Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
lI. Structure of U.S. Export Trade and International Sales as They Relate to the Export Licensing System In order to analyze the influence of the export control licensing process on U.S. firms' competitiveness, one first needs to understand the general characteristics of U.S. export trade and international sales as they pertain to the export licensing system. The first section of this chapter provides this overview, while the second section looks at the relationship between the coverage of the export license system and U.S. trade. Information collected from the Department of Commerce (DOC) licensing database, DOC trade statistics, and from the Academy survey of U.S. exporters forms the basis for the analysis presented in this chapter. Referring to Figure 5 again, what will be developed is an estimate of the scope of U.S. foreign economic activity covered to any degree by export controls. The extent of foreign firm activity influenced by the U.S. control system is not captured by these estimates. There is a further limitation on the scope of coverage of the data that will be presented. Whereas the system covers U.S. foreign merchandise transactions, i.e., both 11
t manufacturing and nonmanufacturing-related activity, the data presented will cover only manufactures-related activity. The reason for this limitation is based on the appraisal that the central concern regarding effects of the export controls on competitiveness of U.S. firms centers on the manufacturing sector. The extent to which the manufactures versus total merchandise distinction matters can be gauged by looking just at the data for the value of Individual Validated Licenses (IVLs) issued by the Department of Commerce in Fiscal Year (FY) 1985.9 In that period, the value of all IVL application approvals for all exports was $88 billion. Of the total, only about $SO--or 57 percent of the total-- was for manufactures exports.l°,ll The balance, or nearly $40 billion, was represented by exports of commodities such gives are only one form of validated license. A different form is called bulk license. Refer to Figures 4 and 5 to note the limitation being applied in developing trade coverage related to IVEs. 1OLicense records are usually reported for fiscal periods, while trade data is usually summarized for calendar years. This necessitates switching back and forth between FY and CY. ilThe Department of Commerce tabulation of TV~s by dollar value was the basis for these numbers. We sorted the Ills by manufacturing vs. nonmanufacturing categories to estimate the manufactures share. -12-
as petroleum, petroleum products and bulk materials of various types, even horses. IT.A. U.S. Foreign Sales As indicated in the simple schematic of Figure 5, the export controls cover a portion of U.S. exports and foreign sales made through affiliates. This section looks at the structure of U.S. foreign business activity in these different channels. Exports are discussed first, followed by an examination of total U.S. international sales. In 1985, countries in the multilateral international Coordinating Committee (COCOM) accounted for the destination of about 56 percent of total U.S. exports, while trade with China and the Bloc countries accounted for 3 percent.12 (See Table 1.)13 The balance of U.S. export trade was with other non-COCOM, non-Bloc countries. This last category was a significant component of U.S. 12This is total trade, not just trade identified as requiring any validated license. 13Table ~ shows how U.S. exports for 1980 and 1985 were distributed by regions grouped according to a meaning relevant to this study. Bloc countries were defined to be, using the Export Administration country groupings, country groups W and U. with exceptions being that Laos, Latvia, Estonia, Lithuania, and Mongolia were omitted. -13-
manufactures export trade, representing over two-fifths of U.S. manufactures exports, though over the past five years, it has declined from roughly half of total manufactures export trade. The segment of U.S. manufactures trade most likely to require some type of validated license--whether individual or bulk--can be roughly identified as equivalent to U.S. high technology exports.l4 In 1985, high technology exports represented about 38 percent of total U.S. manufactures exports, or about $68 billion.15 Collecting U.S. high technology export trade into the same geographic groups used in Table 1, exports to COCOM country destination are only about half of total U.S. high technology exports with nearly an equal amount going to Other Country destinations (the East Asian Newly Industrialized Countries (NICs) account for 20 percent, LDCs 44 percent, and developed countries about 20 percent 14High technology products are defined as having at least 5 percent R&D content. 15As will be explained, this definition includes commodities that do not require a validated license. Thus, it overstates the license coverage associated with these trade categories. -14-
