House of Ikpresentat.ives March 1990 FOREIGN INVESTMENT Analyzing National Security Concerns RESTRICTED--Not k be released outside the General Accounting Of&e unless s-e approved by the Office of Congressio~ Belatione. National !Security and International AfWn3 Division B-234611 March 29,199O The Honorable John D. Dingell, Chairman Subcommittee on Oversight and Investigations Committee on Energy and Commerce House of Representatives Dear Chairman Dingell: This is one of a series of reports responding to your request that we examine the effects of foreign direct investment in the United States. This r-e&t discua~ the national security concerns related to foreign investment and the availability and types of data neededto analyze them. The first report in this series was Foreign Investment: Federal Data Collection on Foreign Investment in the United States (GAO/NWW%BR, October 1989). Other reports are in preparation. As you requested, we did not obtain formal agency comments on this report, but we did discuss our findings with agency representatives. Unless you publicly announce its contents earlier, no further distribution of this report will be made until 30 days from its issue date. At that time, we will provide copies to other interested parties. This report was prepared under the direction of Mr. Allan I. Mendelowitz, Director of Trade, Energy, and Finance Issues,who may be reached on (202) 275-4812. Other major contributors to this report are listed in appendix I. Frank c. ccmahan Assistant Comptroller General Executive Summaxy The possibility that key segmentsof defense-related industries could come under foreign control is one of the central concernsin the debate about ‘increasedforeign investment in the United States. At the request of the Chairman, Subcommittee on Oversight and Investi- gations of the House Committee on Energy and Commerce,GAO examined national security concernsrelating to increased foreign invest- ment in the United States. More specifically, GAO examined (1) the definition of industry sectors and technologies that are national security related, (2) the government’s system for learning of foreign investments before they take place, (3) the difficulties the government encounters in analyzing specific investments, and (4) the broader ques- tions not addressedin the present system of analyzing foreign investments. In responseto concerns about the national security implications of for- Background eign investment, in 1988 Congressenacted the Axon-Florio Amendment to the DefenseProduction Act. This amendment gave the President new authority to investigate and block foreign investments threatening to impair national security. To exercise this blocking authority, the Presi- dent must fiid that (1) credible evidence exists that the foreign interest might take action that threatens to impair U.S. security and (2) provi- sions of law, other than the International Emergency Economic Powers Act, do not provide adequate authority to protect the national security. The President delegated his authority to review transactions to the existing interagency Committee on Foreign Investment in the United States, which is chaired by the Treasury Department. Resultsin Brief National security has not been precisely defined, in terms of identifying specific technologies and firms critical to U.S. defense leadership. The DefenseDepartment can identify its direct contractors, but it cannot systematically identify the firms supplying these contractors or foreign investments in these fii. There is no information base or government agency systematically tracking foreign acquisitions of component sup- pliers. It is thus difficult to answer the general question of how much of the defense industrial base has been acquired by foreign-owned firms. As a result of the 1988 amendment, the interagency Committee now receives information on specific foreign investments; and, basedon Page 2 GAO/NSIALMWB4 Forem Investment Executive Sommary recent experience, it appears the Committee will have advance knowl- edgeof a wide range of proposed investments relating to national secur- ity. However, the Committee is not performing analyses of foreign investment by industry sector. Foreign investments in start-up compa- nies are excluded from review by the Committee, becausethe Exon- Florio Amendment refers only to mergers and acquisitions of ongoing businesses. Difficulties experienced by the Committee in analyzing specific invest- ments do not stem from lack of information about the transactions. Rather, they stem from uncertainties about (1) how the investment may affect the future direction of technology transfers, (2) the likelihood of continued supplies to the DefenseDepartment or its direct contractors, and (3) the existence of reliable, alternate suppliers. Broader concerns about the national security implications of foreign investment are not addressedin the present system of analyzing spe cific, proposed investments. Important questions persist, for example, about how to preserve the commercial competitiveness of U.S. industry sectors underpinning defense technology leadership, or how to assess overall technology transfer effects. Principal Findings Defining National Security The defense industrial base supporting national security includes some fii whose business is primarily commercial but whose leading edge technologies are important to U.S. defensetechnology leadership. These fii may not be direct contractors to the DefenseDepartment and would not necessarily be included in information baseson critical defense suppliers, becausethe DefenseDepartment cannot systemati- cally identify the firms supplying its direct contractors. Neither the Committee nor any other US. government agency systematically tracks foreign investments on a sectoral basis, to provide analyses of whether foreign acquisitions of such commercially driven fii have contributed to any erosion of component suppliers in the defense industrial base. A key issue is whether the defense industrial base should be defined to include foreign-owned firms which maintain production in the United States. On the one hand, foreign investments can bring new capital and P8ge 2 GAO/-NSIAD&MM For&n Invatment technology to U.S.