Electronic components and equipment - 1991 U.S. Industrial Outlook

Sarah Cooper Hall

Electronic Components and Equipment

ELECTRONIC COMPONENTS

The term "electronic component" is used to describe a wide array of products, ranging from the simplest coil to the most sophisticated integrated circuit. An enermous amount of human and financial capital is invested every year by companies competing to develop smaller, faster, more reliable, and more complex components. These advances fuel the developments in the computer, telecommunications, aerospace, robotics, and other related industries.

Before reading this chapter, please see "How to Get the Most Out of This Book" on page 1. It will clarify questions you may have concerning data collection procedures, factors affecting trade data, forecasting methodology, the use of constant dollars, the difference between industry and product data, sources and references, and the Standard Industrial Classidication system ISIC). For other topics related to the subject of this chapter, see chapter 18 (Advanced Materials), 24 (Industrial and Analytical Instruments), 28 (Computer Equipment and Software), and 30 (Telephone and Telegraph Equipment).

The sluggish overall growth of the domestic economy has weakened demand for components while overcapacity has placed downward pressure on prices. Producers, therefore, have cut costs sharply in order to stay profitable while maintaining market share. The uncertain business conditions also have led to further consolidation within the industry, with a number of U.S. companies divesting themselves of component operations.

Semiconductor producers continue to forge ahead in the design and production of new generations of memory circuits. Mixed analog-digital ASICs (application specific integrated circuits) are finding increased use in the automotive, consumer, and communications equipment markets. Constant improvement in materials, production processes, and quality controls has allowed capacitor, resistor, and connector manufacturers to produce high quality, low defect-rate components for use in just-in-time delivery arrangements.

Outlook for 1991

Current-dollar shipments of electronic components are anticipated to increase 6.6 percent in 1991, with semiconductor shipments experiencing a rise of 9.0 percent. European and Asian sales are expected to outperform the domestic market. This trend, when coupled with a weak dollar, will help expand U.S. exports.

Semiconductor producers will continue to develop integrated chip sets that provide complete system functions. This upstream move in the product development chain will necessitate close interaction between chip producers and users. The move to surface mount technology (SMT) will continue in the passive component sector, with more companies investing in SMT design and production equipment. Component producers will compete for sales on the basis of "cost of ownership" to their buyers, which includes price, inventory, testing costs, and defect rates.

Long-Term Prospects

Component industry shipments are forecast to grow at an annual rate of 6 to 8 percent through the mid-1990s. U.S. semiconductor manufacturers will have to generate or access enormous capital pools to fund the continuous R&D and capital investment needed to remain competitive. Semiconductor manufacturers that are not part of large, vertically integrated companies will have to form alliances with competitors and customers in order to survive. Consolidation of the passive components industries is expected to continue, with the rate of acquisitions and mergers in the connector and capacitor sectors accelerating.

The continued development of the Pacific Rim region will provide growing sales for U.S. producers. The European Community will remain the primary overseas market, with U.S. suppliers increasing local production capacity and seeking distribution channels into Eastern European markets. - Robert Scott, Office of Microelectronics and Instrumentation, (202) 377-2795, October 1990.

SEMICONDUCTORS AND RELATED

DEVICES

The U.S. semiconductor industry (SIC 3674) faced a market downturn in 1990, with the value of shipments declining 1.1 percent from 1989 levels. Price erosion took place in most product sectors, with memory chip prices falling by approximately 20 percent. Shipments growth for 1991 is anticipated at 9.0 percent.

Key semiconductor end-user markets such as computers, communications equipment, and instruments are all experiencing sluggish growth. The current Mideast tensions have only added to the business uncertainties and semiconductor purchasers have scaled back orders, particularly original equipment manufacturers (OEMs). The semiconductor book-to-bill ratio declined below 1.0 in the second half of 1990, indicating that new order levels were lagging. The semiconductor industry also was hit by a number of major layoffs and corporate restructurings in 1990. Unemployment in Santa Clara's Silicon Valley has increased steadily in the past 18 months, although it has not approached the levels of the mid-1970s and early 1980s, when the electronics industry was in a recession.

Many semiconductor producers became embroiled in patent infringement suits involving both domestic and foreign competitors. One U.S. semiconductor company has extracted an estimated $600 million from patent licensees since 1986 by aggressive patent enforcement. The question of proprietary rights was highlighted by the U.S. Patent Office's award of a architecture. If upheld, the patent will cover many of the semiconductor industry's most important chips and could entitle the inventor to future royalties.

The National Advisory Committee on Semiconductors (NACS), established by Congress in 1988, issued a series of reports warning that the U.S. semiconductor industry was falling behind its foreign competitors because of insufficient investment. The NACS reports stated that the Japanese semiconductor industry had outspent its U.S. counterparts by $12 billion in 1984-90. The reports also called for federal grants and loan guarantees as well as R&D tax credits and changes in semiconductor equipment depreciation rules. A shortening of the current five-year depreciation schedule for semiconductor manufacturing equipment was cited as the single most important incentive to encourage domestic producers to invest in more state-of-the-art equipment.

Product Development

Semiconductors are classified in two product groups, discrete devices and integrated circuits (ICs). Discrete devices, such as transistors and diodes perform a single function, are not as complex as ICs and account for approximately 30 percent of domestic semiconductor industry shipments. ICs are by definition composed of two or more devices and come in a wide variety of forms, functions, and sizes. One major U.S. producer has over 50,000 IC types listed in its product catalog.

All ICs must, however, fall into three broad categories: memory, logic, and logic and memory (microprocessors). Memory circuits store and give back data, while logic circuits perform specified operations on incoming data, such as addition and multiplication. Microprocessors, which form the heart of electronic systems, combine logic and memory functions on one chip.

The fastest growing and yet most volatile segment of the semiconductor market continues to be memory devices. The five largest semiconductor manufacturers relied on the memory market for an average of 30 percent of sales. While memories traditionally have been considered the "cash cows" of semiconductors, recent severe price erosion and global overcapacity have injected uncertainty in this market.

