. MNC's that originated within the MNC's edged down from 36 percent to 34 percent (table 3, column 8). Underlying this stability were offsetting changes by industry. For example, in industrial machinery and equipment, the share of output originating within MNC's decreased from 46 percent to 36 percent.
The share of MNC output that was accounted for by U.S.-parent gross product edged down from 28 percent to 26 percent. The share accounted for by MOFA gross product edged up from 8 percent to 9 percent.
In 1997, U.S.-MNC-associated trade—U.S. trade involving U.S. parents, their foreign affiliates, or both—accounted for 63 percent of all U.S. exports of goods and for 40 percent of all U.S. imports of goods (table 4 and chart 2)./6/ U.S. MNC's accounted for a larger share of exports than of imports, partly reflecting the intrafirm sourcing patterns of MNC's; parents tend to be a much more important source of supply to their affiliates than the affiliates are to their parents.
Of the $434 billion in U.S.-MNC-associated exports in 1997, 42 percent represented trade between U.S. parents and their foreign affiliates—intra-MNC trade—and 58 percent represented U.S.-MNC trade with others. Data on trade by broad product category from the most recent benchmark survey of U.S. direct investment abroad, covering 1994, indicate that most of the intra-MNC exports consist of machinery (57 percent of total intra-MNC exports in 1994), chemicals (13 percent), and road vehicles and parts (12 percent)./7/ Of the $251 billion in MNC trade with others, 87 percent represented exports shipped by U.S. parents to foreigners other than their foreign affiliates, and 13 percent represented exports shipped to foreign affiliates by U.S. persons other than their U.S. parents. Data from the 1994 benchmark survey indicate that U.S. parents' exports to foreigners other than their foreign affiliates consist mainly of machinery (27 percent in 1994), "other" transportation equipment, mainly aircraft (14 percent), and chemicals (12 percent) and that exports shipped to foreign affiliates by U.S. persons other than U.S. parents consist mainly of machinery (53 percent).
Of the $350 billion in U.S.-MNC-associated imports of goods in 1996, 42 percent represented intra-U.S.-MNC trade, and 58 percent represented U.S.-MNC trade with others. Data from the 1994 benchmark survey indicate that most of the intra-MNC imports consist of road vehicles and parts (38 percent in 1994) and machinery (37 percent). Of the $203 billion in MNC trade with others, 85 percent represented imports shipped to U.S. parents by foreigners other than their foreign affiliates, and 16 percent represented imports shipped by foreign affiliates to U.S. persons other than their U.S. parents. Data from the 1994 benchmark survey indicate that U.S. parents' imports from foreigners other than their foreign affiliates consist mainly of machinery (35 percent in 1994), petroleum and coal products (15 percent) and "other" manufactures (15 percent) and that U.S. imports shipped by foreign affiliates to unaffiliated U.S. persons consist mainly of petroleum (27 percent), other manufactures (19 percent), and machinery (18 percent).
This section examines the following selected aspects of the domestic (U.S.-parent) operations of U.S. MNC's: The change in U.S.-parent gross product by industry and by source of change in 1996–97; the U.S.-parent share of private GDP in 1989 and 1997; and the origin of U.S.-parent output in 1989 and 1997.
The gross product of all U.S. parents increased 6.1 percent in 1997, to $1,570 billion, compared with an average annual increase of 5.1 percent in 1989–96 (table 5).
By industry.—In 1997, U.S.-parent gross product increased most rapidly in wholesale trade (23.6 percent), in "other" industries (15.4 percent), in finance (except depository institutions), insurance, and real estate (14.0 percent), and in services (12.0 percent)./8/ U.S.-parent gross product decreased in manufacturing industries; the decreases were particularly sharp in food and in industrial machinery.
The increase in wholesale trade primarily reflected increased sales to meet strong demand. This increase and the decrease in manufacturing also reflected the reclassification of some U.S. parent companies from manufacturing to wholesale trade./9/ The increase in "other" industries was concentrated in electric, gas, and sanitary services and in retail trade, mainly reflecting the entry of U.S. companies that acquired or established their first foreign affiliate in 1997. The increase in finance (except depository institutions), insurance, and real estate primarily reflected increased revenues as a result of favorable financial market conditions. The increase in services mainly reflected the entry of new parent companies, but it also reflected increased sales to meet strong demand for new products, such as enhanced computer software.