-basedproduction facilities. On the other hand, ques- tions can be raised about the long-term consequences of possible tech- nology outflows and dependenceon foreign-based decision-making. Advance Knowledge of The Committee is the central point for gathering and analyzing informa- tion about specific, proposed foreign investments in existing U.S. firms Foreign Investments relating to national security. The Exon-Florio Amendment does not spe- cifically require foreign investors to notify the Committee of proposed investments. However, the consequencesof not notifying the Committee may be severe,becausethe investors may risk divestiture orders later. Becauseof this risk and becausethe concept of national security is not tightly defined, it is likely that the Committee will have advance knowl- edge of a wide range of proposed investments. Investments in start-up companies are not being reviewed by the Com- mittee, becausethe amendment refers only to mergers, acquisitions, and takeovers of “persons engagedin interstate commerce.” Difficulties in Analyzing The Committee has generally had adequate information to analyze spe- cific investments as provided for by the amendment. Although official Investments government investment statistics do not provide detailed information about defense relationships and alternate suppliers, the Committee has been able to gather neededinformation from a variety of public and pri- vate sources. The more difficult questions arising in specific caseswere those requir- ing judgments about the likely future behavior of foreign investors. These required assessments of foreign investor intentions regarding technology transfer, continued supply to the DefenseDepartment or its direct contractors, and use of any commercial advantages gained through the investment. Broader Policy Questions The Committee does not focus on more general questions which have arisen in public debate about the possible effects of foreign investment on the U.S. industry’s commercial competitiveness. Yet, for commer- cially driven industries that are critical to defense technology leader- ship, important questions persist about (1) how to evaluate the importance of a firm’s commercial strength to its defense role, (2) what competitive factors lead US. firms to discontinue operations in sectors P8ge 4 GAO/NSIAIMMM4 Fo~ign Lnvestment Executive Shunmuy that foreign investors appear to value highly, (3) how to track the direc- tion of technology transfers, and (4) whether and how the U.S. govern- ment may need to respond to these competitive factors. These questions need to be addressedat a higher policy-making level and in a broader context than the case-by-caseapproach afforded by the Committee. This report contains no recommendations. Recommendations As requested, GAO did not obtain official agency comments on its report. Agency Comments However, responsible officialswere consulted during the review, and their views were incorporated where appropriate. P8gel% GAO/NSIAD-WW Foreign Investment Contents Executive Summary Chapter 1 8 Introduction Ekon-Florio Amendment 9 Objectives, Scope,and Methodology 9 Chapter 2 11 Defining National ’ Security Chapter 3 Advance Knowledge of Foreign Investments Chapter 4 16 Difficulties in Guidelines for CFIUS Reviews Government Statistics on Aerospace and Semiconductors 16 17 Analyzing Specific Types of Questions Raised by CFIUS 18 Investments Sourcesof Needed Information 19 Difficulties in Deciding Cases 20 Chapter 5 24 Broader Policy Conclusion 26 Questions Persist Appendix Appendix I: Major Contributors to This Report 28 Abbreviations CFIUS Committee on Foreign Investment in the United States DOD Department of Defense GAO General Accounting Office PUe6 GAO/BMAB8@94 Foreign Investment Chapter 1 Introduction Preserving U.S. industrial capabilities in sectors critical to national security has been a traditional U.S. policy goal. An important concern in the debate on foreign investment in the United States is the possibility that key segmentsof industries critical to the national security could come under foreign control through foreign investments here.’ Because US. defense strategy relies on the deterrent effects of technological rather than numerical superiority, concern about foreign investment focuseson the U.S. government’s ability to identify technologies crucial to defense systems and to act to preserve and promote US. leadership in them. Specific restrictions are in place to protect classified defense informa- tion from foreign acquisition and to ensure continued U.S. production of vital defense goods in a crisis. Foreign investments in U.S. firms per- forming classified defense work are monitored under the DefenseIndus- trial Security Program (based on Executive Orders 10460,10866, and 11662). Known as Foreign Ownership, Control, or Influence restrictions, these provide authority to restrict or deny foreign accessto classified information. Although they do not authorize denials of foreign invest- ments, they can, in effect, deter potential foreign investors who are seeking accessto classified information. The possibility that foreign owners might suddenly close down U.S. production of vital defense goods in a crisis is not a major concern, becausethe U.S. government has statutory authority to ensure production at any facility in the United States. The U.S. government, in addition, prohibits or restricts foreign investment in certain sectors, such as energy resources,coastal and domestic shipping, air transport, and broadcasting. Foreign investments in U.S. firms producing unclassified but technologi- cally sensitive military and civilian items are not reviewed under the Defense Industrial Security Program. These investments are of principal concern, in terms of questions about . how to define the range of industry sectors and technologies that are national security related, l whether investments in crucial segmentsof such industries would come to the attention of the US. government before they take place, and . whether the government has adequate information on foreign invest- ment to be able to analyze the implications, particularly the technology ‘Foreign investment refers to foreign direct investment; that is, investment resulting in foreign own- ership or control of 10 percentor more equity interest in a US. business. P8ge 8 GAO/NSUD-@WdForeign Investment transfer effects, of specific investments as they occur and of overall for- eign investment levels in a sector. The basic question, as many have posed it, is whether the U.S. defense industrial base is being gradually “hollowed out” as foreign investment takes place, not necessarily in plants directly producing defensesys- tems, but in the lower tiers of component suppliers producing goods with both military and civilian applications. The congressional responseto this concern was the enactment of the Exon-Florio Exon-Florio Amendment as part of the Omnibus Trade and Competitive- Amendment ness Act of 1988 (amending Title VII of the DefenseProduction Act, 50 U.S.C. App. 2170), which provides presidential authority to investigate and block or suspend new foreign acquisitions or mergers that threaten to impair the national security. This act also formalized an existing gov- ernment review process,conducted by the interagency Committee on Foreign Investment in the United States (CFIUS).