Several U.S. producers have announced their exit from the static random access memory (SRAM) market, citing uncertain business conditions and increased foreign competition. U.S. producers of erasable programmable read-only memories (EPROMs) are currently matching their Japanese competitors, with each nation controlling approximately 35 percent of the market. Japanese producers continue to dominate the dynamic random access memory (DRAM) market. Currently, only three domestic merchant semiconductor companies are producing these devices. DRAMs are currently shifting from 1 Megabit (1M) to 4 Megabit (4M) production, with 1M DRAM prices averaging $6 and 4M DRAMs averaging $34.

Shipments of DRAMs are driven primarily by personal computers (PCs), workstations, and associated peripherals, which accounted for nearly 60 percent of 1990 sales. The increased memory requirements of sophisticated software packages, the surge in networking and associated data buffers, and substantial aftermarket motherboard and add-in module sales are accelerating consumption of DRAMs. The market for 1M DRAMs probably will expand until 1992, when 4M chips will have reached the necessary price levels to supplant the previous generation of chips.

New generations of DRAMs will follow quickly, with one U.S. captive producer anticipating the development of 64M chips in the mid-1990s. DRAMs and SRAMs continue to be the "technology drivers" in the semiconductor industry. The need for faster and denser memories has pushed engineers constantly to improve semiconductor process technology. These advances tend to move in cyclical patterns as innovations incorporated into memory chips spur developments in microprocessors, which in turn require superior performance from memory and logic chips.

Standard logic ICs have seen a lot of their functions transferred directly into microprocessors. The largest growth area for logic is now in ASICs which provide a custom or semi-custom variation of circuits designed for a specific application to the buyer.

U.S. producers continue to dominate the microprocessor segment, accounting for approximately 65 percent of the world market. Microprocessor technologies are gravitating toward reduced instruction computing. The dominant current technology is known as complex-instruction-set computing (CISC) and relies on several hundred instructions to process information. New reduced-instruction-set (RISC) and minimum-instruction-set (MISC) microprocessors, which increase computing speed by stripping away instructions, are also entering the market.

Automotive electronics represents a growing market, with the average semiconductor content per vehicle expected to increase from $112 in 1990 to $126 by 1994. Semiconductor manufacturers are adapting existing products and developing new ones to operate in engine control, safety, and driver information systems. Semiconductor applications are not limited to the automobile itself, but are found in emerging vehicle navigation and "intelligent" highway technologies.

Another growing market is for new generations of chips designed to compress the huge amount of data in digitized images and motion video into compact data streams. This trend toward producing chip sets for specific applications has spread throughout the semiconductor industry. Semiconductor designers are now becoming involved in the system level, rather than focusing exclusively on individual pieces. This move upstream means that closer interaction between semiconductor and systems producers is becoming key to the success of both industries.

The cost of developing and manufacturing new ICs has more than doubled with each succeeding generation. Companies can expect to spend from $200 million to $1 billion just to begin producing samples of a chip which will become obsolete in three to four years. Faced with these high costs and rapid changes, companies are in a constant search for methods to increase production efficiencies while remaining on the leading edge in the development of smaller, faster, and more complex chips. Semiconductor companies recoup their development costs through the high initial prices of new ICs. Companies that lag in bringing the new products to market often encounter rapid price declines and are unable to remain profitable.

The cost associated with product development in the semiconductor industry also has led to an increased interest in collaborative research. Consortia such as SEMATECH, the Microelectronics and Computer Technology Corporation (MCC), Semiconductor Research Corporation (SRC), and the aborted U.S. Memories were intended to stretch research dollars and strengthen domestic producers. While SEMATECH, SRC, and MCC are active in semiconductor production equipment, basic semiconductor research and information technologies respectively, U.S. Memories failed to generate enough interest from domestic companies to get started.

The advantage provided by size and ready availability of capital are underscored by the gradual move toward product standardization. End-users in the computer and telecommunications sector are asking for increased networking opportunities between systems. This pressure for open systems and standard products provides an advantage to large companies with mass production capabilities. Small innovative companies are still able to hold profitable niche markets, but may eventually be supplanted by their large competitors as the products' market size increases.

INTERNATIONAL COMPETITIVENESS

The U.S. trade deficit in semiconductors dropped from $2.5 billion in 1989 to an estimated $1.5 billion in 1990. U.S. exports grew 9.7 percent in 1990 while imports fell approximately 2 percent (based on value). Exports to East Asian countries other than Japan and to the European Community (EC) led the growth, while exports to Japan increased slightly. The EC represents the third largest regional semiconductor market, accounting for 18 percent of worldwide semiconductor consumption, with Japan consuming 40 percent, and the U.S. 32 percent.

The 14 percent loss in total value of imports from Japan, shown in Table 1, is due in large part to a significant drop in value for three categories of semiconductor imports from Japan: metal-oxide semiconductor (MOS) memory (affected by falling DRAM prices), other MOS devices and parts for semiconductors. The number of semiconductors imported from Japan dropped only 0.7 percent from 1989 to 1990. [Tabular Data Omitted]

Western Europe

Western Europe - EC and other countries - continues to be the largest regional market for U.S. semiconductor manufacturers, accounting for approximately 18 percent of exports. U.S. companies, through a combination of exports and local production, control over 40 percent of the Western European market.

The importance of the Western European - and specifically EC - semiconductor market to U.S. and other non-European companies has been highlighted by the recent wave of investment in the region. Since 1985 at least 10 leading semiconductor companies have opened or announced plans for placing production and assembly facilities in the EC. The average cost of such facilities exceeds $100 million.

Smaller companies that do not have the capital to build a facility in the EC are looking at joint ventures as a way to establish local production. Semiconductor producers are aware that their buyers, faced with a 14 percent tariff on imported chips and uneasy about local content requirements and point of origin rules, prefer a European source for components.

In mid-1990 the European Commission dropped its proposal for a value-added rule of origin on printed circuit boards assemblies (PCBAs) this would have had the effect of reducing U.S. component content on assembled boards entering the EC. The rule of origin issue is currently being discussed in the General Agreement on Tariffs and Trade (GATT) Uruguay round and the final EC determination on how to structure any PCBA rule of origin will be shaped by the final GATT agreement.