By source of change.—In 1997, the gross product of U.S. parents increased $89.9 billion; $53.4 billion of this increase was attributable to changes in the operations of U.S. companies that were parents in both 1996 and 1997 (table 6). Even so, the $22.3 billion that was attributable to the entry of new parent companies was much stronger in 1997 than in 1995 or 1996 and largely reflected the entry of new parents in electric, gas, and sanitary services and in retail trade.
The gross product of U.S. parents accounted for 26 percent of the gross product of all private U.S. businesses in 1997, the same as in 1989 and in 1996 (table 7)./10/ The U.S.-parent share was highest in manufacturing, partly reflecting the firm-specific advantages (such as superior production or marketing techniques) of U.S. manufacturers that allow them to overcome the additional costs entailed in producing in foreign markets.
The U.S.-parent share in services remained low. U.S. direct investment in some service industries may be inhibited by the structure of the industries in some host countries. For example, U.S. direct investment in health care services is constrained or precluded in countries where the government plays a prominent role in the delivery of health care. In addition, some service industries that are characterized by small-scale production (such as dry cleaners and hair stylists) may lack the firm-specific advantages that often provide the basis for direct investment in other industries.
Although the overall U.S.-parent share of the gross product of private U.S. businesses was stable from 1989 to 1997, there were offsetting changes among industries: The U.S.-parent share in manufacturing decreased from 63 percent to 60 percent; the share in services increased from 6 percent to 8 percent; and the share in all other industries combined increased from 16 percent to 19 percent.
The increase in services partly reflects the entry of new U.S. parent companies. These additions have been widespread across service industries, but they have been pronounced in computer and data processing services, motion pictures, and management and public relations services.
The output of U.S. parents consists of both the gross product that originates within the parents themselves and the gross product that originates elsewhere and is incorporated in the intermediate inputs purchased by parents from foreign affiliates and from unaffiliated suppliers.
In 1997, gross product originating in U.S. parents accounted for 32 percent of parent output, compared with 33 percent in 1989 (table 8, column 11). The industries with the highest shares were services, "other" industries (mainly communication and electric and gas utilities), and "other" manufacturing (mainly instruments and related products). The share in services is high partly because the production of services tends to be labor-intensive and is not as easily outsourced as the production of goods. The high share in communications and in electric and gas utilities partly reflects industry regulations that have historically granted exclusive production and distribution rights to a few large companies. The high share in instruments and related products may partly reflect the proprietary knowledge that is embodied in these products and that firms may protect by internalizing production.
In manufacturing, the origin of U.S.-parent output shifted away from internal production, as the share of U.S.-parent output that was accounted for by their own gross product decreased from 38 percent to 34 percent; the shift was pronounced for parents in industrial machinery, particularly in computers and components, and in electric equipment, particularly in consumer electronics. In these industries, the movement to unaffiliated suppliers was partly in response to increased global competition; to improve their competitiveness, parents specialized in areas in which they had an advantage and relied on unaffiliated suppliers for other inputs or finished products.
In petroleum, the origin of U.S.-parent output shifted slightly toward internal production. The share of U.S.-parent output that was accounted for by the parents' gross product edged up from 28 percent to 30 percent.
The share of U.S. parents' output that was accounted for by local (U.S.) content—U.S.-parent gross product and purchases from U.S. suppliers—was 93 percent in 1997, compared with 94 percent in 1989. The industries with the highest U.S. content were services (100 percent) and food and kindred products (98 percent)./11/ The high share in services partly reflects the tendency towards internal production. The share in food and kindred products is high partly because the United States is a major agricultural producer, so most of the inputs are available locally at competitive prices.
In wholesale trade, the local-content share of the output of U.S. parents edged up from 85 percent to 87 parent. (In this industry, local content consists primarily of purchases from domestic sources.) In manufacturing, the share edged down from 93 percent to 91 percent. In petroleum, it edged down from 92 percent to 90 percent.
