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.
Overview Although the competitiveness of key industries most notably, semiconduc- tor manufacturing has seriously declined, the computer sector as a whole retains much vitality. U.S. firms dominate the fastest growing, and perhaps most lucrative, markets for computer-related goods and services. For example, more than 60 percent of the world software market, totaling about $65 billion in 1989 and growing at an annual rate of about 25 percent, is controlled by U.S.- based suppliers.! 2 Similarly, domestic enterprises and their overseas sub- sidiaries reap the majority of the global revenues earned for designing and inte- grating computer systems, a market whose annual value colloquium participants estimated to be between $25 billion and $40 billion. Finally, U.S. computer manufacturers are by far the world's largest producers of computing and periph- eral equipment; in 1989 they earned approximately half of the $134 billion gen- erated worldwide by the production of computer systems per se and perhaps 45 percent of the $300 billion in total global revenues when systems software, ser- vice, maintenance, and leasing are included.3 4 This position reflects, among other things, U.S. strengths in relatively large and advanced-technology com- puters. Whether this nation retains a vigorous presence in computer-related technologies depends on how it responds to changing terms of competition in a global economy. For the U.S. computer sector, the rapid rate of technological change and the direction of that change have been key determinants of its competitive position. Recognizing that technological leadership does not automatically transfer from one generation of products to the next, the U.S. computer industry has tradition- ally invested heavily in research and development. In a recent study of 897 5
6 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE firms in over 40 industries, computer manufacturers averaged the highest investment as a percent of sales (8.2 percent) for all industries except one, soft- ware, which as an industry invested an average of 13.3 percent of sales revenues into R&D activities.5 High rates of technological advance, a hallmark of computer manufacturing, "change the structure of the industry in spans of 10 years or so," explained col- loquium participant Richard S. Rosenbloom, professor at Harvard University's Graduate School of Business Administration. And the commercial manifesta- tions of the change often defy prediction. Underscoring his point, Rosenbloom noted that a major 1977 study on the future of the computer industry addressed the development of microprocessors and large-scale integration but did not anticipate the impact of personal computers. Rapid rates of technological change in computer hardware have been used as a competitive advantage by U.S. computer makers. U.S. manufacturers of workstations, for example, have managed to stave off foreign competitors in the $6 billion market by rapidly incorporating the results of research into their prod- ucts. According to the U.S. Department of Commerce, U.S. makers of "low- end" workstations can now develop and introduce new products in only six months, while "high-end" manufacturers have reduced this period to two years.6 In so doing, U.S. firms have maintained a two-generation lead over Japanese firms vying in the workstation market. If trends that are already well established in the computer-manufacturing base continue, however, much, if not all, of the U.S. computer sector may falter. Even such value-added activities as software development, system design and integration, and after-sale maintenance may be in jeopardy. "Falling behind in computer technology . . . ," said Lawrence G. Tester, vice president of advanced technology at Apple Computer, Inc., "is a very serious issue. The country got very galvanized about the idea of falling behind in space exploration 30 years ago, and it's just as important today that we do not fall behind in computer tech- nology." Yet, according to Tester, "We are losing our lead in software . . . slow- ly." Jeffrey M. Helter, senior vice president and head of the technical services group at Electronic Data Systems, explained that the individual industries are interdependent, and so their fates are linked. "Our business [services and sys- tems integration]," Heller said, "cannot exist on a long-term basis successfully without a healthy domestic industry in both the hardware and software busi- ness." Meanwhile, political and economic developments worldwide are shaping computer-related markets of tomorrow. In particular, with the integration of the European Economic Community in 1992, Western Europe, with a population of about 320 million, will surpass the United States as the world's largest market. Four decades ago, the U.S. market was eight times larger than the next biggest, providing the nation's manufacturers with a large "home audience" for their
