首页    期刊浏览 2024年11月09日 星期六
登录注册

文章基本信息

  • 标题:Valuation of the U.S. net international investment position.
  • 作者:Landefeld, J. Steven ; Lawson, Ann M.
  • 期刊名称:Survey of Current Business
  • 印刷版ISSN:0039-6222
  • 出版年度:1991
  • 期号:May
  • 语种:English
  • 出版社:U.S. Government Printing Office
  • 摘要:This article reviews the issues surrounding the valuation of the U.S. net international investment position and presents revalued estimates for direct investment, for U.S. gold reserves, and for the international investment position. The article describes two alternative methods for valuing direct investment in prices of the current period, presents estimates of the direct investment totals for 1982-89 that are prepared using these methods, and compares these estimates with BEA's existing historical-cost estimates and with current-value estimates from several earlier studies. (Estimates for 1990 and revised estimates for 1987-89 will be presented in the regular article on the international investment position next month.)
  • 关键词:Corporations;Foreign investments;International business enterprises;Multinational corporations

Valuation of the U.S. net international investment position.


Landefeld, J. Steven ; Lawson, Ann M.


Valuation of the U.S. Net International Investment Position

This article reviews the issues surrounding the valuation of the U.S. net international investment position and presents revalued estimates for direct investment, for U.S. gold reserves, and for the international investment position. The article describes two alternative methods for valuing direct investment in prices of the current period, presents estimates of the direct investment totals for 1982-89 that are prepared using these methods, and compares these estimates with BEA's existing historical-cost estimates and with current-value estimates from several earlier studies. (Estimates for 1990 and revised estimates for 1987-89 will be presented in the regular article on the international investment position next month.)

In the mid-to-late 1980's, concerns began to arise about the mix of valuation methods used by BEA in deriving the net international investment position. Although many of the assets in the U.S. international investment position (such as portfolio investment and most reserve assets) were being valued at current-period prices, other assets (such as direct investment and U.S. gold reserves) were being valued at the historical costs at which they were purchased. In 1990, BEA suspended publication of the net international investment position of the United States and announced that it was undertaking a review of alternative methods of valuing international investment to reflect current-period prices.(1)

The BEA review focused on direct investment because the largest differences between historical and current costs in the international investment position were thought to have resulted from a significant nmisstatement of the relative positions for U.S. direct investment abroad (USDIA) and foreign direct investment in the United States (FDIUS). Because most USDIA in the 1989 stock occurred in the 1960's and 1970's, it seemed likely that these assets would require a significantly larger adjustment for the cumulative effects of inflation than would those for FDIUS, most of which occurred in the late 1970's and 1980's.(2)

Revaluation of direct investment.--As a result of its review, BEA has developed two measures--current-cost and market-value--to revalue its estimates of the USDIA and FDIUS positions in prices of the current period. The current-cost method revalues the U.S. and foreign parents' share of their affiliates' investment in plant and equipment using a perpetual inventory model to estimate the net stock of direct investment capital at current costs, revalues direct investment in land using general price indexes, and revalues direct investment in inventories using estimates of their current replacement cost. The market-value method revalues the owners' equity portion of the direct investment position for USDIA and FDIUS using indexes of stock market prices. Thus, the two methods can be viewed as revaluing, respectively, the asset side of a balance sheet and the liabilities and owners' equity side of a balance sheet (see the box "Revaluation of Direct Investment in a Hypothetical Balance Sheet"). The market value differs from the current-cost value in that it is an estimate of firms' aggregate net worth, including not only the current value of tangible assets, but also the market value of intangible assets--such as patents, trademarks, management, and name recognition. The market value may also reflect changes in the general economic outlook or in the outlook for a particular industry--changes that may not be related to the prices of tangible assets.

BEA's revaluation of direct investment assets from historical cost to current

cost raises the value of the USDIA position at yearend 1989 by $162.4 billion, to $535.9 billion, and raises the FDIUS position by $56.7 billion, to $457.6 billion (chart 4 and table 1). Revaluation of owners' equity from historical cost to market value raises the value of the USDIA position at yearend 1989 by $431.1 billion, to $804.5 billion, and raises the FDIUS position by $142.9 billion, to $543.7 billion. On a historical-cost basis, the U.S. net direct investment position at yearend 1989 was -$27.4 billion. Revaluation to current cost raises the net position to $78.3 billion; revaluation to market value raises the net position to $260.8 billion. The difference between the current-cost and market-value estimates reflects significantly different rates of change in recent years in stock prices and in replacement costs of tangible assets.

Revaluation of U.S. gold reserves.--BEA has revalued U.S. gold reserves from the 1973 par value of $42.22 per fine troy ounce previously used in the international investment position to the yearend market price, as reported for gold on the London fixing. the revaluation puts gold reserves on the same current-cost valuation basis as other reserve assets and values gold reserves on the same basis as gold held in private portfolios.