EC semiconductor producers have responded in part to the increased pace of competition in their domestic market by engaging in collaborative research. The Joint European Submicron Silicon Project (JESSI) was formed in 1989 to help EC companies pool research efforts. Funding for the project is provided by member companies, the EC, and national governments. In 1990 JESSI suffered a setback when Philips, one of the project's main participants, suffering operating losses, withdrew from the advanced memory chip program.

Projects like JESSI and recent collaborative efforts between EC, U.S. and Japanese producers signal that European semiconductor manufacturers are aware of the danger of being left behind technologically. The large size and strong growth potential of the EC semiconductor market ensures that the intense competition between domestic, U.S. and Far East companies will continue, especially as U.S. and Japanese production facilities currently under construction come on line in the mid-1990s.

The United Kingdom and Ireland may surpass Germany as the largest semiconductor market in the early 1990s, largely because of the arrival of foreign computer companies. Germany currently leads EC semiconductor production, accounting for approximately 30 percent of the $3 billion total, followed by the U.K. (22 percent), France (19 percent), the Netherlands (13 percent), and Italy (7 percent). While EC semiconductor sales are experiencing strong growth, the market may become flooded when all the new fabrication facilities come online in the mid-1990s. The EC semiconductor market is expected to grow by 10 percent in 1991.

Eastern Europe and the Soviet Union

The initial optimistic scenarios that grew out of the political and economic changes in the former Eastern Bloc have become more and more tempered by the realization that severe economic problems persist. Consequently, the expected release of pent-up consumer demand has resulted in only a slight increase in the flow of goods and foreign investment into the region. U.S. semiconductor exports to Eastern Europe have traditionally been negligible, totalling just under $1 million for both 1989 and 1990. However, the level of direct exports will accelerate following the relaxation of the Coordinating Committee for Multilateral Export Controls (COCOM) restrictions.

The foreign trade organizations (FTOs) that used to act as the only gateway for imports and exports of electronics goods are having to compete with private importers and former state-run companies that now have import/export licenses. Eastern European buyers will approach U.S. suppliers directly and U.S. companies should be prepared to take advantage of the freer access to Eastern European markets now permitted. Although the current sales volume is small, the creation of supplier/buyer relationships should prove profitable for U.S. firms in the long run.

The technical level of Eastern European semiconductor manufacture remains far behind that of the West. Although the former German Democratic Republic (GDR) did produce a 1M DRAM, yield levels were so low that the enormous expense of the program was not justified. The Soviet Union has established several sophisticated semiconductor production sites, geared primarily to defense needs. The current move towards commercial usage of defense production within the Soviet Union will probably lead to increased efforts to enter the commercial marketplace. U.S. imports of low-end devices from Eastern Europe have increased, although they totaled only $750,000 for 1990.

Japan

Japan is the largest producer of semiconductors in the world. Japan is America's top-ranked import partner, exporting $3.2 billion to the U.S. in 1990 while the U.S. exported $966.3 million in semiconductors to Japan.

Although the Japanese market for semiconductors is experiencing slow growth, demand for microprocessors and controllers, digital/audio converters, and EPROMs will continue to rise in Japan. These markets should remain good for U.S. suppliers.

Japan is the largest single market for semiconductors, representing almost 40 percent of the world market, and expanding. Access to this important market has a major impact on the U.S. industry's ability to compete internationally.

In 1990 the United States and Japan entered the fourth year of the U.S.-Japan Semiconductor Trade Arrangement. While foreign access to the Japanese market improved somewhat, from 11.9 percent in 1989 to an estimated 13.4 percent in 1990, market share remained below industry expectations. Sanctions, imposed in April 1987 for Japan's failure to comply with the arrangement's market access provisions, continue. The sanction consist of 100 percent tariffs on Japanese laptop and high-performance desktop computers and electro-pneumatic hammers.

The U.S. Government and the Government of Japan continued their practice of exchanging views on the Arrangement, and held formal consultations in April and October 1990. Following U.S.-Japan discussions at the government and industry levels, greater number of Japanese semiconductor users formulated and updated market access plans (MAPs) aimed at increasing their purchases of foreign chips. Positive efforts by Japanese firms include creating incentive awards for buying foreign chips; establishing market share targets, which reflect future foreign purchases, and preparing lists of semiconductor products for which an immediate purchasing need exists. To promote greater access to the Japanese end-use markets, such as consumer electronics, telecommunications, high-definition systems, and the automotive markets, joint U.S.-Japanese industry groups have held seminars, conducted studies examining foreign strenghts and market access progress, and circulated product lists identifying Japanese requirements. U.S. companies are aggressively broadening their presence in Japan and have developed Market Promotion Plans (MPPs), informing Japanese semiconductor users about their capabilities and efforts to serve Japanese customers.

In conjunction with the Arrangement, two outstanding independent dumping cases against Japanese companies were suspended in 1986. Under the suspension agreements with Japanese companies, most Japanese producers of Erasable Programmable Read Only Memory semiconductors (EPROMs) and 256K and above Dynamic Random Access Memory semiconductors (DRAMs) are committed to selling their products in the U.S. market at or above their foreign market value (FMV). On August 31, 1990, counsel for seven signatories to the suspension agreements requested that the DRAM and EPROM suspension agreements be terminated. On September 14, 1990, the U.S. Department of Commerce initiated an administrative review to determine whether termination of the suspension agreement is appropriate. A final determination is expected by August 30, 1991. While the U.S. Government has not detected any dumping of semiconductors, the danger of dumping remains in situations where consumption is falling rapidly. In the face of lower demand, the pressure to reduce DRAM prices may lead to dumping.

Several recent studies focus on the competitive position of the United States electronics industry relative to Japan, and of the importance of the semiconductor industry as the supplier of components to this vital industry. These studies conclude that U.S. leadership in these industries has eroded, necessitating steps to enchance U.S. competitiveness. The electronics and semiconductor industries are considered essential to the national defense and economic security of the United States. These studies include the National Advisory Committee on Semiconductors' report to the President and Congress, A Strategic Industry at Risk, and the U.S. Department of Commerce's The Competitive Status of the U.S. Electronics Sector from Material to Systems.

East Asia

East Asia is an excellent market for semiconductors, especially for consumer electronic use. Malaysia, Singapore, South Korea, Taiwan, the Philippines, Hong Kong, and Thailand are all major producers of consumer electronics, thus their demand for semiconductors from U.S. and other sources is strong. South Korea and Taiwan also produce telecommunications equipment, computers, and computer peripherals, three other major uses for semiconductors.