OVERVIEW 7 products. Numbering more than 240 million, American consumers still are an attractive market but have already shown an increasingly strong willingness to buy foreign-made goods. As a result, U.S. firms, no longer assured of domi- nance in domestic markets, must compete for sales at home and abroad. GLOBALIZATION AND COMPETITIVENESS "Technology is a fluid," said John Doyle, executive vice president for busi- ness and development at Hewlett Packard. "It will go anywhere that it is needed" Most computer-related markets have been international from the outset, and this has influenced policymaking in other countries. By the early 1960s, the nation's infant computer industry was already active in foreign markets, earning about 25 percent of its total revenues from sales abroad.7 The success of world leader IBM, in particular, was a major catalyst in the decisions of foreign gov- ernments to shepherd the development of their fledgling computer industries. In 1966, for example, Japan's Ministry of International Trade and Industry des- ignated the growth of the domestic computer industry as the number one priori- ty for future economic growth. Not coincidentally, U.S. computer makers' share of the Japanese market began to fall, a decline abetted by formal and informal trade barriers. In the 1960s, sales of IBM computer equipment accounted for about 40 percent of the Japanese market; today, IBM's share is less than 15 percent.8 Starting in the late 1970s, the Japanese share of the U.S. market increased dramatically, and within a decade, the United States had a $4 billion deficit in computer trade with Japan. Computers are perhaps the best and most tangible evidence of a world econ- omy. Generally, machines "made in the United States" are made from compo- nents, subassemblies, and assemblies manufactured by many different domestic and foreign firms, including offshore subsidiaries of U.S.-based companies. In its most extreme form, this internationalization of the computer has resulted in entirely foreign-built machines that are sold under the label of an American company. Several colloquium participants referred to cost advantages and other incen- tives motivating U.S.-headquartered firms to establish manufacturing and other operations overseas. "I'm told that, if I want to build a semiconductor fabrica- tor," explained Patrick A. Toole, senior vice president and general manager of technology products at IBM Corporation, "I can get practically everything paid for in some countries, like Singapore. You can get long-term tax incentives, you can get training assistance, you can get them to pay for all the tools.... That is their national policy with regard to growing and engaging in an emerg ing industry." Although embodying different mechanisms, policies associated with integra- tion of the European Economic Community will achieve similar results, pointed out Gordon E. Moore, chairman of the board at Intel Corporation, a U.S. semi
8 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE conductor manufacturer. In addition to eliminating national trade barriers between the 12 member countries, "Europe '92" will establish rules of origin and local content requirements that will require today's exporters of integrated circuits to build the devices in Europe. "Intel avoided for 20 years doing any manufacturing in Europe, because it Edid] not make sense," Moore said. Now, the company has "two teams running all over the continent, looking for the place we ought to build our plants. We export over a half a billion dollars Lof products] to Europe every year. We are going to add a half billion dollars to our trade deficit because of '1992."' "What we are looking at is not 'fortress Europe,' but 'magnet Europe,"' added Hewlett Packard's Doyle. C. Gordon Bell, vice president for research and development at Stardent Computer, Inc., a U.S.-based, Japanese-f~nanced manufacturer of workstations, suggested that superior manufacturing skills available in foreign nations also motivate offshore location of plants. "The bottom line why we manufacture in Japan is, we are trying to transfer manufacturing technology to the United States, and we want quality," Bell said. Benefiting from the resources and support of overseas environments is nei- ther automatic nor enduring. IBM's Toole noted, for example, that a company that subcontracts too much of its manufacturing to offshore plants "will soon lose the expertise to design and the ability to innovate, because it won't get the feedback it needs." There are strategic dangers as well. Toole said that an underlying motive of host countries is to acquire the technology, and, he added, "it is not long before a competitor has been developed." Moreover, the cost advantages of offshore manufacturing can be transitory. A strong dollar, local wage levels, and the absence of trade barriers are strong incentives for offshore manufacturing, according to Toole, but"none of these factors is within a com- pany's control." Nevertheless, most colloquium participants said they expected offshore assembly of computer hardware to increase. Similarly, U.S.-headquartered companies are expected to invest more in foreign R&D. According to prelimi- nary figures from the National Science Foundation (NSF), U.S. fines invested $5 billion in overseas research facilities in 1987, up 6.4 percent from the previ- ous year.9 U.S. computer manufacturers appear to be among the leaders in this trend. For example, IBM has two of its four major research centers overseas, and Hewlett Packard's four-year-old Integrated Systems Center in Bristol, England, accounts for about a third of that firm's central research effort. William J. Spencer, vice president and head of the corporate research group at Xerox Corporation, described Hewlett Packard's move as an "example of a major company taking a strategic view in its R&D investment." At the European facility, Hewlett Packard can `'take advantage of a large cadre of high- ly ted en ted people and an ability to build-add local value to compete effective- ly in Europe," Spencer said.