The following tabulation provides the historical values for U.S. gold reserves based on the 1973 par value and the current values based on market prices. Revaluing U.S. gold reserves to the yearend 1989 market price of $401.50 per fine troy ounce raises the 1989 value of these reserves in the investment position by $94.1 billion, from $11.1 billion to $105.2 billion.

U.S. international investment position.--After the revaluations of direct investment and U.S. gold reserves, the major components of the international investment position may be viewed as valued at or near current-period prices (table 2). The following list summarizes the valuations used for the major investment position components:

* Direct investment has been

revalued to current-period prices using

both stock market prices for

equity investment and current-cost

values for tangible assets.

* Portfolio investments in foreign

and U.S. securities are valued at

current-period prices; for these

frequently traded assets held in

private and public portfolios, the

position estimates are based on

changes in stock market prices

and, in the case of bonds, on

changes in bond prices.

* Short-term loans and other short-term

liabilities to banks and non-banks

are recorded at historical

cost because the face, or claim,

value recorded on a firm's books

is normally roughly equal to the

current-period value.

* Official reserve assets are valued

at current-period private market

prices; U.S. gold reserves have

been revalued to current-period

private market prices.

* Long-term loans and other long-term

liabilities are valued at

historical cost. For loans held to

maturity, the maximum claim a

lender can collect is the book value

of the principal on the loan, so

loans and other long-term

liabilities generally need not be revalued

to reflect inflation.

In recent years, the Third World

debt problem and the U.S.

savings and loan problem have

indicated that there may be sizable

differences, reflecting increased risk

of default, between market

values and book values.

Unfortunately, the available estimates

of market value--from secondary

markets, appraisals, or indirect

methods--are of limited value.

BEA's revaluation of the U.S. direct investment position and the U.S. reserve gold position from historical cost to current cost reduces the deficit in the U.S. net international investment position at yearend 1989 by $199.8 billion, to -$464.0 billion. The revaluation to market value reduces the deficit by $382.3 billion, to -$281.4 billion (table 3).

It should be noted that unrecorded capital inflows could have a significant impact on BEA's position estimates. During the 1980's, there was a large and persistent statistical discrepancy between the current and the capital accounts in the U.S. balance of payments. The cumulative statistical discrepancy, which amounted to $178 billion, indicated either an overstatement of the current-account deficit or an understatement of net capital inflows into the United States. To the extent that this statistical discrepancy was due to unrecorded capital inflows, particularly of portfolio capital, the foreign investment position in the United States is understated. The Economic Statistics Initiative in the Administration's fiscal 1992 budget calls for improving the estimates of U.S. capital flows. Under this initiative, the measures of international flows of portfolio capital would be strengthened to take into account new channels of financing and new types of financial instruments, and the measures of direct investment would be strengthened by including estimates for small reporters and nonreporters.(3)

Position estimates and measures of wealth.--The current-cost estimates presented in this article put the U.S. international investment position estimates on a basis comparable with BEA's current-cost estimates of total U.S. fixed reproducible tangible wealth and with the Federal Reserve Board's estimates of U.S. domestic net worth--that is, the sum of tangible assets located in the United States, including plant and equipment, inventories, and land.(4) With consistent current-cost estimates of the value of foreign assets in the United States and of U.S. assets here and abroad, it is possible to evaluate changes in the size of national net worth, the distribution of net worth between foreign and domestic saving and investment, and changes in the rate of return to such investments over time.

At yearend 1989, domestic net worth in the United States was $16,017.2 billion.(5) After BEA's revaluations, the current-cost value of domestic assets owned by foreigners was $1,579.3 billion, and the current-cost value of U.S. assets abroad was $1,025.1 billion, and the value of U.S. monetary gold and of special drawing rights was $115.1 billion. Subtracting the current-cost value of domestic assets owned by foreigners from domestic net worth and adding the current-cost value of U.S. assets abroad and the value of U.S. monetary gold and of special drawing rights produces a national net worth of $15,578.1 billion at yearend 1989.

Valuation of Direct Investment

The question of undervaluation of the U.S. direct investment position abroad relative to the foreign direct investment position in the United States was first explored in a series of papers beginning in the late 1980's; the most comprehensive were by Ulan and Dewald, Eisner and Pieper, and Lederer.(6) These authors used a variety of techniques to estimate the current-cost value of direct investment: Revaluation of the cumulative direct investment flows by using a replacement cost index for capital goods or by using various stock market indexes; capitalization of the annual earnings flows from FDIUS and USDIA by a common discount rate to derive an implicit current value of the positions; and use of the ratio of current-cost value to historical-cost value for the U.S. stock of property, plant, and equipment (PP&E) and inventories to estimate the current replacement cost value of tangible assets related to USDIA and FDIUS. In producing the current-value estimates of the direct investment position, BEA has built upon and refined the methods used in these exploratory studies. The remainder of this section describes BEA's methodology and estimates and then compares them with these studies.