* As of February 1,1990, only six proposed foreign investments were selected as warranting the complete CFIUSinvestigation process,out of a total of about 240 investments consideredby CFIUS under the Exon- Florio Amendment. Of these six, the President blocked one investment- by ordering a Chinese firm to divest all of its interest in a U.S. aircraft parts manufacturer. None of the other five investments were formally blocked or dissolved, although two were withdrawn once the investiga- tion phase began. (One of these was later resubmitted in a restructured form approved by CFIUSmembers.) Three of the six investments involved U.S. fums producing high technology electronics components, two involved U.S. aerospacefirms, and one involved a U.S. manufac- turer of extra-high-voltage transformers. The proposed foreign purchas- ers were West German, Japanese,Indian, French, Chinese,and a Swiss- Swedish partnership. At the request of the Chairman, Subcommittee on Oversight and Investi- Objectives, Scope,and gations of the House Energy and CommerceCommittee, we’examined Methodology national security concernsrelating to increased foreign investment in 2By executive order, the President dekg&ed his authority to review Mnsactions to CFlUS. CFWS is stillchairedbythelhasuryDepartment and haa as other membem the Departments of State. Com- merce, and lkfeme, the U.S. Trade Representative, the council of Economic Advisors, the Attorney General,andtheDirectoroftheOf&eofM an@ment and EMget. Other agencies can be brought into CFIUS discussions as appropriate. Page 9 GAO/NSIADMJ-94 Fodgn Investmen the United States. Our objectives were to examine (1) the definition of industry sectors and technologies that are national security related, (2) the government’s system for learning of foreign investments before they take place, (3) the difficulties the government encounters in ana- lyzing specific investments, and (4) the broader questions not addressed in the present system for analyzing foreign investments. To accomplish this, we reviewed relevant literature and talked with rep resentatives of CFIUSmember agenciesand with academic and industry experts. We focused on the data available in two industry sectors gener- ally agreed to be of national security interest, aerospaceand semicon- ductors, and we examined the CFIUSfries to analyze the types of questions arising in the CFIUSprocessand the availability of data needed to answer these. We did not review CFIUS’decision-making or compliance procedures. We conducted our review from January through October 1989 and updated information in February 1990, in accordancewith generally accepted government auditing standards. As requested, we did not obtain formal agency comments on this report, although we discussedour findings with agency representatives and considered their views in completing this report. This report is one in a series of recent reports on foreign investment.” Page 10 GAO/NSUD9@94 Fore&n Investment Chapter 2 Defining National Security Concernshave arisen about how broadly to define the industries and technologies that are national security related. A narrow definition would include firms that do the majority of their businesswith the Department of Defense(DOD) or as subcontractors to DOD prime contrac- tors. A broader definition would include industries and firms whose bus- iness is driven by the civilian commercial sector but, becauseof their leading edge technologies, are important to overall defense technology leadership. Systematic information is available on foreign ownership of DOD’S direct contractors, but not on foreign ownership of lower-tier con- tractors or of commercially driven firms developing technologies critical to defense. The importance of a broad definition of the defense industrial base and an improved government capability to track developments in this base was highlighted by several recent advisory reports to DOD. For example, the DefenseScienceBoard, in its 1987 report on “Defense Semiconduc- tor Dependency,” noted that, despite the semiconductor sector’s crucial importance to national defense,DOD businessis of relatively small importance to the sector, and advancesin the industry are driven princi- pally by commercial considerations. As a result of basically commercial factors, it noted, U.S. semiconductor technology leadership has deterio- rated to the point where U.S. defensemay becomeunacceptably depen- dent on foreign sourcesfor stateof-the-art semiconductor technology. The Defense Policy Advisory Committee on Trade, in its 1986 policy statement on the defense industrial base, also noted (1) the need for a broader concept of the industrial basethan simple surge capacity in wartime, (2) the inadequacy of DOD funding and procurement by them- selves to ensure sufficient research and development in some sectors, and (3) the need to consider the interrelationships between the indus- trial base and general US. economic and trade policy. The Exon-Florio Amendment did not define “national security,” but the accompanying conference report noted that it is to be interpreted broadly and without limitation to particular industries. The proposed implementing regulations, issued in July 1989, also do not define national security and thus neither encouragenor preclude use of a broad definition of the defense industrial base. In our review of CFIUScases under Exon-Florio, we did not find evidence that the absenceof a spe- cific definition of national security affected CFIUS ability to investigate investments. As described in Chapter 4, CFIUScan review foreign invest- ments in a broad spectrum of industries. . Page 11 GAO/NwADBogQ Foreign Investment Cbpter 2 Deflning~tl~seearity As a result of the 1988 amendment, CFIUS now receives information on specific foreign investments; and, basedon recent experience, it appears that ems will have advance knowledge of a wide range of proposed investments. However, CFIUS is not performing analyses of foreign investment by industry sector. Such analyses, addressing questions about whether sectors of the defense industrial base are being “hollowed out” by foreign investment, are more difficult to perform. One problem is the lack of systematic information identifying key suppliers to firms considered to be in the defense industrial base.DOD can identify its direct contractors but it can- not systematically identify lower-tier suppliers to its direct contractors and, therefore, cannot keep track of changesin their ownership to pro- vide an information baseon foreign acquisitions of these firms. Someof these lower-tier suppliers are commercially driven firms developing technologies valuable to national security. Representatives of several DOD direct contractors in the aerospacesector told us that they do not systematically track foreign investment in their lower-tier suppliers. For example, a representative of one major contrac- tor told us his company is concernedwith quality, price, and delivery experience, and that it could not keep track of changesin ownership in the numerous tiers of its suppliers. For major systems these lower tiers can include thousands of suppliers. The DefenseScienceBoard, in its 1988 study on “The DefenseIndustrial and Technology Base,” found, as well, that neither DOD nor industry has the means of specifically measur- ing the extent of U.S. dependenceon foreign sourcesfor defense components. One important difficulty in analyzing national security effects is in deciding whether foreign acquisition of a U.S. firm that maintains pro- duction in the United States should generate concernsabout foreign dependencesimilar to those raised by U.S. imports. On the one hand, foreign investments can bring new capital and technology to U.S.