OVERVIEW 9 Tapping foreign talent and technology is essential for firms that do business in international markets, and many companies see their investments in overseas R&D as a necessary ante for full acceptance in foreign markets, according to observers. These investments also help counter the so-called "not-invented- here syndrome," the parochial view that all innovations originate in the United States. "The post-World-War-II era, when it was just automatically assumed that the United States led the world in everything, is over," NSF Deputy Directory John H. Moore has said.~° "We need to wake up to the fact that there's a lot of good work going on elsewhere in the world. And we need to appreciate that fact and take advantage of it." Investments overseas are also expanding beyond research and manufactur- ing, integrating the operations of firms headquartered in the United States more completely into the fabric of foreign economies. As Toole of IBM explained, "It is very important for U.S. computer companies to establish and grow market share in all of the markets where their products have a value to prospective companies. In the computer industry, it is extremely difficult to survive for long in either a market niche or a single geographical area orientation." What is perhaps confusing to outside observers is how these international linkages are blurring the traditional concept of a "U.S. firm." Is a firm with its headquarters in the United States but with most of its capital assets and employ- ees overseas truly a U.S. company? How does that firm differ from a U.S.- located subsidiary of a foreign company? Obviously, the answers to these ques- tions have important implications for federal policy and for business. Colloquium participants grappled with these questions, offering a variety of opinions of what constitutes a U.S. firm (see box). Despite the diversity of views, most agreed that for a company to be considered a U.S. firm, it must contribute to building capital and raising the standard of living in this nation. One way to define the U.S. benefit is the value added by capital, management, design, engineering, manufacturing, distribution, sales, or service functions per- formed by people from the United States. In a heavily interdependent set of worldwide industries, a given country's benefit is likely to be measured by the value added, which supports employees, taxes, infrastructure, reinvestment, and ultimately the standard of living. COOPERATION Colloquium participants repeatedly addressed the issue of cooperation- cooperation between firms and among universities, government, and industry. Cooperative efforts abroad clearly stimulated examination of cooperation, but there was a strong perception that emerging technological and market chal- lenges are creating a new mandate for cooperation in this country. "Japanese companies, in particular, have mastered the art of cooperating in order to compete," explained colloquium participant David J. Teece, professor
10 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE What is a U.S. Firm? Answers from Colloquium Participants A reasonable definition [of a U.S. firms for many purposes would be one with significant U.S. ownership and a fair share of technology manufac- turing and development in the United States. [A U.S. firm is] one where the majority of the board, top management and owners take a U.S.-centered view of the welfare of the firm (whether it is multinational or not) and can be expected to consider the effects of their actions on the welfare of the United States first. It is clear we know what a foreign firm is since we keep Philips (Signetics) out of SEMATECH. U.S. fops are the others. What is meant, will be meant, should be meant are all important ques- tions. The current connotation of a U.S. firm is one that is incorporated in the United States and has a plurality of its business interest in the United States or one whose tax payments are significant to the United States. A suggestion: A U.S. few is one that is important to U.S. society and con- tributes to the well-being of the United States. A U.S. fun is a company incorporated and headquartered in the United States, which designs, builds, and sells computers on a worldwide basis. Company control is with U.S. nationals. Design is within the United States to exploit particular technology skills. Financing is increasingly likely to come from outside the United States given the large supply of "cheap" dollars. Manufacturing may be done outside the United States using manufacturing, manufacturing-engineering, and assembly skills and foreign capital. U.S. computer companies (and other large companies) become increas- ingly "global" in their operation and development/manufacturing, even research efforts, in addition to their traditional worldwide marketing. I would call a company a "U.S. Fox" as opposed to Japanese, or German, [for example,] if the majority of ownership is U.S. based and, thus, its profits contribute to building capital and the standard of living in this country. Today's products and services are marbled with international involve- ment. A high-tech product may well be designed in London, the software developed in France, the parts fabricated in Taiwan and the entire product assembled in Torino for sale in the United States. Which is more German,
