Technology and Korea’s Business Systems in Action

Televisions

Japanese expatriate engineers supervised the installation and start-up of the TV production line in order to minimize time lost in trial and error. But within a year the local technical personnel trained at Hitachi had acquired enough tacit knowledge through production and product design experience to take over from them. When the company turned to later consumer electronics products, such as cassette recorders and audio systems, they were able to move into assembly without foreign assistance.

Three other firms entered TV production at about the same time, and acquired and assimilated production capabilities through much the same process. Instead of drawing on foreign expatriates for experience-based knowhow, however, subsequent entrants hired experienced engineers and technicians away from existing firms. LG Electronics, as the first and largest producer, was a major source of experienced personnel for new entrants (Kim, 1980).

The role of the state

The eagerness of these other firms to enter the electronics industry owed much to government policy. In the first decade, the state’s role in electronics was the relatively limited one of tightly controlling imports, contraband goods in the black market, and foreign direct investment, opening an opportunity for local firms to meet the needs of the domestic market. But 1969, when the government designated electronics as a strategic export industry, marked a major change in the government’s role.

In that year, the government promulgated the Electronics Industry Promotion Act and released an ambitious Long-Term Electronics Industry Promotion Plan. It also created the Electronic Industry Promotion Fund, which offered preferential financing to build up scale economies in production. The Plan also provided grants to develop and upgrade public support systems for industry standards and R&D. The government specifically identified 95 products for promotion, offering preferential financing and other incentives to their producers. Yearly production targets were established, and progressive local content requirements were set in order to promote the parts and components industry. End products for the local market were completely protected from foreign competitors, and foreign investment allowed only for the production of key parts and components and for re-export. The government also created an electronics industrial park in order to give rise to inter-firm learning and scope economies.

A key element of the Plan was promoting the industry as a leading exporter. In 1969, when the industry exported a mere $42 million worth of products, the government set an ambitious export goal of $400 million for 1976 (the last year of the plan). The government not only set specific export goals and directives, forcing local firms to be competitive both in price and quality in international markets; it also provided incentives. Preferential financing, tax concessions, foreign loan guarantees, and the control of entry of new firms formed the heart of the export drive. That is, this ambitious program induced a crisis, compelling local firms to acquire technological capabilities quickly, and at the same time it provided the support to make the crisis creative rather than destructive. Since marketing of exports was largely in the hands of foreign OEM buyers, local firms concentrated on the acquisition of product design and production capabilities. In 1976, exports exceeded $1 billion, more than twice the target, indicating the rapid expansion of the industry’s capabilities.

Black and white televisions were the first major product to be exported in volume. As black and white sets reached at the declining stage in the major export markets, the color TV became the next target for development. With black and white TVs, the Korean companies had moved up the production learning curve on the strength of the protected domestic market, prior to competing internationally. However, Korea did not broadcast in color, so export markets were the target from the beginning. None of the foreign color TV producers were willing to license technology to Korean producers, who had so convincingly demonstrated their ability to invade what was the largest and most important market, the United States. LG Electronics and two other major producers turned to domestic sources of technology, embarking on a joint research contract with KIST in order to expand their knowledge base in color TV technology. The combined experience from these efforts and their earlier learning in black and white televisions strengthened their bargaining power in the eyes of foreign technology sources, and RCA licensed its core patents to LG Electronics in 1974.

This was a common experience in the success of Korean firms. They have often found it easier and less expensive to license a new technology if they reverse-engineer the technology beforehand. For instance, Korea’s reverse engineering of the VCR virtually forced the Japanese firms to change their policy and license VCR technology to Korea, in order to have a return on their technology investments. In these cases, the purpose of technology licensing was less to gain technology than to pave the way into the export market. In 1999, Korea is the second largest producer of color televisions with 20 percent share of the global market.

Microwave Ovens

The video cassette recorder and the microwave oven were the next targets for development. In both cases, Korean firms were faced with the reluctance of Japanese firms to transfer their technology to Korean firms, who were increasingly seen as potential competitors. Rebuffed by foreign firms in their attempts to license the technology, the Korean firms reverse-engineered the product. Only after they produced successful commercial models were they able to persuade foreign firms to license the technologies, thereby opening the paths for export.

Reverse engineering the microwave oven, however, was a formidable task. After its unsuccessful attempts in 1976 to license microwave technology, it took the industry’s first major producer, Samsung, two years of intensive work, with teams of engineers putting in 80-hour weeks of redesigning, readjusting, and trial-and-error, to produce a successful commercial prototype. The complexity of the components, particularly the magnetron tubes, posed serious problems. There were only three producers of magnetrons in the world, two in Japan and one in the US. Samsung sourced its magnetrons from Japan.