-based production facilities. On the other hand, foreign control means that deci- sions affecting research, product choice, and plant modernization can be made abroad. Assessingthe foreign owners’ intentions regarding the direction of technology transfer and the continuation of advanced research and development programs in the United States can be very difficult, even on a case-by-casebasis. P8ge 12 GAO/NCiIAD-WM Foreign Investment Chapter 3 Advame Knowledge of Foreign Investments Before the Exon-Florio Amendment was enacted, legitimate questions could be raised about the ability of the government to learn about sensi- tive foreign investments before they occurred. The amendment does not specifically require foreign investors to notify CFIUSof proposed invest- ments in existing U.S. firms having any bearing on national security. However, the consequencesof not notifying CFIUSmay be severe, becausethe investors may risk divestiture orders later. Foreign inves- tors in start-up companies are not subject to the Exon-Florio Amend- ment and are, therefore, excluded from CFIUSreview. Foreign investors in firms performing classified work are required, under the DefenseIndustrial Security Program, to provide advance notice to DOD so that arrangements can be made to isolate or to safe- guard the classified work. Under the ems processprior to the amendment’s enactment, proposed foreign investments in other U.S. firms would come to U.S. government attention only when those firms voluntarily notified CFIUS.They might also have been indirectly referred to CFIUS as a result of (1) filings with the Securities and Exchange Commission, required when a foreign firm, like any U.S. firm, makes stock purchases of 5 percent, a level consid- ered to signify intent to seek control, (2) premerger notifications to the Justice Department and the Federal Trade Commission, which perform antitrust reviews for foreign and domestic acquisitions, and (3) industry or intelligence agency notifications. Before August 1988 the President had the authority to block a foreign acquisition of a U.S. firm through theInternational Emergency Economic Powers Act (50 U.S.C. 1701-06), but only if the President declared a national emergency in responseto an unusual and extraordinary threat to the U.S. national security, for- eign policy, or economy. These types of prior notification, however, could not ensure that all for- eign investments in key, high technology U.S. firms would come to the attention of the U.S. government before completion. Filings with the Securities and Exchange Commission are generally required only for publicly traded companies, not for privately owned fil. Filings for antitrust review are not required for transactions involving smaller firms4 SomeU.S. firms in the semiconductor and aerospacesectors are 4Under the~artsoOtt-Rodino Antitrust Improvements Act, notiaz of the intended acquisition must be madewherethefollowingconditionearemet:(1)atleastonepartytotheacquisitionhas%1oOmil- lionormoreinannualnetsaiesortotalassets,(2)theoWerpartyhasSl0millionormoreinannual net sales or total assets; and (3) the acquiring paty will hold either 16 percent or S15 million of the acquired party’s voting securities or assets (UU.S.C. sec. Ma). P8ge 13 GAO/NSlAD-WM Foreign Investmen iidvmm Knowledge of Foreign Investment small, privately owned companies,and foreign investors would not be required to file with the commission or for antitrust review. Nor would foreign acquisitions of such companies necessarily come to the attention of the government through industry or intelligence sources.Even if for- eign investors complied fully in reporting to the CommerceDepartment under the reporting requirements of the International investment and Trade In Services Survey Act of 1976” , prior notification is not required under the act. The Exon-Florio Amendment does not specifically require foreign inves- tors to notify CFIUSof their investments in national security related firms. However, the consequencesof not notifying CFIUSmay be severe, becausethe parties may risk divestiture orders later. The investment that the President ordered blocked on February 1,1990, was one that had been completed before CFIUSfmished its investigation. Therefore, blocking the investment in this caserequired divestiture. The proposed regulations have been criticized as too broad, because they do not provide a clear definition of national security or the criteria to be used in evaluating proposed foreign investments. Consequently, as noted in public comments filed in responseto the proposed regulations, attorneys representing potential foreign investors feel compelled to clear most foreign investments with CFIUS before completing the transac- tions. Thus, at this point, it appears likely that CFWSwill have advance knowledge of a wide range of proposed investments, although only some will receive its full attention. So far, the vast majority of the approxi- mately 240 investments considered by CFIUSsince enactment of Exon- Florio have been reported to CFIUS voluntarily, rather than referred by interested agencies. One type of foreign investment that is excluded from CFIUSreview is “greenfield” investment, i.e., investment in start-up companies.As Trea- sury stated in its proposed regulations, the Exon-Florio Amendment refers only to mergers, acquisitions, and takeovers of “persons engaged in interstate commerce in the United States. Since ‘greenfield’ invest- ments are not ongoing businessesand therefore not ‘persons engaged in interstate commerce’, they are not subject to [the amendment]“. “This act (P.L. 94-472, JZ2USC. 3101 to 3108, as amended) requires foreign investors to supply information on their ihwtments to the Bureau of Economic Analysis of the Commerce Depanment. Individual investor responses are considered business proprietary information and only aggregated data are publicly released. Page 14 GAO/NSI,4lMW4 Foreign Investment Advance Knowledge of Foreign Invabnen~ This exclusion may result in some national security related investments not being reported to CFKJS. One law firm, commenting on the proposed regulations, noted that “this exemption thus potentially removes the Exon-Florio controls on someof the most advanced technologies being developed today.” While such start-up investments are generally pre- sumed to be beneficial in bringing new capital and technology into the United States, there are some technology transfer concernsabout the possibility that U.S. firms’ most skilled scientists and researchersmay be hired by these new, foreign-owned companies. P8ge 15 GAO- Fodgn Investment Chapter 4 Difficulties in Amilyzing Specific Investments CFWS,under the authority delegated to it by the President, is the focal point for government reviews of specific foreign investments relating to national security. Becausenational security is not specifically defined in law, CFIUScan review investments in a broad spectrum of industry sec- tors. The President’s authority to block an investment, however, is more narrowly defined. CFWShas generally had