OVERVIEW for example-a Volkswagen produced in Rio, or a Ford made in Stuttgart? Is the Nobel Prize awarded to a pair of Swiss scientists working for an American company in Lucerne a source of pride for Switzerland, or for the United States? Or both? In today's global marketplace, ownership is often irrelevant as a factor in a company's success. What really matters are the kinds of contributions a fun can make to the localities where it resides. [I]n the rapid internationalization of business it is also important to ask, "What is the United States?" It is a country almost entirely populated by people who came from around the world to better their life. With the coming decline in work force size and existing deficiencies in U.S. educa- tion, we should be encouraging immigration of educated people from all countries. It is no longer meaningful to talk of U.S. fobs. It may be somewhat more meaningful to talk of "U.S.-based funs," if their headquarters are located in the United States, and at least half of their business is conducted there. The fact that the classic distinctions between domestic and foreign firms has all but disappeared raises important policy questions. Policies that are aimed to help "U.S. firms" can no longer be easily implemented. Accordingly, it is probably necessary to think about policies in terms of their impact on U.S. economic activity, rather than on U.S. fobs. 11 of business administration and director of the Center for Research on Management at the University of California, Berkeley. Teece continued, "There has been, and there remains, a recognition that by acting together, sometimes with government assistance but without governmental barriers Japanese firms can do better than they might by competing alone. U.S. firms must also recognize that cooperation can be used to catch up with competitors, stay ahead of competitors, and respond to foreign industrial policies." Sometimes called "alliance capitalism," Japan's efforts have inspired cooper- ative projects in many nations, including the United States. Some observers, like Teece, would argue that Japan's success makes cooperation an almost mandatory condition for long-term survival in global markets. Two corollaries follow from this argument. First, U.S. firms and the federal government must monitor the progress of the growing number of cooperative R&D projects under way overseas. Second, the government must be vigilant about trade barriers and other protectionist measures that other nations may take to nurture the growth of their computer industries. Many colloquium participants questioned whether some of the relatively
12 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE recent foreign cooperative research efforts, such as Japan's Fifth-Generation Computer project, would translate into commercial successes for participating firms. "No doubt exists whatsoever regarding the seriousness behind these efforts," said Sam R. Willcoxon, president of AT&T business markets group. "Whether they're a challenge remains to be seen." Collective industrial research projects in the U.S. computer sector are a recent development. Their formation was shaped by widening recognition that the complexity, scale, and risk of advanced computer-related technologies increasingly demand more resources than individual fops, even market leaders, can afford. Industry representatives at the colloquium commented most on SEMATECH, the semiconductor-manufacturing technology consortium jointly funded by private industry and the Department of Defense. They expressed optimism, but they noted that it was premature to judge the performance of this and other cooperative efforts in the computer sector. Regarding such efforts in general, Gordon Bell of Stardent Computer noted that cooperative research pro- jects attract and sustain scientific and engineering talent. While joint R&D projects have a longer history in this country than is typi- cally recognized, they represent only one