of this category).l6 (See Table 2.) High technology trade with China expanded at a rapid rate between 1980-85, though the level remains relatively small. The level of U.S. trade with the Bloc countries remained insignificant between 1980-85. The high tech trade categories most likely to be covered by the export licensing system include: communication equipment, aircraft and parts, computers/ office equipment, scientific instruments, components, and machine tools. These six categories accounted for about 70 percent of U.S. total high technology exports. (See Table 3.) Firms responding to the National Academy of Sciences Questionnaire (to be reviewed in Chapter IV) were predominately from these six trade categories. Trade with the Soviet Union and Bloc countries is an area where U.S. firms report they are especially confronted with unequal treatment with regard to what can be licensed and the ease of getting a license approved relative to 16For electronics trade with some regions of the world, especially South East Asia, a large part of U.S. export trade consists of subassemblies which are assembled and then reexported back to the U.S. Thus, the importance of these regions ("All Other Countries') as consuming areas is overstated in Table 2. -15-
their other COCOM-based competitors.l7 Table 4 shows trade for the European Community (EC) and Japanl8 with the Soviet Union and Eastern Bloc countries compared with U.S. trade levels to those same areas. The EC and Japan export 25 times the value the U.S. does to the Bloc. Relative to total exports, Bloc trade for these countries represents 3 times what it does for the U.S. Still, it should be noted that Bloc trade does not represent a very large component of EC and Japanese export trade, only averaging around 3 percent of their total annual exports. Because the coverage of U.S. national security controls extends to the international sales of U.S. firms (where appropriate) as well as their direct exports from the United States, one needs to look at the global span of U.S.-based firm sales, including exports, if one wants to develop a more complete measure of the scope of U.S. business activity reached by the national security controls. In 1982, the worldwide sales of U.S. firms 17See Appendix B. the case study of the machine tool sector, for example. Some of the constraints cited arise solely from foreign policy controls. 18This is approximately total COCOM trade, excluding Canada and the U.S. -16-
totalled $3,491 billion,l9 of which 70 percent was U.S. parent sales, and nearly 30 percent was foreign sales of U.S. affiliates.20 (See Table 5.) Sales of manufactures constituted 40 percent of U.S. parent sales, and 35 percent of the foreign sales. In order to develop some rough estimates of the scope of the total foreign sales and related U.S. exports that are likely to come under the U.S. national security controls, the set of U.S. manufacturing sectors specified earlier as the most likely to be covered by the controls is used.21 These categories accounted for $77 billion or 25 percent of total U.S. foreign sales in 1982.22 (See Table 19This discussion draws heavily on data taken from a 1982 U.S. government benchmark survey of U.S. foreign operations. See U.S. Department of Commerce, Bureau of Economic Analysis, U.S. Direct Investment Abroad: 1982 Benchmark Survey Data, (U.S. Government Printing Office, Washington, DC), December 1985. See Table 7, p. 16 20This excludes export sales to unaffiliated parties. 2iSee Table 2. The categories were communications, aircraft/parts, computers/office equipment, instruments, components, machine tools. These were matched to categories defined in the BEA benchmark survey. 22We use this grouping to illustrate likely U.S. foreign sales and report patterns and orders of magnitudes. Tt should be recognized that there is no way to define a perfect mapping between the scope of U.S.-based enterprise trade and foreign sales covered by the controls and readily available statistics on exports and sales. -17-
5.) By 1985, given the growth of the world economy between 1982-1985, it is estimated that U.S. foreign sales in these categories probably increased to about $93 billion. This figure is exclusive of direct export sales by the parent to unaffiliated purchasers. Again, referring to Table 5, of the total of $77 billion of foreign sales of U.S. affiliates in these high tech categories, $65 billion (84 percent) of the sales were concentrated in developed countries and $58 billion (90 percent) was estimated to occur in COCOM countries. Restating that slightly, almost three-quarters of the foreign sales of U.S.-based enterprises which are most likely to be covered by national security controls occur in COCOM countries. To obtain a total of all U.S. foreign sales, the value of direct export sales must be included.23 For 1982, direct exports for the high tech sectors were $30 billion. Combined with sales of foreign affiliates this indicates that U.S. high tech foreign sales in 1982 totaled $107 billion. (See Table 6.) For 1985, the figure would have 23These are export sales by the U.S. parent which do not pass through a related foreign affiliate. -18-