Even after it finally developed the commercial prototype, Samsung still faced problems in volume production. Although it had worked with local bakeries in field tests of the new product, microwave ovens were too expensive to find a significant domestic market. At a time when more than 5 million ovens were being sold worldwide, Samsung’s first major order came from Panama in 1980, for 1,000 microwave ovens. That year, however, proved to be the turning point for Samsung: J.C. Penney asked Samsung if it could produce a low-priced microwave oven for the U.S. market. This order meant a completely new design and heavy losses (because Samsung had yet to develop the production scale economies that would bring its costs down). But it would give Samsung a foothold in the largest and most sophisticated market in the world and the opportunity to turn a primitive assembly line into an efficient high volume operation. Penney provided technical assistance to the Samsung team to ensure that its ovens met Penney’s technical specifications.

Samsung’s engineers and technicians worked around the clock, manufacturing by day and tuning the line at night to fill the order. It was successful enough for Penney to more than triple its order within three months. However, to bring its costs down as other producers developed even greater scale economies, Samsung decided to develop its own magnetron tube, which was still sourced from Japan. After being turned down in its requests for technical assistance by both Japanese producers, Samsung in 1982 bought and transported to Korea the only U.S. factory producing magnetron tubes (it was going out of business because of the Japanese competition). Samsung also invested heavily in improving productivity by automating its production processes.

Samsung’s rapid assimilation of the design and production technology for microwaves was rooted in the quantity and quality of those critically important human resources. Samsung hired Korean engineers who had graduated from leading U.S. engineering schools, as well as from Korean universities. It soon built a microwave technical group that outnumbered their counterparts in GE Appliances, which began providing technical support for sourcing microwave ovens from Samsung. This gave the Korean engineers further opportunities to absorb world-class skills (Magaziner and Patimkin, 1989, 88-89). These engineers and technicians were willing to work long hours and with intense commitment to succeed.

Samsung’s R&D activities have led to 74 local patents and 6 overseas patents related in microwave oven technology, enabling Samsung to become one of the leading producers of microwaves in the world. By 1994 Samsung was the world’s second largest producer, manufacturing four million ovens in Korea and 0.8 million more abroad each year and accounting for 17% of the global market.

Samsung’s success prompted LG and Daewoo to follow suit. They too were turned down in their efforts to enter the field by licensing Japanese technology. The later entrants were able to poach experienced engineers and technicians from Samsung, thereby spreading microwave oven technology throughout the rest of the electronics industry in Korea. While it took Samsung four years to develop its first successful prototype, it took LG only eight months when it set up its own task force in 1980. Then LG Electronics acquired the licenses necessary to open up export markets. Samsung’s development of magnetron technology even helped LG to license the magnetron tube technology from Hitachi, which had previously refused to license it to Samsung. Now LG produces nearly as many microwave ovens as Samsung, and in 1999 Korea accounts for 40 percent of the global market.

Korea’s semiconductor industry, now one of the country’s most dynamic industries, had its beginnings in the mid-1960s, when several multinational semiconductor firms – Signetics, Fairchild, Motorola, Control Data, AMI, and Toshiba -- began assembling discrete devices in Korea to take advantage of its low labor costs. These operations involved only simple packaging processes: parts and components were imported from the parent companies, assembled in wholly-owned subsidiaries by relatively unskilled workers, and re-exported. Little design or engineering capability was transferred to Korea.

In 1975, as part of its drive for rapid industrial transformation, the government formulated a six-year plan to promote the semiconductor industry. However, the initial enthusiasm of Korean companies crumbled in the face of difficulties in obtaining technology from the more advanced countries and the accelerating risks of shortening product life cycles in this far-moving industry. Most of the firms chose to pursue consumer electronics instead.

Korea’s first semiconductor firm was actually established before this government initiative: in 1974, a Korean-American scientist with a Ph.D. from Ohio State University and semiconductor design experience at Motorola, Dr. Ki-Dong Kang, established the Korea Semiconductor Company. It experienced financial problems almost immediately, and Samsung acquired it during its first year of operations, as a source of semiconductor knowhow for its growing consumer electronics business. By 1983, however, the critical role of semiconductor technology in a range of industries was becoming increasingly clear, and the four largest chaebols -- Samsung, Hyundai, LG, and Daewoo -- each decided to enter VLSI production.