realization of the cooperative links that must extend horizontally and vertically from individual firms, according to IBM's Toole. "We must improve cooperation at levels within companies- between functions such as marketing, manufacturing, development, and research," Toole said. Relationships must also be developed with suppliers, cus- tomers, and, "more importantly these days, competitors." Toole also extended his web of industrial cooperation to encompass universities and government agencies. "With the fierce competition we face throughout the world, we can no longer cling to traditional . . . arms-length, business-as-usual dealings," Toole said. "We must find new ways and new areas in which to cooperate at all levels." The prospects for doing this may be poor, several speakers suggested that U.S. firms, grounded in the ideals of free enterprise and competition, still do not know how to cooperate. STANDARDIZATION Standardization, the process of specifying technical requirements for com- puter hardware, software, and networking, has emerged as a strategic competi- tive issue. The market-shaping power of standards in the computer sector was first demonstrated 25 years ago. In 1964, IBM folly entrenched itself as the world's leading computer manufacturer by establishing its own internal standard for computer components and peripherals, which was showcased in its 360 System of compatible computers. Software developed for one computer model in the family would run on any other machine in the IBM product line. This compatibility assured IBM's customers that all of their computer equipment
OVERVIEW 13 would perform in essentially the same manner and that their hardware and soft- ware would not become obsolete with the next series of new product introduc- tions. For IBM, standardization permitted the company to spread its R&D costs over a wide array of products that served a large-volume market. By virtue of IBM's large market share, the company's proprietary standard became the industry standard. But because IBM controlled the standard, it could make changes, making it difficult for competitors to overtake the industry leader. Today, with the proliferation of computing equipment and applications, there is a well-accepted need to develop standards that attain at least a certain level of compatibility and connectivity among the machines of different manufacturers. Indeed, more than 1,000 standards pertaining to computer-related technology have either been adopted or are under consideration by national and internation- al standards-setting bodies.12 Computer-related firms throughout the world are keenly interested in the development of these nonproprietary standards, with many trying to influence the outcome of the decisions. Some firms (e.g., Sun Microsystems), for exam- ple, seek to have their own standards adopted by the entire industry. But debate is not limited to specific standards; it surrounds the entire issue of standardiza- tion. Colloquium participants also engaged in the debate. At issue is where to draw the line between necessary standardization and excessive standardization. Virtually all agreed that standards are necessary for data storage and exchange and for communication. Beyond that point at the level of operating systems and hardware elements the consensus deteriorated. The relevant issues will be discussed in greater detail in each of the following chapters; the costs and bene- fits from the manufacturer's perspective will be summarized only briefly here. Standardization expands markets and lowers costs. Moreover, by providing users with a greater array of compatible machines from which to choose, it increases competition. On the negative side, standardization, particularly if it is done too early in a technology's evolution, can freeze the technology at a "prim- itive stage," said Apple Computer's Tester. In addition, Tester and others noted that standardization can eliminate innovation as a prerequisite for entry into computer markets, thus placing greater emphasis on high-volume, low-cost pro- duction-a strength of Asian competitors. It also places a premium on protect- ing the intellectual property embodied in whatever innovative elements are pre- sent. Adequate safeguards (e.g., patents or copyrights) must be maintained and enforced, especially internationally, to ensure that the innovator receives appro- priate return on investment. 13 Given the global push for standardization, all colloquium participants agreed, the issue cannot be ignored. U.S. firms, they said, must confront the question of what level of standardization is best, or they risk having the matter settled by the competition.