adequate information to analyze specific foreign investments as provided for by the law. Although official government investment statistics do not provide the type of detailed information about defense relationships and alternate suppliers neededby CFIUSin evaluating individual cases,CFIUS has been able to gather neededinfor- mation from a variety of public and private sources. In caseswhere agency views have initially differed, the key concerns related to difficult questions about the desirability of foreign depen- dence and the potential for technology transfers out of the United States. Many of the investments reviewed by CFIUS have been in the aerospace or semiconductor sectors, and so we were able to identify the types of questions arising in the review processfor these investments and to see how they were answered. CFWSwas created in 1975 by executive order to review foreign invest- Guidelines for CFIUS ments that “might have major implications for United States national Reviews interests.” Before the Exon-Florio Amendment was enacted in 1988, CF+IUSoperated, under the chairmanship of the Treasury Department, on an informal, ad hoc basis, mostly reviewing investments by foreign gov- ernments in US. firms and never formally making negative determina- tions about investments. The Exon-Florio Amendment provided the President with new authority to investigate, block, or suspend foreign mergers, acquisitions, or take- overs that may threaten to impair the national security. It authorizes the President to exercise this blocking authority only if he finds that (1) there is credible evidence that the foreign interest might take action that threatens to impair the national security, and Page 16 GAO/N!3IAD6@64Foreign Investment w-4 Dlfncllltlerln Andping Spedfk lnvestmenta (2) provisions of law, other than the International Emergency Economic Powers Act, do not provide adequate authority to protect the national security. The amendment also formalizes the CFIUSreview processby providing a maximum go-day review of individual foreign investments. This period includes 30 days to determine whether to initiate an investigation, 45 days to complete an investigation, and a final 15 days for the Presi- dent to act. CFIUSissued proposed regulations for implementing the Exon-Florio Amendment in July 1989, but these have not yet been finalized. Both aerospaceand semiconductors are generally recognizedas sectors Government Statistics that drive defense technology and that are crucial to U.S. technological on Aerospace and leadership. Yet, under the U.S. government’s official foreign investment Semiconductors data system (maintained by the CommerceDepartment’s Bureau of Eco- nomic AnalysW ), data cannot be broken out for either of these sectors. Aerospace is included in the larger category of “Other transportation equipment,” and semiconductors fall in the larger “Electronic compo nents and accessories”category. The CommerceDepartment’s Intema- tional Trade Administration collects public announcementsof foreign investments in different sectors (including aerospaceand semiconduc- tors) and publishes the data yearly, but this information is not consid- ered complete and does not show cumulative foreign investment positions in the industry sectors. Industry analysts in Commerce’sInter- national Trade Administration also may follow foreign investment trends, but their focus is primarily on export promotion. The DefenseDepartment, as noted in chapter 2, does not have a compre- hensive database that would show foreign investments in national security related sectors, although it doestrack foreign investments in firms doing classified work through the DefenseIndustrial Security Pro- gram. Defenseis now attempting to improve its knowledge of the extent of dependenceon foreign sourcesin certain weapons systems, but this effort is focused on identifying imports rather than foreign investments. “U.S. government fondgn investment databass a~ discussed in Collection on Foreign Investment in the United States (GAO/NS Page 17 GAO/lSUDWM Foreign Lwmtment Industry analysts in the government and private sector told us that, becauseCommercedata are not useful to them, they rely on private sec- tor data sourceswhen considering investment questions. Someof the private sector data is considered very useful. Foreign direct investments in the US. aerospacesector are generally thought to be minimal, although increasing over the past l-1/2 years. Other types of technology- and production-sharing arrangements are of relatively greater concern to businessand government officials, mostly becauseof their implications for national security and the aerospace industry’s commercial competitiveness. Foreign investments in the semiconductor sector, however, are consid- ered significant and have been the subject of recent scrutiny in the CFTUS process.As a result, the government’s knowledge base in this particular sector has advanced well beyond the officially collected data It is known, for example, that foreign firms have been active in the semicon- ductor materials and equipment sector. For example, key manufacturing processes,such as silicon wafer manufacturing and ceramic packaging, are now dominated by foreignowned firms. Many industry analysts maintain that segmentsof the semiconductor industry have been “hollowed out” incrementally as a result not only of foreign acquisitions but also of US. firms’ difficulties in competing against foreign firms. These commercial competitive factors include superior foreign technologies and product quality, but they also reflect differing U.S. and foreign industry structures. For example, large, verti- calIy integrated Japaneseconglomeratescan better withstand the steep downturns in business cycles that are characteristic of this industry, becauseprofits earned from consumer product sales can sustain the firms’ other operations. Once an investment comesto the attention of CFIUS, either through self- Types of Questions notification or through agency referral, CFWSexamines the investment Raised by CF’IUS to determine whether to initiate a full investigation. For casesselected as warranting a full CFIUSinvestigation, the same types of questions examined in the initial review are pursued in greater depth. CFIUS’main interest in reviewing investments has been to learn the nature of the U.S. firm’s relationship with defense-related work-i.e., whether it performs classified work for DOD, what contracts it may have with DOD, what percent of its production is defense-related, and what Page 18 GAO- Fore&n Investment portion of its research and development is defense related. CFIUStries to learn whether the foreign purchaser would be acquiring (1) sensitive U.S. technology subject to U.S. export licensing, or (2) control over a scarcesupply of goods that bear on national security. It also seeksinfor- mation on the U.S. and foreign firms’ market shares and on the availa- bility of alternate suppliers, both domestic and foreign, for the US. firm’s product. Occasionally, it needsto evaluate allegations regarding unfair trade or anticompetitive