This section examines selected aspects of the foreign (foreign-affiliate) operations of U.S. MNC's. First, the 1996–97 change in employment by all affiliates is examined, and the patterns of newly acquired or established affiliates in 1997 are presented. The remainder of the section focuses on selected aspects of the operations of majority-owned foreign affiliates (MOFA's).
The broadest measures of the foreign operations of U.S. MNC's cover all affiliates, regardless of the degree of U.S. ownership. The discussion of all-affiliate operations uses data on employment because estimates of gross product are available only for MOFA's.
The total employment of nonbank foreign affiliates increased 6.3 percent to 8.0 million in 1997, compared with an average annual increase of 1.9 percent in 1989–96 (table 9). The unusually large increase reflected both new investments and the expansion of existing operations in industries—such as personnel supply services, computer component assembly, and telephone utilities—that typically employ many workers.
The increase was widespread by geographic area. By industry, most of the increase was accounted for by affiliates in manufacturing and in services.
Newly acquired or established affiliates.—In 1997, 489 new affiliates were acquired or established by U.S. MNC's, up significantly from 294 in 1995 and 413 in 1996 (table 10). These affiliates had a combined employment of 228,000, up from 138,000 in 1995 and 150,000 in 1996. The rapid increase in new investments in 1995–97 partly reflected the new investment opportunities and the improved business climates created by economic and legal changes, such as the privatization of electric utilities in the United Kingdom and Australia, the liberalization of foreign direct investment policies, the end of hyperinflation in Brazil and Argentina, and the reinstatement of the U.S. foreign tax credit on investment in South Africa.
In 1997, new affiliates continued to be primarily located in high-wage countries. These affiliates accounted for 71 percent of all new affiliates and for 59 percent of their employment. This large share suggests that U.S. direct investment abroad tends to be attracted more by access to large and prosperous markets than by access to low-wage labor. The United Kingdom, Germany, and France were among the high-wage countries that attracted significant amounts of new investment in 1997.
Low-wage countries have nevertheless been attracting a rising proportion of the new investments. Their share of new investments rose steadily from 18 percent in 1989 to 30 percent in 1996 before slipping to 29 percent in 1997. South Africa, Brazil, and China were among the low-wage countries that attracted relatively large amounts of new investment in 1997.
Manufacturing continued to be the primary industry for new investments. In 1997, it accounted for 35 percent of all new affiliates and for 41 percent of their employment. "Other industries" also accounted for some large new investments; for example, some U.S. providers of electric power and telecommunications services acquired foreign affiliates as a result of host-country privatizations.
In 1997, majority-owned foreign affiliates (MOFA's) accounted for 90 percent of all foreign affiliates. The MOFA share of the employment of all affiliates was 81 percent, up from 77 percent in 1989 (table 1). These high percentages are consistent with the "internalization" theory of the origins of MNC's, which suggests that MNC's tend to have firm-specific advantages that must be preserved by strict control of operations./12/
In all but 20 host countries, more than three-fourths of all affiliates are majority owned. Saudi Arabia, at 43 percent, and Israel, at 56 percent, are among the countries that had a relatively low percentage of MOFA's in 1997. In some countries, laws constrain, or have constrained, the level of foreign ownership of domestic businesses or have assessed lower taxes on, or provided other benefits to, businesses that have majority local ownership.
In India, the percentage of MOFA's was relatively low, at 58 percent, in 1997, but it was up substantially from 22 percent in 1989. The rapid rise in the MOFA share partly reflected the Indian Government's New Industrial Policy, adopted in 1991, which significantly liberalized the country's foreign direct investment policy. One change raised the maximum permitted foreign ownership of newly acquired or established Indian businesses to 51 percent in most sectors. Despite these changes, the share of the Indian economy accounted for by production by MOFA's remained relatively low in 1997. (For details, see the section "MOFA share of host-country GDP.")