Singapore, the Philippines, Hong Kong, Malaysia, and Thailand are assembly and testing locations used by Japanese and U.S. semiconductor manufacturers. While prospering from the increase in jobs from semiconductor assembly and testing plants, these East Asian countries have not developed or acquired sufficient technology to develop a domestic industry.

However, South Korea and Taiwan developed not only a domestic semiconductor industry, but also rival Japan in the DRAM market. Although both countries suffer from high interest rates and an undersupply of workers trained in electronics, they are continuing to develop their domestic industry. Their companies are trying to diversify their product lines because too many companies offer the same products, particularly low-end memories.

South Korea is one of the fastest growing markets for semiconductors. Japan is the major supplier to the Korean market, followed by the United States. South Korea will continue to be an excellent market for U.S. semiconductors, especially in the high-end market.

Following current trends in electronics exports, South Korean companies are moving from consumer electronics into higher-end products. The value of South Korean-produced electronic parts and components, including semiconductors, has surpassed that of consumer electronics.

To help South Korean electronics companies compete with their Japanese and U.S. counterparts, the Government of Korea has implemented a five-year plan that provides approximately $40 billion for technology development and plant investment in sophisticated industries, including semiconductors.

South Korean semiconductor companies primarily produce memory devices. Korean technology in DRAMs has reached the 1M and 4M level (although they are experiencing some development problems in 4M) and research continues in the 16M and 64M DRAMs. Wafer production has reached the 6-inch level. South Korean companies are entering the ASIC and custom market to cushion the effects of the falling prices of memory semiconductors. At least one major Korean company is shifting its focus from DRAMs to SRAMs and Mask ROM production.

Problems in South Korean semiconductor production include a low yield rate, very little cooperation between companies, and a history of labor unrest. Because most technological advances in semiconductors have come from outside South Korea, information on new technologies is difficult to obtain. Labor costs are low, but as South Korea becomes more prosperous, wages are likely to rise.

In 1990, the U.S. exported $749 billion in semiconductors Taiwan. Japan leads in exports to Taiwan, with the U.S. ranking second. Taiwan's IC market, valued at $1.9 billion in 1988, is expected to grow at a rate of 40 percent per year. Demand for semiconductors will reflect rising production of telephones, calculators, televisions, and audio products. Semiconductors for computers and peripherals will continue to be a good market for U.S. companies since most of the locally made computer products are OEM items for U.S. firms. Chip sets for computers are already becoming key products for Taiwanese semiconductor producers and exporters to Taiwan.

Taiwanese companies are feeling the effect of offering overlapping product lines, underutilized wafer fabrication facilities, slow return on investment, and high capital costs involved in semiconductor production. Some analysts predict a shakeout in the Taiwanese market in the near future.

However, all of the major Taiwanese electronics companies have invested heavily in semiconductor production. The number of semiconductor fabrication plants in Taiwan has expanded over the last 10 years from 0 to 7. By 1993, the number could double if the Taiwanese broaden their product lines.

Outlook for 1991

U.S. exports are expected to increase by approximately 10 percent in 1991, fueled by increased demand in the Far East and EC. Domestic shipments are projected to grow by 8 percent in 1990-1991, with the telecommunications sector providing the most active market. If, however, the Gulf crisis continues and the economy worsens, semiconductor shipments growth would be adversely affected.

As the overseas markets for semiconductors grow offshore production and exports contribute to the globalization of the industry. Several U.S. semiconductor companies received over 40% of their revenue from outside of the U.S. in 1990. Competition from foreign companies, and the need for better customer service, quality, and just-in-time delivery are driving U.S. companies to invest abroad. Overseas markets, especially East Asia and the EC are growing faster than the United States.

Long-Term Prospects

Demand for semiconductors will continue to be driven by the computer, telecommunications, and automotive markets. The use of chip sets and custom devices, such as analog to digital circuits in the automotive industry, is expected to soar. The EC and East Asia will be the fastest growing market for semiconductors from 1990-1994, out-performing the domestic U.S. market.

Capital expenditure by semiconductor manufacturers is expected to increase most rapidly in the EC during 1990-1994, with that region's 20-percent increase outperforming the U.S. and Japan, both of which will each experience 15 percent growth. - Dorothea Blouin (202)377-1333, Robert Scott (202)377-2795, Marguerite Markey Nealon (202) 377-8411, Office of Microelectronics and Instrumentation, (202)377-8411, October 1990.

ELECTRONIC COMPONENTS OTHER

THAN SEMICONDUCTORS

Slow, uneven growth characterized 1990 for U.S. producers of components other than semiconductors (SICs 3671-2, 3675-79). Domestic shipments totaled approximately $38 billion in 1990, which represented a 2.5 percent constant dollar increase over 1989 levels. A growth rate of 4.2 percent is expected in 1991, with shipments anticipated to rise by 5 to 6 percent. Exports will lead the expansion in both 1991 and 1992, primarily with sales to the European Community and Pacific Rim markets. Growth will be augmented by a gradual upturn in the domestic market during this period.

The component categories covered in this section are: electron tubes (SIC 3671); printed circuit boards (SIC 3672); capacitors (SIC 3675); resistors (SIC 3676); coils and transformers (SIC 3677); connectors (SIC 3678), and electronic components, not elsewhere classified (SIC 3679).

In the second half of 1989 and throught 1990, component producers experienced slack demand in the domestic market, with accompanying downward pressure on pricing and profit margins. Companies have implemented business strategies that they expect will bring them through the near-term difficulties and enable them to take advantage of any future upturn in the market. The industry has sharply reduced personnel, consolidated product lines, increased the use of quality management techniques, and emphasized exports.

Total employment in 1990 declined to 350,000 workers, a 2.2 percent drop from 1989 levels and a 8.9 percent reduction from 1988. The majority of layoffs has occured among administrative and non-technical personnel. The industry continues to undergo a consolidation process driven by the cutback in component suppliers by original equipment manufacturers (OEM), overcapacity, and the abundance of inter-changeable products being offered by competing suppliers.