practices on the part of the foreign firm. Identifying and assessingthe new owner’s plans for the U.S. firm consti- tutes another set of CFIUSinterests. CFIUS questions foreign investors about their intentions to continue production and research and develop- ment in the United States, to pursue particular product lines, to continue supplying DOD, and to relocate U.S. plants. In somecases,specific techni- cal questions arise regarding the nature or uniquenessof the product or technology in question. Virtually none of the information used in the CFIUS review processcame Sources of Needed from formal U.S. government statistics on foreign direct investment. Information Commercedata, even if it were to exist for aerospaceand semiconduc- tors, do not provide the kind of detail about defense relationships, the firm’s technology, alternate suppliers, or the new owner’s intentions that are needed in the review process. The DefenseDepartment can identify whether the U.S. firm performs classified work, and though it cannot systematically provide informa- tion on lower-tier subcontracting, this can be learned for specific firms. To learn whether there may be “credible evidence” that the foreign interest might take action threatening the national security, Commerce, Defense,and State can search their export control records for licensing and enforcement information, and the intelligence agenciescan be called on to check, for example, for known unauthorized technology transfers. The Justice Department, the Federal Trade Commission, and Commerce can provide information regarding evidence of anticompetitive behavior. Information on alternate suppliers doesnot readily exist in U.S. govem- ment databasesbut can be obtained from government and private sector industry analysts for some specific products. Technical questions that arise in the review processcan be addressedby consulting government and industry analysts and technical specialists. P8ge 19 GAO/Nf3IMMW4 Fondgn Investment CFIUSalso seeksinformation directly from the investing and selling com- panies through correspondenceand discussions,particularly regarding plans for continued US. production and research efforts. In the several CFIUS caseswe examined where agenciesraised serious Difficulties in Deciding questions about the foreign investment, the key concernscould not be Cases allayed by further investment data collection. Rather, the concerns related to the future consequencesof U.S. dependenceon foreign-owned suppliers resulting from reduced US. commercial competitiveness and technology transfers in some sectors. The following casesillustrate the types of concerns which have been raised by CFIUS members both before and after enactment of the E&on-Florio Amendment. Semiconductor The 1989 sale to a West German firm of the last major U.S. merchant Investments producer of silicon wafers--the basic element used in semiconductor chips-raised difficult questions about erosion of important parts of the U.S. semiconductor industry. It was also the first full CFIUSinvestigation to take place under the Exon-Florio Amendment. At the time of the CFTUS investigation, it was well known that the U.S. firm was the last major U.S. merchant producer of silicon wafers (although some large U.S. electronics firms still have captive units that produce for internal use). It was also public knowledge that Japanese and German firms had purchased other U.S. silicon and wafer fabrica- tion firms in recent years and that, with the sale of this U.S. firm, the share of the world market held by U.S.-owned firms would drop from 14 percent to 4 percent, and their share of the U.S. market would drop from 45 percent to 8 percent. The 1987 DefenseScienceBoard study had already pointed out the importance of semiconductor materials and equipment suppliers in determinin g the competitive state of the art in the semiconductor industry. It had also revealed that the United States was leading its closest competitor, Japan, in only 1 of the 14 materials and processing technologies the Board had studied. An additional important issue was the investment’s effect on the newly created business-governmentconsortium Sematech,which Congresshad authorized in 1987 to promote a U.S. capability to manufacture advanced technology semiconductors. The question was whether the United States neededto have a U.S.-ownedfirm supplying Sematech. The U.S. firm was not a large supplier directly to the DefenseDepart- ment, but it was to be a key supplier to Sematech.Questions that arose Pye20 GAO- Foreign Investmenr Chapter 4 Dlfllcnltiea in AndyAng Sped& Invumntr related to the possible technology outflow that might result from a for- eign firm being a supplier to Sematech,the capability of remaining U.S.- owned firms to supply Sematech,the capabilities of alternate foreign suppliers, and the reliability of continued supplies of wafers to Sematech and to other U.S. semiconductor firms. Answers to someof these questions could be provided by private sources,but someques- tions required judgments about the future intentions of foreign firms. The law did not provide a clear basis for blocking the investment, becausethere was no “credible evidence that the foreign interest might take action that threatens to impair the national security”; nor could other laws, such as export control laws, be shown to be inadequate in protecting the national security. The casedid raise questions about what could be cited as credible evidence regarding the future intentions and behavior of foreign investors, particularly those from allied countries. Oncean investment casebecomespublic, it can be awkward for an agency to argue that the foreign fm’s intentions are malevolent, in terms of “threatening” national security, particularly if the foreign firm makes a formal statement of positive intentions. One other important CFIUScasein the semiconductor sector-occurring in late 1986, before the Exon-Florio Amendment was enacted-involved a Japanesefirm’s proposed purchase of a major U.S.-basedsupplier of semiconductors to DOD. The U.S.-basedfirm was already foreign owned (by a French oil field services firm). But its acquisition by a large, verti- cally integrated Japaneseelectronics firm that competed directly with U.S. firms raised questions about technology transfer, reliability of sup ply both to DOD and to U.S. businesses(which included a key U.S. supercomputer firm), and effects on the overall commercial competitive ness of the U.S. semiconductor industry. The U.S.