Changes in gross product.—The gross product of MOFA's increased 4.2 percent in 1997, to $519.3 billion, compared with an average annual increase of 6.5 percent in 1989–96 (table 11). The slowdown was primarily accounted for by affiliates in Europe and in Asia and Pacific, and it mainly reflected a substantial decline in the U.S.-dollar price of foreign currencies. In contrast, MOFA gross product in Latin America and Other Western Hemisphere, Canada, and Africa grew faster in 1997 than in 1989–96. The increase in Latin America and Other Western Hemisphere was strong enough to overcome the dampening effect of the significant appreciation of the U.S. dollar against the two primary currencies in the region—the Brazilian real and the Mexican peso; the increase was widespread across industries, reflecting robust economic conditions in the host countries and the addition of newly acquired or established MOFA's. In Canada, the increase was concentrated in transportation equipment manufacturing and largely reflected increased production to meet strong demand in the United States. In Africa, the increase was concentrated in manufacturing and mainly reflected newly established affiliates in South Africa. In Eastern Europe, affiliate operations continued to grow rapidly, but they remained quite small; MOFA gross product in this area increased 26.7 percent, to $3.9 billion, but the level was only 1 percent as large as that in Western Europe.
By industry, affiliates in services and manufacturing accounted for most of the increase in MOFA gross product.
In 1997, $14.3 billion of the $21.0 billion increase in MOFA gross product was attributable to changes in existing operations (table 12). The entry of new MOFA's contributed $7.0 billion, up significantly from 1996.
MOFA share of host-country GDP.—In 1997, the gross product of MOFA's accounted for 7 percent or more of the gross domestic product (GDP) of four of the host countries shown in table 13: Ireland (17 percent), Singapore (9 percent), Canada (9 percent), and the United Kingdom (7 percent).
The relatively high MOFA shares of host-country GDP in these countries can be traced to some of the following factors: A common language with the United States, marketing and commercial legal systems similar to those in the United States, geographic proximity to the United States, the availability of a skilled work force, political stability, and low corporate tax rates.
The MOFA share of host-country GDP was less than 1 percent in four of the host countries shown in table 13: India, China, the Republic of Korea, and Japan. The low shares in most of these countries partly reflect past or present barriers to investment, including limits on foreign ownership.
Origin of output.—The output of MOFA's consists of both the gross product that originates in the MOFA's themselves and the gross product that originates elsewhere and that is incorporated in intermediate inputs purchased by MOFA's from U.S. parents, other foreign affiliates, or from unaffiliated suppliers.
In 1997, gross product originating in MOFA's accounted for 26 percent of MOFA output, compared with 31 percent in 1989 (tables 14 and 15, column 12). The industries with the highest shares of MOFA gross product were "other" manufacturing (mainly tobacco products), services, and petroleum. The high MOFA share in tobacco products partly reflects the fact that these products have relatively few material components that could be outsourced. Like the U.S.-parent share, the MOFA share in services is high partly because the production of services tends to be labor-intensive and is not as easily outsourced as the production of goods. The share in petroleum is high partly because the large fixed capital costs that must be incurred to reach a profitable scale of operation allows a small number of highly integrated companies to dominate production.
The shift to unaffiliated suppliers from 1989 to 1997 was widespread across geographic areas. It was also widespread across industries, but it was most pronounced in manufacturing. MOFA's in some industries probably sought to decrease their reliance on internal production in order to meet rising global competition by specializing in areas in which they had an advantage and by relying on unaffiliated suppliers for other inputs and finished products.
The U.S. content of MOFA output was 11 percent in 1997, compared with 10 percent in 1989. The geographic areas with the highest U.S. content were Canada and Latin America and Other Western Hemisphere, particularly in Mexico. In Canada and Mexico, the high U.S. content mainly reflects these countries' proximity to, and strong economic ties with, the United States.
In Canada, the U.S. content of MOFA output increased from 22 percent in 1989 to 27 percent in 1997; this increase was widespread across manufacturing industries and may have been related to the initial implementation of the Canada-United States Free Trade Agreement in 1989 and the North American Free Trade Agreement in 1994. In Latin America and Other Western Hemisphere, the U.S. content of MOFA output rose from 13 percent to 16 percent, partly as a result of currency-related valuation changes./13/
Real gross product of MOFA's in manufacturing.—Changes in the current-dollar measures of MOFA operations can sometimes be difficult to interpret because they can be strongly influenced by changes in prices and exchange rates. (The effects of exchange-rate changes that are unrelated to the relative prices of goods and services in various countries can be especially problematic.) To overcome these limitations, BEA recently began producing estimates of real gross product for MOFA's./14/ These new estimates provide more meaningful comparisons of gross product across countries because they are based on purchasing-power-parity exchange rates rather than on market exchange rates, and they provide more meaningful comparisons across time because they are in chained dollars.