BEA's current-cost estimates

Method.--The current-cost method revalues tangible assets using a petual inventory model for plant and equipment, general price indexes for land, and special adjustment factors for inventories. The model used for revaluing the direct investors' shares of investment in plant and equipment by affiliates is the same one used to derive BEA's estimates of total U.S. fixed reproducible capital. The parent's share of equity in FDIUS and USDIA affiliates has averaged about 80 percent in recent years.

The perpetual inventory model first revalues each year's plant and equipment investment from historical cost to constant cost using U.S. capital goods price indexes for FDIUS and a weighted average of country-by-industry price indexes for USDIA. The constant-cost gross capital stock of plant and equipment for a given year is then obtained by cumulating past investment in plant and equipment and deducting the cumulated value of plant and equipment investment that has been discarded, using estimated average service lives and retirement patterns. The constant-cost net capital stock of plant and equipment is obtained in a similar manner, using a depreciation formula to write off the value of the assets over their service lives. The constant-cost net capital stock is then revalued to current cost using the appropriate capital goods price indexes.

The current-cost values for the net capital stock of plant and equipment derived by this method are added to current-cost estimates of the parents' share of their affiliates' land and inventories. Land is revalued using U.S. and foreign gross national (domestic) product price indexes. Inventories are revalued using ratios of current-cost to historical-cost values for U.S. inventory stocks. The sum of the revalued plant and equipment, land, and inventories produces a current-cost replacement value for all tangible assets.

One of the major advantages of the perpetual inventory model is that it explicitly takes into account current-cost depreciation, as well as the timing pattern of investments and differences in prices across industries and countries. Nevertheless, uncertainties about the appropriate choice of service lives and pattern of depreciation can have a large impact on the resulting estimates of capital stocks of plant and equipment. The sensitivity of the estimates to changes in underlying assumptions, as well as a more detailed discussion of the methodology, is presented in the "Technical Notes."

Estimates.--Although revaluation to current costs significantly changes the relative levels of the USDIA and FDIUS positions, the trend in the current-cost estimates is similar to that in the historical-cost estimates--both show a smaller increase in the USDIA position than in the FDIUS position during the 1980's. From 1982 to 1989, the USDIA position in current costs grew $161.9 billion, from $374.0 billion to $535.9 billion. Over the same period, the FDIUS position in current costs grew $284.3 billion, from $173.2 billion to $457.6 billion. As a result, the net direct investment position dropped from $200.8 billion in 1982 to $78.3 billion in 1989.

The sources of change in the year-to-year USDIA and FDIUS positions in current costs are presented in table 4. In the table, changes attributable to capital inflows and outflows are distinguished from changes attributable to valuation adjustments for price changes, exchange rate changes, and "other changes."

The price change adjustment reflects changes in capital goods prices (either from movements in the price of, or from shifts in the mix of, capital goods) that cause changes in the average age and price of the stock. This price change adjustment is generally negative when PP&E prices are declining--as they were in the United States in 1982-84--or when current-period PP&E investments are large enough, relative to earlier period investments, to lower the average age of the PP&E stock. The price change adjustment is generally positive under the opposite circumstances.

The exchange rate adjustment reflects the effect of translating the current-cost estimate into U.S. dollars using the yearend exchange rate times its percent change from a year earlier. The exchange rate adjustment to the USDIA position moves inversely to changes in the value of the U.S. dollar relative to other major currencies: The rise in the dollar in 1982-84 and in 1988-89 reduced the value of USDIA in foreign currencies, and the decline in the dollar in 1985-87 raised the value of USDIA in foreign currencies.

The "other changes" adjustment is a statistical entry that includes revisions due to changes in coverage, statistical discrepancies, the effect of the interaction between exchange rates and price changes, and other statistical adjustments to the value of assets.

The change in the current-cost USDIA position was $36.4 billion in 1989, compared with $14.3 billion in 1988. Capital outflows contributed $31.7 billion to the 1989 change in position. Valuation adjustments for price changes and for "other changes" increased the position by $8.7 billion, and adjustments for exchange rate changes lowered it by $4.0 billion.

The change in the current-cost FDIUS position was $73.6 billion in 1989, compared with $61.3 billion in 1988. Capital inflows contributed $72.2 billion to the 1989 change in position. Valuation adjustments for price changes increased the position by $2.2 billion, and adjustments for "other changes" decreased it by $0.8 billion. (Because U.S. affiliates of foreign parents generally maintain their financial accounts in U.S. dollars, the adjustment for changes in exchange rates is negligible.)

BEA's market-value estimates

Method.--The market-value method for estimating the value of the direct investment positions in current-period prices revalues the historical-cost value of equity in foreign affiliates of U.S. parents using weighted average foreign stock prices. The method revalues equity in U.S. affiliates of foreign parents using a broad-based U.S. stock price index. BEA's estimates revalue only the owners' equity portion of the position; as noted earlier, the liabilities portion is assumed to be approximately valued at current-period prices.