Samsung was the first to succeed in producing the 64K DRAM. It assembled a task force in 1982 to formulate an entry strategy for the company. The team members spent one of their allotted six months of work in the United States, meeting experts in the industry, particularly Korean-American scientists and engineers working in American semiconductors firms or teaching at American universities. Samsung licensed 64K DRAM design from financially troubled Micron Technologies of Boise, Idaho, and paid $2.1 million for a high speed MOS process from Zytrex of California. Samsung sent engineers to these technology suppliers for training as part of the technology transfer agreement. In 1983 Samsung established an R&D facility in Silicon Valley and hired five Korean-American Ph.D.s in electronics engineering from Stanford, Michigan, and Notre Dame Universities with semiconductor design experience at some of America’s leading firms, including IBM, Honeywell, Zilog, Intel, and National Semiconductor. These scientists, plus about 300 American engineers (including several designers who left Mostek), provided not only a high level of capabilities and a window into the technological networks of Silicon Valley, but also an opportunity for Korean engineers in participate in training and research in the U.S.. Simultaneously Samsung organized a team in Korea to work collaboratively with the California team. It included two Korean-American scientists (both with experience in 64K DRAM development in American companies) and Samsung engineers trained at technology suppliers. Intense interactions between the Silicon Valley group and the team in Korea through training, joint research, and joint problem solving significantly raised the Korean team’s capacity to absorb the VLSI technologies acquired from Micron Technology and Zytrex.

With eight years of experience in assembling LSI chips, Samsung found the VLSI assembly process relatively easy to assimilate: its production operations easily reached a 92 percent yield ratio, on a par with Japanese producers. Its mass production plant was designed and its construction supervised by a Japanese firm that had previously built a Sharp semiconductor plant in Japan. Samsung was able to market 64K DRAM chips early in 1984, about 40 months after the American pioneer and some 18 months after the first Japanese commercial entrant. Korea became the third country in the world to produce DRAMs, and had significantly narrowed the technological gap with the United States and Japan.

Hyundai was the second Korean entrant to this market, despite its lack of previous experience in electronics. Aware of the increasing importance of electronics in its core automobile, shipbuilding, and heavy machinery businesses, Hyundai decided in 1983 to develop an electronics capability in order to strengthen its competitiveness in those businesses. Hyundai first contacted Dr. Kang, the founder of Korea’s first semiconductor firm, who had by then returned to Silicon Valley, to get his help in formulating an entry strategy. Based on this plan, Hyundai recruited four Korean-American PhDs with work experience in semiconductors and computers at Xerox, System Control, Fairchild, and Ford. It also planned to expand its Korean-based capabilities by recruiting an additional 75 Korean-American scientists from the US and 35 high-caliber scientists and engineers within Korea, many from Samsung, to form the core of its new electronics business and its Semiconductor R&D Laboratory in Korea. But in addition, like Samsung, it set up an R&D center in California, staffed by Korean-American scientists and local American engineers.

Despite considerable success on the design side, Hyundai, without previous production experience in electronics, had serious troubles moving into mass production. To raise its yield rate, it pursued two strategies. First, it entered an OEM agreement to assemble 64K DRAMs for Texas Instruments, gaining assembly knowhow from TI’s technical assistance. Second, it purchased designs from Vitelic in the United States. In 1986, two years after Samsung, Hyundai became the second Korean chaebol to mass produce 64K DRAMs.

Despite its longer experience in electronics, LG took a rather cautious approach, focusing primarily on non-memory chips for use in its in-house consumer electronics business. In an attempt to enter VLSI production, LG acquired the R&D and production facilities operated by the government’s KIET (the Korea Institute of Electronics Technology) in 1984, but these had been rendered obsolete by the rapidity of developments in the industry. It then licensed chip designs from Advanced Micron Devices and Zilog in the U.S. and entered a joint venture with A.T.&T.’s Western Electric, but it was far behind Samsung and Hyundai in launching 64K DRAMs.

Daewoo, which had acquired consumer electronics and modest semiconductor production, also attempted to enter VLSI production through acquisition. In 1985 it invested about $13.4 million in the ailing Zymos Corporation, acquiring 51% equity. Daewoo transplanted Zymos’ wafer fabrication equipment to Korea, but drastically scaled back its semiconductor project and eventually gave up the idea of producing memory chips. It focused instead on semiconductors needed in the telecommunications industry, on a small scale.

Samsung was also the first to begin production of the next generation of VLSI chips: the 256K DRAM. Its top Management gave different assignments to local and Silicon Valley teams. To reduce the lead in commercialization held by the U.S. and Japanese firms, Samsung’s local team was to source circuit design from Micron Technology. Although it had enough process experience to avoid having to license process technology, the design and production challenges in taking the purchased designs into volume production were substantial. As before, Samsung involved the local team in the (by now familiar) pattern of intensive, round-the-clock efforts for eight months, resulting in the “working good die” by October 1984, which trailed the first introduction of the 256K DRAM only by two years, compared to four years for the 64K DRAM. Mass production began in early 1986, only 18 months after the first volume production in the advanced countries.