The real gross product of MOFA's in manufacturing grew more than twice as fast in 1997 (7.7 percent) as in 1989–96 (3.5 percent) (table 16)./15/ The faster growth partly reflected the faster growth in industrial production in most host countries. The real gross product of manufacturing MOFA's in the 19 member countries of the Organisation for Economic Co-Operation and Development grew faster (6.5 percent) in 1997 than total industrial production partly because of the entry of new MOFA's (chart 3). Industrial production in these countries grew 4.6 percent in 1997, compared with 1.1 percent in 1996.
The $15.1 billion increase in the real gross product of MOFA's in manufacturing in 1997 was concentrated in Canada (up $2.7 billion), Ireland (up $2.6 billion), and Germany (up $2.2 billion). In Canada and Germany, the increases primarily reflected rising production of motor vehicles to meet strong demand; in Germany, the acquisition of new affiliates that manufacture automobile components also contributed. In Ireland, the increase primarily reflected the rising production of new, higher value pharmaceuticals to meet strong demand, mainly in other European countries.
The real gross product of manufacturing MOFA's in the United Kingdom decreased sharply ($4.2 billion), partly because of the sale of a few large MOFA's.
Box: Data on U.S. Direct Investment Abroad
Footnotes:
1. The year 1989 is used for comparison because it was a benchmark survey year for U.S. direct investment abroad and because before 1994, gross product estimates (which are the basis for much of the analysis in this article) were only available for U.S. parents in the years covered by benchmark surveys. In addition, in 1989, like in 1997, economic growth continued in the United States and in most major host countries.
This article presents highlights from the 1996 and 1997 annual surveys. More detailed estimates will be available later this year (see the box "Data Availability" on page 15).
2. The average real GDP growth rate for major host countries is a weighted average covering host countries that are members of the Organisation for Economic Co-Operation and Development; these countries hosted roughly 80 percent of MOFA gross product in 1997.
3. The preliminary 1996 estimates were published in Raymond J. Mataloni, Jr., "U.S. Multinational Companies: Operations in 1996," SURVEY OF CURRENT BUSINESS 78 (September 1998): 47–73.
4. In most of this section, the examination of the foreign operations of U.S. MNC's uses the data for MOFA's rather than data for all foreign affiliates, because parents and MOFA's are conceptually under U.S. managerial control (other foreign affiliates may be under the control of foreign owners) and because the necessary data items for this analysis are collected only for MOFA's.
Although MOFA's and U.S. parents are under the control of one or more U.S. parents, in some cases the U.S. parent is, in turn, under the control of a foreign parent company; in 1997, U.S. parents that were ultimately controlled by foreign parents accounted for 11 percent of all U.S. parents and for 8 percent of their gross product.
5. In 1997, the weighted average U.S.-dollar price of the currencies of 23 major host countries (in terms of MOFA gross product) fell 7.8 percent. This decline lowered the dollar value of MOFA gross product by a similar amount when the data reported to BEA in dollars was translated from foreign currencies as was generally necessary.
The average rate of price inflation in these host countries was derived as a weighted average (in terms of MOFA gross product), using, in most cases, the GDP implicit price deflators for the countries.
6. MNC-associated trade accounts for an even larger share of U.S. trade in goods if U.S. businesses owned by foreign MNC's are included. In 1996—the latest year for which data are available—U.S. affiliates of foreign MNC's accounted for 22 percent of U.S. exports of goods and 32 percent of U.S. imports of goods. As noted in footnote 4, however, these U.S. affiliate shares overlap the U.S.-MNC shares because some U.S. parents belong to both groups. In 1997, trade by U.S. parents that were also U.S. affiliates of foreign companies accounted for 13 percent of MNC-associated exports and 26 percent of MNC-associated imports. (See the addenda to table 4.)