The market-value method is similar to that used by BEA to value portfolio investment in that both use stock price indexes to revalue equity interests in companies. The major difference is that portfolio investments are composed of frequently traded securities, whereas U.S. and foreign affiliates are often wholly owned subsidiaries, and their stock may not be publicly traded. The key assumption is that revaluation of direct investment using general stock price indexes produces on average a reasonable estimate of the aggregate value of affiliates in a country. See the "Technical Notes" for a more detailed discussion of the methodology.

Estimates.--On the market-value basis, unlike on either the historical-cost or the current-cost basis, the USDIA position increased more than the FDIUS position from 1982 to 1989. Although both U.S. and foreign stock market indexes rose to record levels in the 1980's, stock market prices increased more rapidly abroad than in the United States. From 1982 to 1989, the USDIA position at market value grew $576.2 billion, from $228.3 billion to $804.5 billion. Over the same period, the FDIUS position at market value grew $410.7 billion, from $133.0 billion to $543.7 billion. As a result, the net direct investment position increased from $95.3 billion in 1982 to $260.8 billion in 1989.

From 1982 to 1984, the market-value estimates of the USDIA position were lower than the current-cost estimates. As foreign stock market indexes jumped in 1985, the market-value estimate moved slightly higher than the current-cost estimate. By yearend 1989, the market value of USDIA was $804.5 billion, $268.6 billion higher than the current-cost estimate.

Detailed information on the sources of change in the year-to-year USDIA and FDIUS positions on a market-value basis is not yet available. It is clear, however, that changes attributable to stock prices and capital flows predominated over changes attributable to exchange rates and other factors.

Comparison of BEA's estimates with those of earlier studies

Table 5 presents the alternative valuations of the positions for USDIA and for FDIUS that have been made by BEA and by authors of earlier studies. The methodologies used and results obtained are compared in this section.

Current-cost method.--In addition to using different source data, the BEA current-cost estimates differ from the current-cost estimates from various earlier studies for two methodological reasons.

First, BEA's current-cost measures differ from those of Ulan and Dewald and of Eisner and Pieper because BEA applies the tangible-asset price indexes only to the tangible assets. Both sets of authors applied price indexes for capital goods to the entire direct investment flow. As Lederer pointed out, broad application of the tangible-asset price indexes to all flows is incorrect because these flows are used by affiliates to finance a wide range of investments, ranging from plant and equipment to financial assets, a significant share of which are assets--such as cash and trade receivables--that do not need to be revalued. Among assets other than tangible assets, only equity stock in other corporations and intagible assets such as goodwill might arguably be revalued.

Second, BEA's current-cost estimates, unlike Lederer's estimates, are based on the perpetual inventory model, which explicitly takes into account the timing and composition of investment in plant and equipment and of prices both here and abroad. Lederer's estimates were based on the single ratio of current cost to historical cost for the total U.S. capital stock of plant and equipment and other tangible assets. This approach implicitly assumes that the timing of investment flows, the distribution of assets, and the rate of inflation are the same for U.S. domestic investment, USDIA, and FDIUS; however, three-fourths of FDIUS included in the yearend 1989 FDIUS position occurred in the 1980's and thus requires a smaller revaluation than the USDIA position, a large share of which occurred in the 1960's and 1970's.

Market-value method.--BEA's market-value estimates differ from those of Ulan and Dewald because the BEA method excludes the portion of the movements in stock prices that are attributable to the retention of earnings. In this way, BEA avoids the double-counting of retained earnings in the Ulan and Dewald estimates that resulted from their applying an unadjusted stock price index to direct investment capital flows that included reinvested earnings. Furthermore, BEA's market-value estimates differ from those of Ulan and Dewald and of Eisner and Pieper because BEA's adjusted stock price indexes are applied only to the owners' equity portion of the direct investment capital flows; in contrast, both sets of authors applied their price indexes to the entire flow of direct investment capital.

Capitalization of earnings.--BEA has not produced an estimate based on the capitalization of direct investment earnings because of the large uncertainties involved in choosing an appropriate rate of discount. Given the existence of exchange rate risks, expropriation risks, less than perfect capital mobility, and persistent differences in interest rates across countries, it seems unreasonable to assume that a single discount rate could be appropriate for discounting investment flows from USDIA and FDIUS; further, small differences in discount rates produce large differences in the capitalized value of earnings. In addition, choosing a discount rate predetermines the rate of return one can derive from the capital stock, and thus yields no independent information.

Valuation of Gold and Debt

U.S. gold reserves

In order to more accurately reflect the current value of all assets in the international investment position and to provide consistent current-cost treatment of U.S. gold reserves with other reserve assets and private gold, BEA has revalued gold reserves from the 1973 par value of $42.22 per fine troy ounce to yearend market prices, as reported for gold on the London fixing.