increased to roughly $130 billion.24 The importance of foreign sales to U.S. enterprises is also indicated in Table 6. Approximately one-fourth of all of the sales of U.S.-based firms are foreign sales. Another aspect of the structure of U.S. export trade which is important to the export control process is the role of intrafirm trade.25 U.S. exports are not always transactions between unrelated parties. This reflects the multinational character of many U.S. firms. The extent to which intrafirm trade is a factor in U.S. exports can be examined by breaking it down into the part that flows to unaffiliated versus the part that flows to affiliated parties. The Bureau of Economic Analysis's benchmark survey data on U.S. foreign operations for 1982 indicates that of the total for U.S. merchandise exports of $212 billion, $163 billion, or 77 percent, was shipped by U.S.- based parents. (See Table 7.) Of manufactures exports, 24This estimate is an extrapolation taking into account real growth and general price level changes, but does not take into account any negative factors, such as relative tightening, if any, of U.S. export controls. 25This discussion is partly based on the article "Intrafirm Trade and U.S. Protectionism: Thoughts Based on a Small Survey," by Jane S. Little, New England Economic Review, Jan/Feb. 1986, pp. 43-49. Little estimated that roughly 30-40 percent of U.S. exports are intrafirm trade. _19 _
U.S. firms accounted for $102 billion of the $163 billion total, or 63 percent.26j27 Two-thirds of U.S. parent U.S. exports went to unaffiliated parties. The high tech segments of exports to unaffiliated parties equalled 46 percent. For high tech shipments to affiliated parties, about half went to COCOM country destinations. U.S. exports to affiliated and unaffiliated enterprises vary widely by sector. (See Table 8.) Some sectors have large portions of exports going to affiliated parties; this is true for computing and office equipment, electronic components and instruments. While in the communications and other transport equipment sectors, shipments to unaffiliated parties are more the rule. In turn, for trade with affiliated parties overall, 60 percent goes to COCOM country destinations. If electronic components are excluded to account for subassemblies exported to the NICs, it is almost 75 percent. 260f the $102 billion, $65 billion was shipped to unaffiliated parties and $37 billion was shipped to affiliated parties. 27This raises another point about the U.S. licensing system. A fairly significant share of U.S. export trade is performed by foreign firms. A large number of foreign firms operate in the U.S. and export; obviously, they are subject to U.S. national security controls. -20-
The large intrafirm component of export trade has several implications for measurement of competitive effects of the U.S. license system. First, it suggests that some competitive effects of the U.S. license system may show up only outside the U.S. due to U.S. controls on reexport and resale of U.S.-origin goods and technology within the Free World--controls largely unique to the U.S. control system. To illustrate, for large U.S. multinational corporations, the structure of their complex international manufacturing, distribution ard marketing networks may be affected by changes in U.S. licensing regulations. For example, they may lose unaffiliated distributors due to U.S. reporting and compliance rules.28 Thus, to the degree foreign unaffiliated distributors are lost, ultimately U.S. export trade are affected. This suggests that for multinational firms their competitive position within foreign markets may be adversely influenced by U.S. licensing actions, but the linkages back to how this affects their U.S. operations may be difficult to measure directly. A second implication of the high proportion of intrafirm trade is on the type of license utilized by the 28This, in fact, has occurred. See the response to the questionnaire, Appendix C. -21-
firm. The type of license used is, in part, a function of the structure of the firm. Firms with significant intrafirm trade would be expected to utilize bulk licenses to a greater degree than firms which transact mainly with unrelated parties. The distribution and project licenses are designed around multinational firms with a large volume of international transactions. As will be discussed below, over half of all U.S. foreign sales requiring a license are conducted using a bulk license. II.B. Coverage of the Export Control License System of U.S. Foreign Sales While all U.S. exports require a license, only a certain segment requires a validated one--that is, the exporter must receive prior written approval from the Department of Commerce before exporting certain goods or technical data. Typically, the segment requiring a license covers the "more strategic 'I or more sensitive items. The U.S. national security controls process covers not only products as they flow across the U.S. border, but also covers certain transactions abroad. These transactions might include, for example, products produced and sold offshore by a U.S. affiliate; they might also include