-basedfirm was known to be losing money, however, and the infusion of capital promised by the acquisition could also have the effect of strengthening the company as a supplier to DoD.Onceit became apparent that the investment would be opposed by the Secretary of Commerceon general grounds of unacceptable foreign dependence,the Japanesefirm withdrew its purchase offer. No further U.S. government action was needed (although it is not clear what would have been done, since at that tune there was no specific statutory authority to block a foreign investment). The answers to questions raised in the CFIUSprocesson this casewere, again, not the type to be found in formal U.S. government statistics on P8ge 21 GAO/NsIADBob)l Foreign Invemtment foreign direct investment. They required policy judgments about the consequencesof dependenceon foreign semiconductor firms for both the U.S. civilian and military sectors.Such decisions would require making assumptions about the Japanesefirm’s intentions regarding the market power and technology transfer that it would gain from the acquisition. Aerospace Investments Foreign investments in the aerospacesector have also raised technology transfer questions in the CFIUSprocess.A 1983 aerospacecaseraised important questions about the technology transfer implications of for- eign investment, in terms of foreign accessto controlled, but not classi- fled, U.S. technology and the ability of the export control system to monitor such potential transfers. In this case,a large Japanesefirm was seeking to acquire a U.S. manufacturer of specialty alloys used in air- craft engines. Opponents of the investment argued that since the tech- nology had never been allowed to be exported and was not included in U.S.Japan technology-sharing agreements,the Japanesefirm should not be allowed simply to purchase the manufacturer. Particular questions were raised about the ability of the export control system to protect technology acquired through foreign investments. The export control laws do apply to technology transfers resulting from vis- its of foreign scientists to U.S. plants, but DOD believed that the technol- ogy could be transfered becauseof the difficulty in keeping the Japanesefirm’s scientists isolated from the U.S. firm’s experts. Ques- tions Werealso raised about whether the technology might leak to the Soviet Union becauseof concerns as to whether Japan had procedures to protect the technology. The investment never took place, once DefenseDepartment officials dis- cussedU.S. government concernswith the proposed Japaneseinvestor. But the lack of definitive U.S. policy answers on the technology transfer questions generated an interagency attempt to examine technology transfer issues related to foreign investment. The complexity of the issue and the existence of differing agency viewpoints prevented the interagency group from issuing joint findings or recommendations.A similar attempt to examine technology transfer issueswas made by the National Security Council staff in 1986-87but they, too, were unable to come to specific conclusions. National security concerns about technology transfer and reliability of supply were raised again in the 1984 acquisition by Japan’s largest ball- bearing manufacturer of a U.S. firm considered to be one of the most Pye 22 GAO/NSLADW94 Foreign Investment important suppliers for missile guidance systems and for major military contractors. Debate about this investment included congressionalhear- ings, and concerns were allayed by the Japanesecompany’s stated com- mitments to continue investments and research in the plant and to fulfill military contracts. A 1989 aerospacecase,for which a full CFIUS investigation took place, also focused on technology transfer concerns.In this case,a French firm was seeking to buy divisions of a major U.S. aerospacefirm whose main business involved producing technologically sensitive items for DOD and the National Aeronautics and SpaceAdministration under classified con- tracts. Technology transfer concernswere allayed by obtaining from the French firm a detailed plan for observing U.S. export control laws, in addition to limiting foreign accessto classified information through DOD’S Industrial Security Program. The one investment that the President ordered blocked was the acquisi- tion of a small, privately owned U.S. aircraft parts producer by a Chi- nese company controlled by the Chinese aerospaceministry. This investment was different from these other c~lt~scases,in the sensethat (1) the foreign investor was controlled by the government of a non-allied country, and (2) a past record of the investor’s objectionable activities could be documented, based on classified intelligence reporting, to sat- isfy the “credible evidence” requirement. P8ge 22 GAO- Foreign hmmtment Chapter 5 Broader Policy QuestionsPersist The vast majority of casesconsidered by CFTJS, and of foreign takeovers of U.S. firms in general, have been friendly, not hostile. In someof these takeovers, the U.S. firm had initially sought a domestic buyer, but with- out success.In the semiconductor sector, it appears that the foreign investing firms were interested in acquiring the U.S. firms becausethey would fit into their more comprehensive product structure and would also enhance their U.S. market position. Someof the U.S. firms selling out were discontinuing their efforts in the industry segment as a result of low profitability, market uncertainties, or corporate realignments, and the foreign purchaser did assure continued U.S.-basedproduction. Businessconsiderations, understandably, are of primary concern to managers of U.S. firms. National security questions about specific investments are raised, rather, by government policymakers who try to view the investment from the larger perspective of the country’s defense needs.Sometimes these questions are raised only after a sensitive foreign investment is publicly proposed and CFIUSbegins a review of it or after a large seg- ment of the industry has already been acquired by foreign firms. CFXJS’ reactive, case-bycase approach does bring out certain investment issues,but it is focused on developing information to meet the law’s spe- cific requirements. The President’s recent decision to block an invest- ment illustrates that these requirements can be met for certain investments. But CFIUSis not presently set up to examine other larger questions, such as (1) which industry sectors, technologies, or types of firms to preserve for U.S. ownership, (2) how to evaluate the importance of a U.S. firm’s commercial strength to its defense role, (3) why U.S. companies have found it desirable to discontinue operations in certain sensitive sectors, (4) why foreign investors place more value than U.S. firms do on devel- oping predominance in certain high technology sectors, and (5) what parts of the US. defense industrial base are being “hollowed out,” as foreign investors acquire U.S. firms which are lower-tier suppliers to DOD. For example, the concerns raised with us by semiconductor industry representatives and industry analysts, as noted below, are not the types of questions on which the CFVJS processhas focused. They are not focused strictly on the firms’ relationships with DOD. Rather, they reflect concerns about preserving long-term U.S. commercial competitiveness - in the face of foreign firms’ competitive businesstactics - in order, Page 24 GAO/NSID-SO-94 Foreign Investment over the longer run, to preserve a U.S. production base available for defense needs. These industry representatives and analysts expressedthe following concerns about foreign competitors’ practices. They also raised ques- tions about what, if