Using the yearend 1989 market price of gold of $401.50 per fine troy ounce raises the 1989 value of U.S. reserve holdings of gold by $94.1 billion, from $11.1 billion to $105.2 billion. Revaluation to market value significantly raises the value of gold reserves throughout the 1982-89 period. The physical U.S. gold stock changed little throughout 1982-89, so virtually all of the changes in the year-to-year position of gold at current cost reflect changes in the price of gold. From 1982 to 1989, the current-cost value of U.S. gold reserves declined from $120.7 billion to $105.2 billion.

Long-term loans and other long-term debt

The valuation of debt, particularly that of heavily indebted nations, is a major issue for the 1990's, both here and abroad. In the past, valuation at historical cost seemed reasonable for debt that was unlikely to be sold in secondary markets--for example, government or bank debt. Bad debts, when deemed uncollectible, were written off by banks or forgiven by governments, and these writeoffs were reflected in the position estimates. Although a large dollar volume of debt to Third World nations was written off or forgiven during the 1980's, much debt that may yet have to be written off or forgiven is still being recorded at book value. In recent years, the rescheduling, selling, repurchasing, and swapping of such debt has led to development of a secondary market for the debt of these nations.

While there is some default risk attached to the debt of a substantial number of countries, market attention has focused on the debt of heavily indebted countries. For these countries, the secondary market value of their long-term bank debt has been estimated at about one-third of the book value of that debt.(7) Ulan and Dewald, using these secondary market values, estimated that discounting bank loans to less developed countries would reduce the value of claims reported by U.S. banks by $40-50 billion in 1989. Such estimates are speculative because secondary markets are extremely thin; any large purchase can substantially change the secondary market price. Indeed, when Brazil bought back a portion of its own debt in March 1988, the secondary market price of Brazilian debt doubled. In addition, these secondary market discounts cannot simply be applied to bank debt to produce market-value estimates, because the value of bank claims varies substantially according to the extent to which loans have been collateralized and/or subordinated. Moreover, many of these loans have been written down substantially from face value, and the true market value of current bank claims may be only half of the amount implied by such estimates.

Although revaluation of debt was not attempted in the work reported in this article, BEA intends to examine the question further. The issue will face BEA--for both domestic and international debt--in the more general context of moving to an integrated set of national and international income and wealth accounts.(8)

Technical Notes

This section provides additional detail on the two methods--current-cost and market-value--used by BEA to revalue the USDIA and FDIUS positions. The discussion covers the assumptions underlying each method, including tests of the sensitivity of the estimates to several of these assumptions.

Current-cost method

Under this method, U.S. and foreign parents' shares of affiliates' tangible assets--inventory stocks and PP&E--are revalued to current costs. Inventory stocks are revalued using ratios of current-cost to historical-cost inventory stocks for nonfarm corporate business from the U.S. national income and product accounts (NIPA's); these adjustments convert inventories from historical costs to current replacement costs. For FDIUS, land is revalued using the implicit price deflator for gross national product; for USDIA, land is revalued using country-specific implicit price deflators for gross national (for domestic) product. Plant and equipment is revalued using a perpetual inventory model.

Perpetual inventory model.--The current-cost method uses a perpetual inventory model to estimate the gross and net stocks of plant and equipment for foreign affiliates of U.S. parents and for U.S. affiliates of foreign parents, by industry and geographic area.(9) The model starts with plant and equipment investments in current and constant dollars and obtains the gross plant and equipment capital stock for a given year by cumulating past plant and equipment investments and deducting the cumulated value of plant and equipment that has been discarded or retired, using estimated average service lives and retirement patterns. Net plant and equipment capital stocks are derived by deducting depreciation for plant and equipment from the gross stock. The depreciation estimates are based on the straight-line formula used in the NIPA's, in which annual depreciation for a fixed asset is equal to its gross value divided by its service life.

The constant-cost estimates measure the net plant and equipment stocks in the prices of a base year, according to the following equation: [K.sub.n]=[summation]([I.sub.t]-[D.sub.t]) ([P.sub.b]/[P.sub.t]).

In this formula, [K.sub.n] is the constant-cost net stock of plant and equipment in year n, expressed in the prices of base year b; [I.sub.t] is plant and equipment expenditures, net of discards of retired plant and equipment, in year t; [D.sub.t] is the estimated annual depreciation in year n on the plant and equipment purchased in year t; [P.sub.b] is the price that would have been paid in the base year for the mix of plant and equipment purchased in year t; and [P.sub.t] is the price of the plant and equipment in period t. The net plant and equipment stock in a country or region is the summation of net plant and equipment stocks across all industries in the country or region.

Current-cost plant and equipment estimates are derived by multiplying constant-cost plant and equipment estimates by current-period price indexes. Thus, current-cost estimates measure the plant and equipment stocks in prices that would have been paid if the stocks had been purchased in the period to which the plant and equipment estimates refer.