7. In the 1994 benchmark survey, exports and imports of U.S. parents and majority-owned foreign affiliates were desegregated into 12 product categories on the basis of the Standard International Trade Classification, Revision 2, United Nations Statistical Papers, ser. M, no. 34 (New York: United Nations, 1975).
8. "Other" industries consists of agriculture, forestry, and fishing; mining; construction; transportation; communication; electric, gas, and sanitary services; and retail trade.
9. Each U.S. parent is classified in the industry that accounts for the largest portion of its sales or, for holding companies, its total income. Many U.S. parents are involved in a variety of business activities; changes in the mix of these activities can cause a parent's industry classification to change, but reclassifications due to minor or temporary shifts in industry mix are avoided. A parent is reclassified by industry only if the change in the primary activity from the prior year is significant or if the change has occurred for 2 successive years.
The change of a parent company's industry from manufacturing to wholesale trade is related to the percentage of that parent's output originating within the parent itself. Parents classified in manufacturing are primarily engaged in the transformation of materials or substances into new products, and parents classified in wholesale trade are engaged in selling merchandise produced by others to businesses.
10. Generally, at the all-industries level, the estimates of U.S.-parent gross product are conceptually consistent with the estimates of gross product for all U.S. businesses in the national income and product accounts. However, for individual industries, inconsistencies may result from differences in the basis for the industrial distribution of the estimates. The industrial distributions of gross product for all U.S. businesses are based on data collected from establishments, which are classified by the principal product or service produced at each establishment, whereas the industrial distributions of U.S.-parent gross product are based on data collected from enterprises (companies), which are classified by the principal product or service produced by all of their establishments combined. Because the establishments of large companies usually are classified in several industries, the distributions of data by industry of establishment can differ significantly from those by industry of enterprise. In this article, U.S.-parent gross product as a share of the gross product for all private U.S. businesses is computed only at the highly aggregated level shown in table 7.
11. The local content of U.S.-parent output is overstated to the extent that domestic purchases include the services imported by the parents themselves and the imported goods and services that are embodied in parents' purchases from domestic suppliers. (These items were not reported separately and thus could not be identified and included in foreign content.)
12. According to this theory, these advantages, such as superior production or marketing techniques, are necessary so that MNC's can overcome the various barriers to investing abroad, such as foreign languages and unfamiliar business environments.
For a discussion of this theory, see Stephen H. Hymer, The International Operations of National Firms (Cambridge, MA: MIT Press, 1976). For a recent appraisal of the theory, see John H. Dunning and Alan M. Rugman, "The Influence of Hymer's Dissertation on the Theory of Foreign Direct Investment" in The Theory of Multinational Enterprises, vol. 1 (Cheltenham, United Kingdom: Edward Elgar Publishing, 1996).
13. The dollar-denominated measures of the operations of affiliates in the two largest economies in the area—Mexico and Brazil—were affected by changes in the host-countries' currencies. From 1989 to 1997, the Mexican peso lost over two-thirds of its value against the U.S. dollar, and Brazil introduced a new currency, the real. The effects of these changes cannot be precisely measured; however, a devaluation of host-country currency generally depresses the U.S.-dollar value of affiliate sales, but it has no direct effect on the dollar value of U.S. exports to affiliates, so that, all else equal, the U.S.-content share of affiliate output is raised.
The effect on U.S. content may be mitigated to the extent that, after a devaluation, MOFA's substitute locally purchased goods for more expensive U.S. goods. However, it is unlikely that such substitution is significant, at least in the short run, because locally produced substitutes are not usually readily available.
14. For a summary of the methodology used to derive the real gross product estimates and for the 1982–88 and 1990–93 estimates, see Raymond J. Mataloni, Jr., "Real Gross Product of U.S. Companies' Majority-Owned Foreign Affiliates in Manufacturing," SURVEY 77 (April 1997): 8–17.
15. The real gross product grew much faster (8 percent) than the current-dollar gross product (2 percent) in 1997. The slower growth in the current-dollar gross product reflected the dampening effect of the appreciation of the U.S. dollar. (See also footnote 5.)