As example of the consolidation going on in the component industries, Fleck International forecasts that the connector industry will see a reduction from 500 domestic producers currently to only 100 by 1995. While the severity of the shakeout is unclear it has occurred in all product sectors, underscoring how vulnerable producers of mature products lines are to periods of slow growth.

The recent easing of East-West tensions has reduced sales to the military market. The Electronics Industry Association (EIA) has forecast that defense spending will decrease at an annual rate of 2 percent over the next 10 years. As the military balances a loss in numbers with upgraded technological capabilities, the proportion of defense expenditures committed to electronics research, test and engineering, and procurement will increase. The Persian Gulf crisis has further clouded the short-to medium-term outlook, with some component producers experiencing increased military sales attributable to the U.S. buildup. The long-term trend for electronics companies, however, is clearly away from overexposure in the military market. The situation is underscored by one major defense contractor's recent announcement that it intended to fully divest itself of all military electronics divisions.

Some major component end-user markets, among them computers, telecommunications equipment, and automobiles are experiencing their own periods of slack demand. Consequently, a flattening of new orders for components has triggered a decrease in inventories among producers and put downward pressure on pricing. Tighter profit margins have led many companies to curtail capital and R&D expenditures, weakening the U.S. industry's ability to remain competitive in the long run. Some argue that R&D levels are adequate, but that development times must be reduced by the parallel inclusion of manufacturing, design and marketing into the development phase of the product.

The push toward stretching R&D dollars has led some companies to investigate joint research projects with other individual companies, universities, and industry consortia. Although the potential benefits of pooled research are very attractive, the actual implementation of such programs can be problematic. Companies accustomed to their own decision-making process often are initially reluctant to compromise on the procedures and goals of the project. Participants have to identify the scope and purpose of the research, allocate funds and personnel, and agree on the internal use and external dissemination of research results.

In addition to collaborative efforts under way in the private sector, the U.S. Department of Commerce's National Institute of Standards and Technology (NIST) is working to facilitate cooperation among component producers. During 1990 NIST began discussions with U.S. producers of ceramic multilayered chip capacitors (MLCs), as well as their suppliers and customers, on the possibility of forming a consortium to address the issue of mechanical reliability of the MLCs. Being based on ceramics, MLCs fail randomly in service due to cracks that result from defects introduced or exaggerated during manufacture or assembly. A collaborative effort on the part of manufacturers and users should improve the product, with tangible benefits accruing to all involved.

Product Development

Steady technological progress continues in component materials development and manufacturing techniques. TV cathode ray tube (CRT) manufacturers and developing technologies that they hope will allow them to compete with flat panel displays in the high-definition television (HDTV) market. Upgraded CRTs have become more viable because flat panel displays are encountering technological and pricing barriers years away from entry into the commercial market. CRT manufacturers will continue to hold an advantage in the developing HDTV market as long as the more exotic flat-panel technologies, such as plasma, active matrix liquid-crystal displays, and vacuum microelectronics are unable to generate a large display that can compete with a CRT on a cost and quality basis. By increasing the speed at which electron beams scan the screen and controlling the beams more precisely, CRT manufacturers can produce high definition displays of up to 40 inches in diameter.

Printed circuit board (PCB) manufacturers continue to upgrade in-house design capabilities in order too interact with customers at an earlier stage in the production process. Instead of waiting to be presented with board layouts and specifications by the customers, PCB shops now are working increasingly with customers to help create the final design. Although the up-front board costs may be higher with a PCB shop that has committed itself to higher quality and technology by utilizing computer-aided design and manufacturing (CAD/CAM), the savings in inspection as well as rework and repair charges will provide overall price reductions to the buyer.

New PC-based board design software packages have become more sophisticated and are providing PCB manufacturers with a lower cost alternative to packages designed to run on workstations. PC based packages now contain a wide choice of pad and track sizes, automatic and interactive parts placement, and design rule checking. PCB manufacturers also are using CAM stations to control such processes as image transferring and drilling.

Capacitor manufacturers are meeting the challenge presented by customers demanding quality levels of defects per billion, rather than per million. Failure analysis techniques now include fractography, which analyzes entire crack surfaces and helps identify the stresses affecting board-mounted ceramic components. Advances in ceramic materials, such as the introduction of lead magnesium niobate based ceramics in place of barium titanate, have allowed for higher dielectric constants and smaller, more efficient MLCs. Ceramic capasitors represented approximately 45 percent of the global capacitor market in 1990, with chip capacitors for surface mount applications expected to account for 80 percent of ceramics shipments by 1995.

Tantalum capacitors, which represent the second largest market segment by value, are requiring less tantalum powder to maintain necessary electrical properties. This development of new powders will reduce unit costs and is expected to double the capacitance by volume, thereby reducing capacitor size. The rapid increase in types of surface-mount products has brought a need for standardization. The Electronics Industry Association is considering the adoption of four basic size designations for surface mount capacitors incorporated under one standard designation.

Resistor manufacturers also are busy reducing size while upgrading performance. One leading producer recently introduced a surface mount resistor with dimensions of 1.0 x 0.5 x 0.35 mm, approximately the size of a grain of pepper. As is the case for capacitors, resistor manufacturers are developing items for the growing automotive applications market and are experiencing increased demand for surface mount products. The recent softness in the automobile market and evidence that the growth in surface mount sales in simply balancing a decline in leaded resistor sales has slowed investment in surface mount techniques.

Connector manufacturers are finding fiber optics to be the most active markets sector, with increased telecommunications and computer networking applications providing the impetus for strong growth. The advent of surface mount technology (SMT) has brought with it a wide array of connector systems that are fitted directly to the printed circuit board, thereby saving space. Printed circuit board connector sales have benefitted from the increased use of personal computer networks. Newly developed corrosion resistant composite connectors have found markets with the military and automobile industries.

Because of SMT, guidelines in the design, selection, and usage of surface mount connectors are needed. The Institute for Interconnecting and Packaging Electronic Circuits (IPC) has produce a number of documents that provide guidelines and assist designers in effectively interconnecting a package.