PP&E expenditures.--For USDIA and FDIUS, PP&E expenditures are derived from BEA's direct investment surveys of foreign and U.S. affiliates. For USDIA and FDIUS, it is assumed that the parents' share of PP&E expenditures equals the affiliates' PP&E expenditures multiplied by the parents' share of ownership in the affiliates.

Gross PP&E stocks at historical-cost (book) value are also available from BEA's direct investment surveys. Yearend changes in the gross stock of PP&E (also weighted by the parents' share of ownership) that are not explained by current PP&E expenditures or discards are the result of acquisitions or divestitures of affiliates and of benchmark revisions. Such changes are treated as transfers of used PP&E to or from affiliates.

Annual PP&E investments--PP&E expenditures adjusted for discards, acquisitions, divestitures, and benchmark revisions--are distributed into the components of PP&E using detailed information from BEA's benchmark surveys of FDIUS and USDIA. Additional adjustments are made to include expensed petroleum and natural gas exploration and development expenditures in PP&E investments and stocks. Although companies may expense certain petroleum and natural gas exploration and development expenditures for financial reporting, BEA treats these investments as capitalized for the purpose of developing current-cost estimates consistent with NIPA concepts.

For FDIUS, annual PP&E expenditures at historical cost by industry of U.S. affiliate are available from the 1974, 1980, and 1987 benchmark surveys and from the 1977-79, 1981-86, and 1988 annual surveys of FDIUS. Estimates are made for 27 industry groups of affiliates. Because such estimates are not yet available for 1989, PP&E expenditures are estimated by extrapolating the results by industry from the Census Bureau's Plant and Equipment Expenditures Survey. Gross PP&E stocks at historical cost by industry of affiliate are available for 1974 and for 1980-88. Foreign parent ownership shares, by industry, are available from the 1974, 1980, and 1987 benchmark surveys and for large affiliates from the 1981-86and 1988 annual surveys.

For USDIA, annual PP&E expenditures at historical cost by geographic area and industry of majority-owned foreign affiliates (MOFA's) are available from the 1957, 1966, 1977, and 1982 benchmark surveys and from the 1958-65, 1967-76, 1978-81, and 1983-89 annual capital expenditures surveys of USDIA.(1) Gross PP&E stocks for MOFA's are available from the 1966, 1977, and 1982 benchmark surveys and the 1983-88 annual surveys. Parent ownership shares, by geographic area and industry, are available from the 1966,1977, and 1982 benchmark surveys and from the 1983-89 annual surveys.

For the estimates of PP&E expenditures and stocks for USDIA to be consistent with those for FDIUS, data on PP&E expenditures and stocks are needed for both MOFA's and minority-owned foreign affiliates (MINOFA's).(11) PP&E data for MINOFA's are not as complete as those for MOFA's. As a result, the relationships between net PP&E stocks for MOFA's and MINOFA's, by region and industry, as reported in BEA's 1982 benchmark survey are used to proportionally adjust the MOFA's PP&E expenditures and stocks, by region and industry, to an estimated total for MOFA's and MINOFA's combined.

For USDIA, the revaluation adjustments were based on weighted averages of data from the following countries or groups of countries: Canada, France, Germany, Italy, Japan, the United Kingdom, all other countries in Europe, and a residual for all other countries in the rest of the world.(12)

Priceindexes.--For FDIUS, current-and constant-cost values for plant and equipment are derived using the annual price indexes for U.S. investments in plant and equipment, by industry, from BEA's capital stock estimates. Current- and constant-cost estimates of investment in land are derived using the implicit price deflator for U.S. gross national product.

For USDIA in Canada, France, Germany, Italy, Japan, and the United Kingdom, the current- and constant-cost values for plant and equipment are derived using the appropriate country price index, available from the Organisation for Economic Cooperation and Development (OECD), for nonresidential structures and for nonresidential equipment. Current-and constant-cost estimates of investment in land are derived for each country using its price deflator for gross national (or domestic) product.

For USDIA in "other Europe," country price indexes, available from the OECD, are used to develop weighted price indexes for structures, equipment, and gross domestic product. For USDIA in the rest of the world, U.S. price indexes are used because reliable weighted indexes for the developing countries are not available; furthermore, foreign affiliates in developing countries, particularly affiliates in the petroleum industry, are believed to acquire much of their equipment from the United States.

Average service lives.--The average service lives and retirement patterns used for FDIUS plant and equipment are the same as those used by BEA to derive the estimates of total U.S. private fixed reproducible tangible wealth.

The service lives used for USDIA plant and equipment in Canada, France, Germany, Italy, Japan, and the United Kingdom are those used in the national economic accounts of those countries, as reported to the OECD.(13) The service lives for nonpetroleum investments in other developed countries are based on service lives used in selected small European countries and on service lives in Canada, France, Germany, Italy, Japan, and the United Kingdom. The service lives used for nonpetroleum investments in less developed countries are based on those for developed countries, but they have been lengthened because less developed countries are assumed to have slower technological obsolescence and lower labor costs (and maintenance costs) relative to capital acquisition costs. The service lives used for petroleum investments are judgmental estimates and are considerably longer than those used by BEA for the domestic petroleum industry; the use of longer service lives reflects the slower, more efficient rate at which oil is extrated in foreign countries.