products manufactured by a foreign firm incorporating U.S. origin components or technology. As indicated in the preceding section, the foreign sales just of U.S.-based firms potentially covered by the licensing system is estimated to be S130 billion in 1985. This total figure can be divided into transactions under self-license ("noncontrolled") versus transactions under validated license ("controlled"). (This was represented in Figure 5.) The controlled transactions can be divided into transactions controlled for national security or other purposes.29 The national security controls are implemented via either bulk or individual licenses. The largest portion of the total value of U.S. foreign economic activity is transacted under a bulk license. But most of the data on the transactions covered by the licensing system pertain to the IVLs. This is especially true with regard to Commerce Department information on operational parameters of the licensing system. Therefore' estimates 29In estimating the coverage of the export control system, the focus is on the national security controls, but the data used to develop the estimate has included some transactions controlled for other reasons. These other controls only apply to a small proportion of transactions, probably less than ~ percent of the total value. Therefore, the data can for all practical reasons be treated as purely related to national security controls. -23-
are directly developed for IVL coverage while the bulk license coverage is estimated as a residual value. Two separate approaches using different data sets are used to estimate the fraction of U.S. direct manufactures export trade covered by the export control process. This provides a cross-check on the reasonableness of the results. The first approach uses data from both the Commerce Department and the Academy Survey of U.S. exporters. The second approach uses data exclusively developed from companies. In FY1985 (1984.4-1985.3), Commerce records indicate that applications for individual licenses were received for approximately $64 billion of manufactures exports.30 In the same time period, license approvals for manufactured goods totalled about $50 billion.31 This figure includes more than just the value of licenses for direct export from 30In this time frame, Commerce received applications totaling $10S.5 billion. Of this total, $41 billion was estimated to be nonmanufacturing exports. 31Note that this value of processed licenses is in FY85, and there is no assurance as to when the license was actually used. The difference between the $64 billion value of applications and $SO billion value of approvals is accounted for by denials, RWAs, and timing differences, i.e., the inventory of license applications pending decision. -24-
the United States. It also includes the value of reexport license applications approved which equalled approximately $6.4 billion in FY85. Thus, the value of direct U.S. exports is about $43.5 billion. It is likely that only 50 to 85 percent of the value of all approved individual licenses are actually used.32 Therefore, between $22 billion to as much as $36 billion of the authorized $43.5 billion was actually shipped under an IVL. The second approach to estimating the extent of U.S. exports covered by an individual validated license uses only data developed from the Academy Questionnaire sample of firms which have foreign sales and whose sales are to some degree covered by the licensing system.33 Forty-one percent were shipped or sold General Destination (GDEST); that is, without requiring a validated license. Applying this distribution to the estimated total U.S. foreign high tech sales in 1985 of $130 billion indicates that approximately 60 percent, or $78 billion, were transactions 32This estimate is taken from responses to the Academy Questionnaire and interviews. One reason firms request more than they actually use is that there is an asymmetry in the valuation process--i.e., firms have an incentive to be sure the license value is sufficiently large, but there is no penalty if the actual value shipped is less. 33 Chapter ITI discusses the validity of the sample.
covered by export controls. (See Table 9.) The portion requiring a license was divided between 39 percent utilizing an IVL and the balance of 61 percent going under a bulk license. Thus, of the $78 billion of foreign sales covered by a validated license in 1985, about $30 billion was covered by an IVL.34 This compares reasonably well with the range of the other estimate of between $22 to $36 billion for IVL covered shipments. The balance of foreign transactions, $42 to $56 billion, goes under some form of bulk license. Note that these estimates do not reach to the transactions of foreign firms--including distributors, OEMs, etc.--who utilize U.S.-origin commodities and technology which are subject to U.S. export controls. 34To arrive at this number, the following estimates were made. The $107 billion, Table 6, was adjusted to a 1985 value of $130 billion. The distribution of license coverage by category was then applied. See Table 9 for the distributions. IVEs were used to cover approximately 23 percent of total foreign sales or 39 percent if only the portion of trade requiring validated licenses is looked at. This percentage was applied to the $78 billion. -26-