The Samsung team in Silicon Valley, however, was given the mission of developing a new 256K DRAM from circuit design through process design on its own, to enable Samsung to become independent of foreign design suppliers. Intensive efforts produced a circuit design in April 1985 and a “working good die” in July. The quality of this die was superior to the design licensed from Micron Technology on several important performance measures, and Samsung adopted it as the dominant design for mass production. Through training and relocation of personnel, Samsung was able to transfer the growing design capabilities of its California center to its Semiconductor R&D Center in Korea.

Hyundai, faced with the same challenge of speeding up development, tried to set up its production process and acquire design technology concurrently. Hyundai had few problems purchasing state-of-the-art manufacturing equipment from Japan. However, the Japanese semiconductor firms refused to give Hyundai access to their design technology, making it difficult for the company to find a design compatible with the Japanese equipment. Hyundai entered a licensing agreement with Inmos of the US, whose 256K DRAM technology was the fastest available but was not production tested, and when Inmos failed to supply the technology on time, Hyundai canceled the agreement and again turned to purchasing a design from Vitelic in June 1985. But it was not able to get the critically important yield ratio above 30% throughout 1986. Again, it turned to an OEM agreement with Texas Instruments to assemble the latter’s production-tested 256K DRAM, enabling Hyundai to improve its own production technology to achieve a profitable yield rate and allowing TI to move into the more profitable 1M DRAM.

Faced with the Korean chaebols’ entry into 64K and 256K DRAMs, Japanese producers moved quickly to sell their own chips at a fraction of the Korean producers’ cost. This strategy had been successful earlier against their American competitors, but the diversified chaebols were able to subsidize their semiconductor subsidiaries during the financial crisis generated by the Japanese. Then the chaebols encountered a stroke of luck: the US-Japan semiconductor agreement that retrained Japanese exports to the US. This and the subsequent move to the 1M DRAM by Japanese firms opened up new opportunities for the Korean firms in the US market, allowing them to emerge as dominant suppliers of 64K and 256K chips. Increasing demand and short supply also pushed up prices for the 256K DRAM, enabling the Korean firms to generate profits and firmly establishing them in the industry.

Samsung began work on the next generation, the 1M DRAM, in September 1985. This time the company committed both its Silicon Valley team and its Korean center to working on original designs, in a “collaborative competition” involving exchanges of information, personnel, and research results, but two parallel projects. This time, the Korean center completed the task first, by three months, indicating that the locus of R&D capability had shifted to the R&D center in Korea. The company produced a working good die in July 1986, narrowing the gap with the Japanese pioneer to one year. It began mass production in late 1987, one year after the Japanese firms but in time to catch the rapid rise in demand.

Hyundai was a late entrant in 1M DRAM, again acquiring design and process technology from Vitelic. But by 1988, its design and process capabilities were rapidly catching up with Samsung. In contrast, LG turned to Hitachi for 1M DRAM technology: Hitachi provided LG with the technical assistance for the production of 1M DRAMs to secure a reliable OEM source, allowing Hitachi to devote its resources to the next generation DRAMs.

The enormous investments in production facilities and R&D made by the Korean firms attracted several foreign firms to set up design houses in Korea. LSI Logic, for example, set up a design center in Korea to help Korean firms design ASICs. Texas Instruments built a facility to produce bipolar MOS and ASICs. These foreign subsidiaries provided the Korean chaebols with additional sources of expertise.

The road ahead was, however, becoming bumpier. In 1986 Texas Instruments filed a suit against Samsung and eight Japanese chipmakers, charging infringements of patents for DRAM designs, while Intel filed a similar suit against Hyundai and its American design suppliers. Both Samsung and Hyundai ended up paying royalties on past and future sales of their memory products.

Moreover, work on the next generation of chips -- the 4M DRAM -- meant competing neck-and-neck with Japanese and U.S. companies in exploring the frontiers of semiconductor technology. Anticipating the consequent difficulties in acquiring foreign technology, and seeking to avoid costly duplication in research and investment, the government stepped in and designated R&D on 4M DRAMs a national project in October 1986. ETRI, a public R&D institute, served as the coordinator in a consortium of the three chaebol semiconductor makers -- Samsung, LG, and Hyundai -- and six universities. The objective was to develop and mass produce 4M DRAMs by 1989 and completely close the technological gap with the Japanese firms. The consortium spent $110 million for R&D over three years, 57% provided by the government (much higher than in most national projects).