INTERNATIONAL COMPETITIVENESS

The U.S. recorded a $1.3 billion trade deficit in 1990 in electronic components other than semiconductors, an improvement of 24 percent over 1989's $1.7 billion figure. Exports, primarily into the European Community and Far East, increased by 21 percent, while imports rose by only 13. percent. U.S. companies, facing flat domestic demand and assisted by a devaluation of the dollar, have focused on foreign markets for increased revenue.

The largest regional market for U.S. tube manufacturers was the EC, which accounted for 22 percent of the $590 million in total exports. The EC market for tubes, currently at $3 billion, is anticipated to outperform other regional markets in the near term and continue to provide U.S. exporters with a strong market. Exports of TV tubes have concentrated on Mexico, where they are assembled into TVs and reexported into the U.S. market. East Asian producers continue to dominate U.S. imports, with Japan alone accounting for 54 percent of the $800 million total.

Printed circuit board manufacturers continue to find their largest markets in the EC, with U.K., West Germany, Ireland, and France each accounting for over $40 million in U.S. exports. Assembly operations in the Far East, responding to growing regional demand for finished products, also are providing U.S. suppliers with increased sales opportunities. Singapore alone imported nearly $50 million worth of U.S. boards in 1990.

Capacitor and resistor exports have followed similar patterns, with Latin America, East Asia, and the EC each accounting for approximately one-third of U.S. exports. Capacitor exports in 1990 totaled $507 million, a 10 percent percent increase over 1989 levels. Foreign sales of resistors remained flat at $250 million. Connector exports rose 9 percent over 1989, with fiber optic connectors accounting for the majority of the growth. The EC once again proved to be the primary market for U.S. connectors, accounting for 43 percent of total exports.

The recent surge in production and consumption of electronics in the Pacific Rim region has challenged U.S. component producers to establish and increase market share in what is for many of them a non-traditional market. Many U.S. companies, which had until now focused on the EC as their primary export market, are becoming increasingly involved in Malaysia, Thailand, and Singapore.

U.S. firms attracted to the size and growth potential of the EC are finding that a patchwork of marketing strategies and distribution channels are needed to enter the different regional and national markets contained within the EC. Within the EC, Germany represents the largest market for components, accounting for approximately 30 percent of demand. The recent addition of 16 million consumers from the former DDR will only expand this market. The UK (23 percent) and France (17 percent) followed, while Italy has experienced strong demand growth and become an attractive market for numerous U.S. suppliers. To take advantage of the European market, many firms are increasing local production capacity and upgrading marketing and service efforts. Component producers are also following their major buyers into the EC, with a number of them collocating production facilities in Ireland, Spain, and Italy.

Eastern Europe is an attractive potential market, but problems with currency convertibility, private property laws, and distribution and service arrangements remain. Foreign Trade Organizations (FTOs) such as Transelektro and Metrimpex in Hungary, Unitra and Elektrim in Poland, and KOVO in Czechoslovakia have in the past decade relinquished their monopolies on trade in electronics, and currently account for only 40 percent of foreign trade. Individual domestic producers, which are currently in the process of privatization, are setting up their own import/ export operations. U.S. producers will have to identify these new companies as they represent both direct sales opportunities and potential partners for production and distribution arrangements.

Outlook for 1991

A moderate increase of 4.2 percent is expected for components shipments in 1991. Electron tubes and assorted components subsumed under the "not elsewhere classified" category, among them microwave components, switches, and printed circuit assemblies, will experience the strongest growth rates. The growth projection is based on a recession-free economy, and could be adversely affected by prolongation of the Persian Gulf crisis.

Foreign markets should continue to respond to the devaluation of the dollar, which should benefit U.S. manufacturers. The trade deficit for components should continue to decline, reaching an estimated $1 billion in 1991 as opposed to $1.3 billion in 1990.

Long-Term Prospects

Growth in domestic shipments is anticipated to increase by an average of 5 percent annually in the 1990-95 period. A continued consolidation of the producer base with each component segment is expected.

OEMs will continue to ask their suppliers to implement total quality management (TQM) techniques and just-in-time (JIT) delivery. Companies that do not meet these requirements will be dropped from preferred vendor lists. Customers will accept higher prices as long as on-time delivery and excellent quality levels that are good enough for ship-to-stock programs are achieved. Component endusers continue to move toward reductions in inventories, counting on their suppliers for quick delivery capabilities. OEMs are also setting up Electronic Data Interchanges (EDI) transmittal of order information, thereby cutting down on delays.

Companies will need to become involved in establishing industry standards, with those companies taking an active role benefiting from increased future sales. Strategic partnerships between OEMs and component producers will offer cost reduction and increased product development cycles, but can also place partners in overly dependent relationship. Benefits and risks have to be weighed by the companies involved. - Robert Scott, Office of Microelectronics and Instrumentation, (202) 377-2795, October 1990.

SEMICONDUCTOR MANUFACTURING

EQUIPMENT

Sales growth was minimal in 1990, due to over-capacity in the world market for certain generations of dynamic random access memory devices (DRAMs), and a slow domestic economy. Sales should improve modestly in 1991 (by 5 percent), with the continued growth of markets in the Pacific Rim, and the revamping of fabrication lines for new generations of devices.

The semiconductor manufacturing equipment (SME) and materials industry consists of four primary product areas: wafer processing, test and inspection, assembly equipment, and electronic materials. Other peripheral product segments include computer-aided design, mask/reticle making equipment, and wafer growing equipment. Electronic materials include silicon, gallium arsenide, sputtering targets, photo-resists, gases and chemicals, and packaging materials (ceramics, leadframes, encapsulation resins, bonding wire, hybrid packages, and thick film pastes). The majority of devices are manufactured using silicon wafers as a base. Gallium arsenide devices are slowly gaining in popularity, with applications as integrated circuits and light-emitting diodes.

Wafer processing is the most capital intensive phase of the production process and represents just over 50 percent of the dollar value of equipment sales. Processing steps include epitaxy, oxide preparation, micropatterning (lithography and etch), diffusion, ion implantation, passivation, and metallization. Testing of a device occurs throughout wafer processing, via optical and electrical means. Test and inspection equipment represents approximately 25 percent of equipment sales. Because assembly of devices was labor intensive, many established assembly fabrication lines in developing countries. However, as this process becomes more automated and requirements for particle-free manufacturing extends throughout device processing, there is less incentive to move this production step to low-wage economics and from the core manufacturing facility. Assembly equipment represents approximately 9 percent of the total equipment market.