Alternative service lives and the depreciation formula.--BEA examined a number of alternative assumptions about the appropriate service lives and formulas to use for depreciation. Several of these assumptions are discussed in the following paragraphs.

It is possible that the longer average service lives for used for USDIA do not reflect actual differences in practice between the United States and other countries. If the USDIA position at current costs were recalculated using the shorter U.S. service lives (instead of the OECD service lives) for U.S. affiliates abroad, the current-cost USDIA position for 1989 would be $61 billion lower, as would the resulting net direct investment position.

Various studies of depreciation in the United States suggest that depreciation for equipment may be more rapid in the first years of the service life than that calculated using the straight-line formula; studies also suggest that, for structures, either the depreciation rates are less or the service lives are longer than those used by BEA. BEA tested the effects of such assumptions using a declining balance formula with a depreciation rate of 1.8 times the first year's straight-line rate for equipment and using a straight-line formula with 25 percent longer service lives for structures.(14) Combining these alternatives for equipment and structures would raise the FDIUS position by $1 billion in 1989 and the USDIA position by $23 billion; the resulting net direct investment position for 1989 would be $21 billion higher.

Market-value method

Under this method, owner's equity of foreign affiliates of U.S. parents and of U.S. affiliates of foreign parents is revalued to current costs. Owner's equity included in the USDIA and FDIUS positions is the cumulative total of equity capital flows and reinvested earnings. Owners' equity is revalued to current cost using the market-equity model.

Market-equity model.--In the market-equity model, FDIUS is revalued at the aggregate level, and USDIA is revalued by a weighted average country/region estimate. The revaluation formula for parents' equity in affiliates that maintain their financial records in U.S. dollars is [Mathematical Expression Omitted], where [K.sub.t] is the equity investment in affiliates in year t, valued at yearend stock market prices; [Peoy.sub.t] is the yearend stock market price index and [Pavg.sub.t] is the annual average stock market price index, in year t; [I.sub.t] is the total equity capital flow in year t; and [RE.sub.t] is the yearend ratio of retained earnings per share as reflected in the stock price index for year t.

This formula revalues U.S. and foreign parents' equity in affiliates using end-of-year stock price indexes, while adjusting for changes in annual investment and correcting for the effect of retained earnings on stock market prices during the year. The stock market data are first converted into U.S. dollars, so exchange rate effects are reflected in the market indexes.

An additional adjustment is needed for foreign affiliates of U.S. parents that maintain their financial accounts in another national currency and later translate these accounts into U.S. dollars. Investments made during the year by these foreign affiliates must be revalued from the average exchange rate during the year to the yearend exchange rate.

Equity investment flows.--Data on equity capital flows are generally available from BEA's quarterly and benchmark surveys from 1966 to 1989. For both USDIA and FDIUS, the necessary earnings, dividends, equity capital flows, and equity positions are generally available beginning in 1966 for incorporated U.S. affiliates of foreign parents and incorporated foreign affiliates of U.S. parents.

For FDIUS, the 1966 market value of the foreign equity position in incorporated U.S. affiliates is estimated by multiplying the position by the ratio of market-to-book values in 1966 for the Standard and Poor's Index for 400 Industrial Companies.(15) This method assumes that the relationship between market and book values of incorporated U.S. affiliates is similar to that of a typical large U.S. industrial corporation in 1966.

For USDIA, comparable market-to-book-value ratios for 1966 are unavailable for foreign stock markets. Therefore, the 1966 market value of U.S. parents' equity in incorporated foreign affiliates is estimated by calculating the dividends affiliates paid to U.S. parents, assuming market yields in 1966, and then dividing the value of dividends by the market yield for the year.

Time series data for unincorporated U.S. and foreign affiliates are more limited than data for incorporated affiliates. For FDIUS, distributed earnings, equity flows, and equity positions are available for unincorporated U.S. affiliates of foreign parents from 1980 to 1989. Because these data are not available for earlier years, the valuation of unincorporated affiliates begins with data for 1980. A starting position in current-cost values was created by multiplying the equity position in unincorporated U.S. affiliates by the estimated market-to-book-value ratio of incorporated U.S. affiliates in 1980. In 1989, equity capital flows from foreign parents to unincorporated U.S. affiliates accounted for 8 percent of total equity capital flows to the United States from foreign parents.

For USDIA, complete data for unincorporated foreign affiliates are available from 1982 to 1989. An initial position for 1982 was estimated by using the market-to-book-value ratio for incorporated affiliates. In 1989, equity capital flows from U.S. parents to unincorporated foreign affiliates accounted for 12 percent of total equity capital flows from U.S. parents.