14 KEEPING THE U.S. COMPUTER INDUSTRY COMPETITIVE THE GOVERNMENT ROLE Perhaps the biggest gap in cooperation to bridge is the one between govern- ment and industry. Here, too, the nature of the relationship remains to be defined, even though the need for more constructive links between the two has been emphasized in nearly every one of the numerous studies that have exam- ined the flagging competitiveness of American industry. Ironically, the U.S. computer sector is a creature of government-industry cooperation, motivated by a strategic interest, namely, national security. During and after World War II, the Department of Defense funded well over half of the research that resulted in the first electronic computers, and it supported much of the work that led to subsequent refinements and innovations that made the machines commercial products. As Kenneth namm has noted, only 4 of the 12 major U.S. producers of digital computers in 1956 exist today.14 Yet the techni- cal roots of many of today's firms, old and new, he writes, "can be traced to experience accumulated in the pioneering days of computing" when almost all R&D was sponsored by the nation's military services. A congressional commit- tee estimated that in 1959, 85 percent of electronics R&D in the United States was funded by the federal govemment.lS By 1986 federal support had slipped, accounting for less than 30 percent of the funding for U.S. electronics R&D.16 More than half of IBM's R&D during the 1950s was conducted under govern- ment contract, and even after the commercialization of computers had begun, over a third was govemment-supported in 1963.17 The U.S. govemment's influence extends over many of the important techno- logical developments in computer hardware and software, sometimes creating marketing niches effectively exploited by IBM's U.S.-based competitors. For example, time-sharing, a product of work supported by what is now known as the Defense Advanced Research Projects Agency, was a boon to the Digital Equipment Corporation, whose PDP-6 computer was the first commercially available machine to offer the feature. Today the federal government, primarily through the Department of Defense, continues to fund much of the basic research in computer science and engineering, such as studies of gallium arsenide and other nonsilicon semiconducting materials. Some of today's lead- ing-edge computing technologies among them, parallel processing and artifi- cial intelligence are outgrowths of military-supported projects. This situation reflects the strong dependence of U.S. national security on advanced technolo gY Although the precedent of cooperation between the government and the com- puter sector was established long ago, and the government continues to fund basic research important to the sector, the relationship that has evolved is not commensurate with changing global conditions. Indeed, many colloquium speakers observed that the cornerstone of this relationshi~military sponsor- ship of computer-related R&D is much weaker than it was two or three
OVERVIEW 15 decades ago, while at the same time it is not sufficiently complemented by civil- ian support. "In the absence of a national appreciation of the developing computer role," said Alan J. Perlis, professor of computer science at Yale University, "the sup- port by the military of computer development has been essential and far- sighted." Yet Perlis and others noted the divergence of the miluary's needs and interests such as "radiation hardening" of electronic devices from those of the marketplace. "There has been too little effect on our economy, our society, our industry, and our quality," Perlis said. Military sponsorship tends to channel "computer development in far too narrow a way and prevents our society from adequately exploiting its potential." In addition, factors crucial to success in commercial markets are not often high priorities in the development of products for the military. "There is a prob- lem with using military funding to drive the direction of computer innovation," said James H. Morris, professor of computer science at Carnegie Mellon University. "People who work on such projects are force-fed with money and never get a feeling for costs or markets." Further skewing the government-industry relationship is the fragmented reg- ulatory and policymaking structure of the federal government. In the executive branch, as many as 12 federal agencies help shape and carry out policies that influence the activities of firms in the U.S. computer sector. In Congress, 9 of the 13 appropriations subcommittees are involved in determining the research budget and specific allocations. This fragmentation and the lack of a coherent perspective on the cumulative effect of actions taken by widely dispersed units within the federal bureaucracy pose serious obstacles to achieving the coordina- tion and cooperation that colloquium participants, as well as the Council on Competitiveness and other groups, argue is needed. "Those who are likely to be in a position to implement any remedies," said Robert M. White, president of the National Academy of Engineering, "are found in many different organizations spread across the federal government and private industry. They deal in relative isolation with many separate and relative issues whether in