The importance of the semiconductor device, equipment, and materials industries to U.S. national and economic security is being recognized by more and more officials in Government and industry.

A number of efforts have been taken to stem this decline. These include: the creation of SEMATECH in 1987, an advanced-technology industry research consortium; an attempt by industry between 1989 and 1990 to structure a production consortium for advanced DRAMs; and the creation of the National Advisory Committee on Semiconductors (NACS) by the National Advisory Research and Development Act of 1988.

This committee has highlighted the decline in competitiveness of the U.S. semiconductor, equipment, and materials industries. Their report released in 1990 on the U.S. equipment industry concluded that explanations for this erosion include: the decline of the U.S. device industry; lack of adequate capital due to its high cost and lack of investor confidence; and the inability of companies to achieve a critical volume of overseas sales.

SEMATECH was established in 1987 as a joint government-industry effort to boost manufacturing capabilities of U.S. semiconductor producers, largely by improving relationships between the producers and their equipment and materials suppliers. The research consortium also has as its goal achieving 0.5 micron line widths by 1992 (achieved in 1990) and 0.35 micron line widths by 1994. Only U.S. equipment and materials will be used to achieve these production parameters. SEMATECH works closely with semiconductor manufacturers, semiconductor equipment and materials producers, academic institutions, and government agencies, including the Defense Advanced Research Projects Agency (DARPA), the Department of Energy's research laboratories, and the National Institute for Standards and Technology. Half of SEMATECH's budget comes from member device companies, and half from DARPA.

One of SEMATECH's strategies to improve the supplier/ customer dialogue is through technology transfer. Manufacturing techniques are transferred via numerous workshops and seminars, and through the employees that member companies assign to SEMATECH. SEMATECH also has established joint development and equipment improvement contracts with U.S. semiconductor manufacturing equipment (SME) and materials producers. Projects include the development of a sub-micron reticle and mask exposure system, an advanced metal deposition procedure, new automatic test equipment, and gas control technology. The transfer process is also typified by SEMI/SEMATECH, an organization of U.S. equipment, materials, software and services companies established in 1987.

SEMATECH's contribution to technology developments should prove very useful to U.S. equipment manufacturers. The life cycle of most semiconductor equipment is getting shorter and shorter. With each new generation of dynamic random access memory devices (DRAMS), lithography equipment must achieve smaller line-widths, deposition and etch equipment must become more precise, and test equipment must achieve higher pin counts and speed. With companies spending an average of 16 percent of revenues on research and development, SEMATECH's research base should prove an invaluable additional resource.

SEMATECH's efforts to change the way device producers view the equipment and materials suppliers will be beneficial to both industries. These efforts focus on making the semiconductor manufacturers realize the advantages that can accrue from long-term and continuous communication of technical and performance parameters to U.S. equipment providers. Device manufacturers should consider providing equipment companies with data on the performance of their product, and entering into longer development contracts with their suppliers. Such a relationship would stabilize and improve the performance of U.S. equipment and materials companies just as such relationships have bolstered Japanese equipment manufacturers. Technology Opportunities.

A new semiconductor generation is typically introduced every three years. Most recent is the shift from the one (1.0 micron geometries) to the 4-megabit (MEG) DRAM (0.8 micron). Soon companies will shift to producing 16 (0.6 micron) and then 64-megabit DRAMs. The most critical technology for these sub-micron devices is pattern delineation, specifically the lithography and etch processing steps.

The industry is debating which lithography process is most appropriate for DRAM generations following the 64-megabit chip. Optical methods have survived far longer than anticipated, with I-line steppers replacing H/G-line equipment for use in 4 MEG DRAM production. Future options as light sources include Deep ultra-violet (UV), E-beams, and X-rays. Japanese companies and government agencies have invested heavily in hard X-ray sources, and have built at least a dozen large synchrotron. There are only two large synchrotron rings in the United States, with two more in the planning stage. U.S. companies and researchers, including SEMATECH, are focusing on deep UV, and "soft" X-ray sources, methods which are less capital intensive.

Another high-profile technology debate is over the future of the cluster tool concept. As densities of circuits increase, and line widths decrease, the integration of different processes under vacuum (such as deposition and etch) is becoming a necessity. Although more and more companies are pursuing the concept of cluster tools, its general industry acceptance is being delayed until integration standards are fully adopted by the industry.

Another "niche" application is for the manufacture of flat panel displays. The processes used to manufacture the displays mirror those needed to produce semiconductors. However, adaption of semiconductor production technology will require heavy financial investment, which many companies are unwilling or unable to pursue.

INTERNATIONAL COMPETITIVENESS

Two-thirds of the equipment producers are U.S.-based. However, they supply only 40 percent of worldwide sales, and their market share has declined steadily since the early-1980s. The majority of U.S. companies are small, entrepreneurial firms with a niche product, and sales of $5 million to $25 million. Only a few have established a strong position in the international market, and have been able to generate sales of $200 million and above.

The high costs of doing business contribute to the concentration of market share among a few large companies. Especially costly are researching and developing new product advances and establishing overseas sales and production facilities to enhance foreign market penetration and ensure prompt product servicing.

U.S. companies have a strong position in the following product areas: ion implantation, plasma chemical vapor deposition, etch wafer processing equipment, and linear testers (particularly mixed signal test equipment). Certain U.S. companies also are very competitive in reticle and mask inspection tools (metrology equipment). The only materials segment in which U.S. companies remain competitive are gases, photoresists and thick film pastes.

U.S. companies have lost significant market share to the Japanese in optical steppers, resist equipment, memory testers, certain assembly equipment areas such as tape automated bonding and die bond, and the majority of packaging and processing materials.

The efforts of SEMATECH and other research consortia such as Semiconductor Research Corporation may help to advance U.S. technology. Moreover, the research efforts of certain U.S. companies in the field of lithography, such as step and scan equipment, soft X-ray tools, and E-beam write equipment may begin to erode Japan's dominating position in this critical technology area.