Market indexes.--For FDIUS, Standard and Poor's composite stock market data are used to revalue foreign parents' equity in U.S. affiliates. For USDIA, stock market data from Morgan Stanley Capital International are used to revalue U.S. parents' equity in foreign affiliates. OECD stock market data are used for years in which the Morgan Stanley stock market data are incomplete or missing. Investments in countries where country-specific stock market data are not available are revalued using the Morgan Stanley World Index for stocks.

The market-value method, like the current-cost method, is sensitive to the assumptions used. For example, FDIUS equity was revalued using the Standard and Poor's 500 stock market index because that index has broader coverage than the Morgan Stanley index for the United States; if the Morgan Stanley U.S. index were used, the 1989 FDIUS position would be raised by $16 billion. [Table 1 to 5 Omitted] [Chart 4 Omitted] [Revaluation of Direct Investment in a Hypothetical Balance Sheet Omitted]

(1)See "International Investment Position: Component Detail for 1989," Surveyof Current Business 70 (June 1990): 54-65. Before its suspension in 1990, an annual estimate of the net international investment position of the United States was published each year. (2)Inflation drives a wedge between values expressed in historical prices andthose in current prices. During the last 30 years, the International Monetary Fund's world price index has risen more than 4 percent a year, amounting to more than a threefold increase over the period. Such an inflation rate may hinder meaningful comparisons of dollar values at different points in time. As a result, measures of flows, which are in current prices, are often restated to constant prices, and measures of stocks, which are valued in acquisition (or historical) prices, are often restated to current (or to constant) prices. Consistent comparisons of business income and assets over time and of rates of return, capital productivity, and capital/labor ratios require such valuations. (3)See "Improving the Quality of Economic Statistics: The 1992 Economic Statistics Initiative" in the March 1991 Survey. (4)BEA has produced estimates of the gross and net stocks of domestic fixed reproducible assets on consistent current- and constant-cost bases since 1972. The Federal Reserve Board uses BEA's current-cost estimates, along with an estimate of the market value of land, to estimate total tangible assets located in the United States, or domestic net worth, in its balance sheets for the U.S. economy. (5)Board of Governors of the Federal Reserve System, Balance Sheets for the U.S. Economy, 1945-90, Board of Governors of the Federal Reserve System, Publication C (Washington, DC: March 1991). (6)Michael Ulan and William G. Dewald, "The U.S. Net International InvestmentPosition: Misstated and Misunderstood," in James A. Dorn and William A. Niskanen, ed., Dollars, and Trade (Norwell, MA: Kluwer Academic Publishers for the Cato Institute, 1989).

Robert Eisner and Paul J. Pieper, "The World's Greatest Debtor Nation?," in The North American Review of Economics and Finance, vol. 1, no. 1 (Greenwich, CT:JAI Press, 1990).

Walter Lederer, "The Valuation of U.S. Direct Investment Abroad," Unpublished (Washington, DC: Board of Governors of the Federal Reserve System, May 8, 1990). (7)Salomon Brothers, "Indicative Prices for Less Developed Country Bank Loans," January 4, 1990. (8)For a description of BEA's plans for moving to an integrated set of national and international income and wealth accounts, see "The United Nations System of National Accounts: An Introduction," in the June 1990 Survey; and "Improving the Quality of Economic Statistics: The 1992 Economic Statistics Initiative," in the March 1991 Survey. (9)For detailed information on the perpetual inventory model, see U.S. Department of Commerce, Bureau of Economic Analysis, Fixed Reproducible Tangible Wealth in the United States, 1925-85 Washington, DC: U.S. Government Printing Office, June 1987): vii-x. (10)MOFA's are foreign affiliates in which the U.S. parent(s) ownership share is over 50 percent. (11)MINOFA's are foreign affiliates in which the U.S. parent(s) ownership share is between 10 percent and 50 percent. (12)PP&E is revalued according to its location rather than to the location of the direct investment claim. This treatment differs from the usual historical-cost treatment so as to allow for the use of of price indexes and currency exchange rates of the country in which the PP&Eis located. (13)Derek Blades, "Service Lives Of Fixed Assets," OECD Working Paper No. 4 (Paris, France: Organisation for Economic Co-operation and Development, March 1983). (14)These assumptions about depreciation of equipment and structures are similar to the parameters suggested in a study by Hulten and Wykoff; see C.R. Hulten and F.C. Wykoff, "The Measurement of Economic Depreciation," in Depreciation, Inflation, and the Taxation of Income from Capital (The Urban Institute Press, 1981): 94. (15)The equity position of FDIUS in 1966 is not separately available. Therefore, an estimated equity position is derived by multiplying the total 1966 direct investment position by the ratio of equity to total direct investment in 1974, the first year equity is reported separately from debt.
联系我们|关于我们|网站声明
国家哲学社会科学文献中心版权所有