trade policy, tax policy, procurement policy, employment and training programs, [or] research support. Until there is a meaningful interac- tion, it will be difficult to fashion a nationally coherent or coordinated response to the challenge that is now facing the U.S. computer sector." These factors may contribute to the perception, expressed by Xerox's Spencer, that"cooperation between government and industry [seems like] an unnatural act. It seems so difficult for us to do." By contrast, broad-based cooperation between government and industry has been a boon to the computer sectors of other nations. '~Foreign industrial policy is affecting the outcomes in American markets all the time," said Teece of the University of California, Berkeley. To assume that market and price mecha- nisms guide firms to pursue individual optima that result in a maximum benefit
16 KEEPING TlIE U.S. COMPUTER INDUSTRY COMPETITIVE for the nation-an assumption that, Teece said, "informs public policy in this nation today" is to ignore other important factors at work, including actions of foreign governments. Some colloquium participants suggested that the nation has a government- directed industrial policy, in the form of a composite of fragmented decisions in areas ranging from defense spending and overall government procurement to antitrust laws. Moreover, several studies suggest that the success of strategical- ly chosen industrial sectors in nations with such policies stems from factors other than direct government intervention.19 While it can be difficult to sort out the nature and impact of government actions in the United States as well as abroad, the computer sector must not be a passive observer in the policymaking process. Clyde V. Prestowitz, Jr., princi- pal adviser on Japan affairs to the U.S. secretary of commerce from 1983 to 1986, said firms should be making greater efforts to present their views and sug- gestions to Congress and federal agencies. Patrick Toole of IBM noted, `'We have to do it not only by meetings like this, but also by very strong industry associations." "The main thing that makes the computer industry different from other industries is that the rapid change of the technology far exceeds the policymak- ers' ability to change," argued Heller of Electronic Data Systems. He suggested that this rigidity is the legacy of institutions shaped by concepts of law, educa- tion, work, and accounting that evolved during the industrial age. Cautioned Helter, "The heavily entrenched, industrial structural model works poorly in a service society and worse in an information-based world." NOTES 1. 2. 3. 4. U.S. Department of Commerce. "Computer Equipment and Software," 1989 Up. Industrial Outlook (Washington, D.C., 1989~; 1989 world market estimate provided via personal communication from a Department of Commerce analyst. Quantitative estimates within the computer sector are particularly sensitive to defi- nitions and methodologies. Comparisons should be made with caution as different estimates seldom, if ever, measure the same market with the same methodology. The numbers used in this report are generally as reported by the Department of Commerce, supplemented where appropriate by industry and trade group data. Gartner Group, Inc., Stamford, CT, 1989. CBEMA. Information Technology Industry Global Market Analysis, Industry Marketing Statistics Committee, Washington, D.C., l9B9. 5. Business Week: Innovation in America, October, 1989, pp. 177-228. 6. U.S. Department of Commerce, 1989 U5. Industrial Outlook, 1989, p. 26-11. 7. Flamm, Kenneth. Creating the Computer (Washington, D.C.: Brookings Institution, 1988), p. 101. 8. Dertouzos, Michael L., Richard K. Lester, and Robert M. Solow, MIT Commission on Industrial Productivity. Made in America: Regaining the Productive Edge (Cambridge, Mass.: MIT Press, 1989), pp. 262-263.
OVERVIEW 17 9. Buderi, Robert. 6'U.S. Companies Hike Investment in Foreign R&D," The Scientist, May 29, 1989, p. 1. 10. Buderi, The Scientist, 1989, pp. 1, 6. 11. One that beers watching, according to Stardent's Bell, is Japan's SIGMA project (see Chapter 3 of this report). 12. Gantz, John. "Standards: What they are. What they aren't," Networking Management, May, 1989, p. 23. 13. A Computer Science and Technology Board report on issues in the protection of intellectual property in software is anticipated in mid-1990. 14. Kenneth Flamm, Creating the Computer, 1988, p. 81. 15. Kenneth Flamm, Creating the Computer, 1988, p. 16. 16. American Electronics Association, 1989, as reported in The Competitive Status of the U.S. Electronics Industry Sector, U.S. Department of Commerce, 1989 (draft version). 17. American Electronics Association, in The Competitive Status of the Up. Electronics Industry Sector, 1989, p. 94. 18. Council on Competitiveness. Picking Up the Pace: The Commercial Challenge to American Innovation (Washington, D.C., 1988~. 19. "Even in Japan and France, countries where the state is reputed to shape the indus- trial structure in accordance with some national strategic vision," reported the MIT Commission on Industrial Productivity, "some of the most recent research is far more skeptical about the actual influence of state policy and more inclined to emphasize the role of private actors." (Dertouzos et al., Made in America, 1989, p. 109~.