The industry has been eroded by a recent avalanche of foreign acquisitions of key equipment and materials companies. Since 1988, the U.S. has lost its recognized leader in sputtering targets; will probably lose its elite gas control manufacturer; has a diminished supply base for polysilicon, and is dependent on foreigners for its supply of silicon wafers. Contributing to this upsurge in foreign investment in the U.S. were a weak dollar, the unwillingness of domestic capital to invest in high-technology fields, and foreign (typically Japanese) interest in U.S. technology.

The Japanese continue to be the dominant investor in SME and materials suppliers. In the area of materials alone, three polysilicon producers, four silicon wafer manufacturers, and two sputtering target producers have been acquired by Japanese companies since 1984. In each of these materials areas, the U.S. has become more dependent on Japanese suppliers. Moreover, many U.S. companies are discovering that a business/investment relationship with Japan can lead to a larger market presence in Japan, as well as provide additional sources of capital.

Japan continues to be both the largest foreign purchaser of U.S. semiconductor manufacturing equipment as well as the most important supplier. To achieve sales volume, U.S. equipment producers must have a presence in the Japanese market (which represents approximately 50 percent of the world market).

The EC has always been a major market for equipment, consistently representing approximately 12 percent of worldwide market share. South Korea is the most rapidly expanding market. Its semiconductor manufacturers are making a major push in DRAM production, attempting to rival the U.S. and Japan. Singapore, Taiwan, and Malaysia also offer sales potential, especially for back-end or assembly equipment.

With the relaxation of some export controls, Eastern Europe and the Soviet Union may now provide some market opportunities for U.S. equipment producers. Their institutions, universities and businesses are potential markets for the less sophisticated production equipment. Selling to these markets can, therefore, generate high profit margins for companies that are able to find a customer with some hard currency.

Changes in the export regulations include: exports below $5,000 do not need CoCom review for licensing; only national discretion is required for a license for exports valued at less than $50,000; most equipment will be decontrolled to the "China Green Line," with a core list of controlled equipment focussing on lithography and etch equipment capable of 2 micron and smaller line widths.

Outlook for 1991

U.S. equipment companies faced minimal to no growth in 1990. The outlook for 1991 is somewhat better, with a forecast of 5 percent expansion (current dollars). Companies that can compete effectively in the Japanese market will be most successful. However, both the Japanese and European markets are approaching saturation, as shown by falling one-megabit DRAM prices and reports of declining capital expenditures by the Japanese device producers. The possibility of war in Iraq, and general softening of the Western economies will further dampen demand in 1991.

Offsetting this downward trend will be the inevitable re-equipping of fabs for 4- and 16-megabit DRAM production; greater demand for flat panel display production equipment; and expansion of some less traditional markets. Taiwan and South Korea will continue to be the fastest growing markets for equipment companies, while Eastern Europe and the Soviet Union will provide a few sales opportunities for the more adventurous companies.

Long-Term Prospects

The long-term outlook for U.S. equipment producers is mixed. U.S. semiconductor equipment companies must have adequate domestic capital sources such as tax credits and/or subsidies, or pooled development/production funds generated by consortia to pursue new technologies. The first source will require change in the tax law, while the second will require relaxation of antitrust rules. Their future also depends on forging much closer business relations with U.S. device producers.

The plethora of joint development contracts between U.S. and Japanese companies often appears a mixed bag. Their ability to provide additional capital is welcome. However, it's questionable whether development or production contracts between Japanese and U.S. companies have opened the Japanese market. Moreover, some of these arrangements have led to intellectual property rights violations that can quickly erode any company's initial market gains.

Support for the U.S. industry comes from the efforts of SEMATECH, the continued ability of U.S. entrepreneurs/ engineers to devise excellent new products, and the growing market acumen of U.S. producers. - Sarah Cooper Hall, Office of Microelectronics and Instrumentation, (202) 377-2846, October 1990.

PHOTO : The fastest growing and yet most volatile segment of the semiconductor market continues to be memory devices, such as this 5-inch 64K static random access memory (SRAM) silicon wafer.

PHOTO : Technicians fine tune the chemical vapor deposition process that helps spawn new device generations.

Additional References

(Call the Bureau of the Census at (301) 763-4100 for information about

how to order Census documents.) SEMATECH's Efforts to Strengthen the U.S. Semiconductor Industry;

General Accounting Office, September 1990. Telephone: (408) 437-0292. CMP Sector Analysis, Electronic Components, Taiwan U.S. Department

of Commerce, Taiwan desk. Telephone: (202) 275-6241. CMP Sector Analysis, Electronic Components, Korea U.S. Department of

Commerce, Korea desk. Telephone: (202) 377-4957. The Competitive Status of the U.S. Electronics Sector from Materials to

Systems, Science and Electronics, International Trade Administration,

U.S. Department of Commerce, April 1990. Available from U.S.

Government Printing Office, Washington, D.C. (stock number 003-009-00574-4)). EDN, The Cahners Publishing Company, 275 Washington Street, Newton,

MA 02158-1630. Electronic Business, Cahners Publishing Company, 275 Washington

Street, Newton, MA 02158-1630. Electronic Engineering Times, CMP Publications, Inc. 600 Community

Drive, Manhasset N.Y. 11030. A Strategic Industry at Risk, National Advisory Committee on Semiconductors,

1555 Wilson Boulevard, Arlington, VA 22209 Telephone: (703)

528-6090. Semiconductors, Printed Circuit Boards, and Other Electronic Components,

1989 Current Industrial Report, MA3Q. Bureau of the Census,

Industry Division. U.S. Department of Commerce, Washington DC

20233. Telephone: (301) 763-4132. CircuiTree, (PCBs) 340 Martin Avenue, Santa Clara, CA 95050. Telephone:

(408) 986-1292. American Electronics Association, 1225 I St., Washington, DC 20005.

Telephone: (202) 682-9110. Electronics Industry Association, 2001 Pennsylvania Avenue, Washington,

DC 20006. Telephone: (202) 457-4900. Fleck International, 1640 Pacific Coast Highway, Suite 217, Huntington

Beach, CA. Telephone: (213) 592-9000. The Institute for International and Packaging Electronic Circuits, 7380

N. Lincoln Ave, Lincolnwood. IL 60646 Telephone: (708) 677-2850.

COPYRIGHT 1991 U.S. Department of Commerce
COPYRIGHT 2004 Gale Group