anything, could be done, since the practices are not specifically prohibited under U.S. laws. Foreign investors in U.S. semiconductor firms, particularly the Japa- nese,tend to substitute their own suppliers for U.S. suppliers once the firm is bought. U.S. equipment manufacturers are typically small, but their products are high in value, and so the loss of even one customer is considered important. U.S. computer fm’ increasing dependenceon component parts sup plied by their more vertically integrated Japanesecompetitors raises questions about whether the most advanced design components would be supplied in a timely manner to the U.S. firms. Foreign investors have bought U.S. firms not only to enhancetheir own position and acquire technology but also to deny their U.S. competitors accessto this technology. The relatively high cost of capital in the United States, together with the increasingly capital-intensive nature of the semiconductor industry, has led U.S. start-up companiesto seek foreign funding or businessalliances that may entail the transfer of commercially valuable technology. It is difficult to track systematically the extent to which research and development efforts of high-tech firms are continued, once a U.S. firm is acquired. Semiconductor industry concernsremain unanswered, however, about how foreign investments may contribute to the continued erosion of the U.S. semiconductor infrastructure. One recent, proposed Japaneseacqui- sition of a U.S. supplier of critical semiconductor materials illustrates that the current CFIWprocessdoes not resolve commercial concerns. This acquisition, involving a mJor U.S. supplier of certain semiconduc- tor materials, underwent the standard 30day initial CFIUSreview, but no CFIUSmember requested that it move to the 46day investigation phase. The U.S. firm was not a direct contractor to Don,although it was a sub contractor on unclassified work. Alternative foreign or domestic suppli- ers were available, and there was no evidence the specific buyer would threaten national security. The US. firm neededcapital to redesign equipment to remain competitive and continue research, and no other U.S. firms had shown interest in acquiring it. One m participant Page 26 oAo/NsuMnrer PondgIl lnvelltme!nt Chapter 5 Broader Policy Questions Pembt noted that, in his view, the casepresented a choice between two undesir- able alternatives: (1) foreign ownership, which could mean that funda- mental scienceand technology would shift offshore; or (2) continued businessdifficulties for the U.S. firm, which would prevent it from aggressively pursuing the next generation of processingequipment. Although such specific casesmay not individually present a threat to national security, the overall decline of U.S. commercial competitiveness in some semiconductor subsectorsdoes generate broader concernsabout preserving the U.S.-owned production base. To some analysts, the commercial health of high-technology sectors is at the heart of the national security issue. The 1987 DefenseScienceBoard study on semiconductors emphasizedthe importance of U.S. leadership in semiconductor technology to the electronic systems underpinning U.S. military systems. The board noted that leadership in semiconductor technology depends on high-volume commercial production becauseof the need for commercially derived profits to fund continued research and development and new investments. It also noted that high-volume commercial production itself dependson maintaining US. and world market shares. Yet neither Defensenor any other U.S. government agency has responsibility for assuring the commercial viability of the U.S. semiconductor industry. As the board concluded, “... the principal factor affecting the relative shift in strength of the U.S. and Japa- nese semiconductor industries is the fact the Japanese established a strategic (long term) goal and effectively brought together all the resources from government, industry and academia, needed to pursue that goal. The U.S., at its own discretion, elected not to pursue such an organized focus and structure, and as a result is find- ing that it is unable to compete in the marketplace as it has been defined by the Japanese.” Conclusion A strong US. economic baseoverall, and particularly in high technology, is widely viewed as important to national security and to the military defense structure. Congresshas taken steps to promote U.S. high tech- nology leadership, such as helping to fund the research and develop ment activities of Sematechand funding DOD research contracts in high technology fields. Concern persists about foreign investment, however, becauseforeign purchases (1) tend to confirm the existing perception that foreign firms value these high-technology sectors more highly than U.S. firms do, and Page 20 GAO/N3LW-W94 Foreign Investment (2) tend to raise larger questions about the adequacy of U.S. government policy in addressing the variety of factors affecting US. commercial competitiveness in sectors critical to national security. In essence,concern about the national security effects of foreign invest- ment is part of the ongoing public policy debate on the need to preserve US. commercial competitiveness in certain sectors. The link between national security and commercial competitiveness concernsis evident if it is recognizedthat the defense industrial base does include industries whose viability dependson commercial successin U.S. and world markets, systemic factors-such as the target&, national development strategies of U.S. competitors, the vertically integmted structure of Japanesesemi- conductor conglomerates,and differences in costs of capital between countries-in addition to specific trade practices, can affect the com- mercial competitiveness of U.S. firms, and . U.S. government agenciescovering trade and investment matters are oriented to protecting against certain unfair trade practices or specific investments clearly threatening national security, but they do not have any directive to maintain the commercial viability of an industry sector. In this sense,preventing the erosion of defense technology leadership is more than a matter of reviewing foreign investments. CFIUSis carrying out its foreign investment reviews as provided for by the Exon-Florio Amendment, but it cannot be expected to provide answers to the types of questions about preserving commercial competi- tiveness that need to be addressedat higher policy-making levels. P8lgC27 GAO/NSMDWM Fodga lnveatmeat Appendix 1 Major Contributors to This Report bow, Assistant Director National Security and ~~~~*~~~Lm-2,arvps\‘u phe9 *.u.uLU~‘* Prninet Manatinr International Affairs Division, Washington, DC. San Francisco Evelyn E. Aquino, Evaluator Regional Office Robert R. Tomcho. Evaluator Seattle Regional Office ThommL. Ktiti, EVtiuatir David L. Hilliard, Evaluator PIOC 28 GAOflW Foreign Investment
Foreign Investment: Analyzing National Security Concerns
Published by the Government Accountability Office on 1990-03-29.
Below is a raw (and likely hideous) rendition of the original report. (PDF)