ETRI invited researchers from the three chaebols to participate in developing core technologies jointly at ETRI facilities. However, the three companies were unwilling to work together, and each pursued their own path: Samsung working on a stack structure, Hyundai on a trench structure, LG on a hybrid structure. Samsung was the first to complete the design of a 4M DRAM in 1988, only six months after Japan. LG was the second. Hyundai had to switch its research to the stack structure.

The government also designated the development of the next generations, the 64M and 256M DRAMs, as national projects, but again, although an ETRI-based consortium was organized, the three companies refused to share their knowledge with each other and the consortium basically became a distributor of funds. In contrast to earlier efforts in HCIs, in which the state played a major role in directing the development of technology in the chaebols, Korea’s success in the semiconductor industry should be attributed to business rather than state initiatives.

In the next generation of semiconductors, Samsung was again the first, becoming the world’s first supplier of commercial samples of 64M DRAMs in the second half of 1994 to such giant users as Hewlett Packard, IBM, and Sun Microsystems. And Samsung was the first to develop the world’s first fully working sample of 256M DRAM, after investing $150 million in R&D over 30 months. In August 1994, Samsung was ahead of the Japanese in using its own patented technology to design a new architecture that overcame operating speed limitations, making major improvements in the capacity to process vast amounts of data. Samsung had also completed product development of a 1-gigabit DRAM in 1996, nearly one year ahead of its rivals. In semiconductors, Korean firms had reached the innovation frontier.

In March 1999, Samsung began mass-producing 256M DRAM for the first time in the world, about two years before industry watchers predicted. LG became the world’s first chipmaker to develop a functional sample of 64M direct Rambus DRAM chip, which is expected to become the next primary memory for high-performance personal computers by 2001. Under the restructuring pressure from the government, Hyundai acquired LG, making Hyundai the second largest memory chipmaker after Samsung. In 1999, Korea is the largest memory chip producing country, accounting for 41% of the global market.

TFT-LCD Development

The development of TFT-LCD or flat panel display (active matrix display) illustrates further how Korean firms have strengthened their technology capabilities to emerge as innovators in world markets. Their move into the TFT-LCD industry in the 1990s resembles the development of the semiconductor memory chip industry in the 1980s. When Korean firms decided to invest in memory chips, Japanese firms dominated the burgeoning semiconductor market. In ten years, Korean firms, led by Samsung, were challenging the Japanese at the leading edge of the market. Korea was determined to repeat that success in TFT-LCDs, where Japan currently dominated.

The development of TFT-LCD is based about 30% on passive matrix LCD technology and about 70% on semiconductor technology. The leading chaebol, with strong technological bases in both, were able to build quickly on those strengths and on their experience in acquiring and assimilating leading edge technology from foreign sources. For instance, Hyundai Electronics’ involvement in passive matrix LCD began in 1988, when it organized an LCD business unit. Lacking capability in the more advanced display technologies, it imported the stick form of twisted nematic (TN) LCD from a Japanese firms, Oprex, in 1990, sending its engineers to Oprex for training in production and LCD design and importing a complete production display from Japan. Training, in-house R&D, and collaborative efforts in a government-sponsored research cooperative, the Display Research Cooperative, led to the development of Hyundai’s own TN LCD within months, and a super TN LCD by 1993. In three years, Hyundai increased its output more than a hundredfold: from 116 thousand units in 1990 to 15.22 million units in 1993.

Based on this experience and its growing capabilities in semiconductors, Hyundai organized a task force team to develop the more advanced TFT-LCD. It also approached Japanese and American LCD firms, including Oprex, for technical assistance, but none were willing to share the emerging technology. As an alternative, Hyundai in 1992 set up a joint venture in San Jose, Image Quest Technology, with a group of leading American LCD engineers, who spun out of Colory, Inc. Hyundai invested over $16 million in developing a 10.4-inch VGA TFT module prototype and in setting up a pilot plant at Image Quest. Participation in the joint research brought Hyundai engineers to a par with their Japanese counterparts.

Two other semiconductor firms -- Samsung and LG -- were as advanced as Hyundai in TFT-LCD technology, if not more so. In 1994, Samsung developed a 14.2-inch TFT-LCD with a thickness of less than 3 centimeters. In 1995, it developed a 3.1 inch polysilicon TFT-LCD, which embeds drive ICs on panels to increase the light transmissive efficiency up to 80%, enabling the producers to extend the display size to 100 inches and to diminish defect rates. In 1995, Samsung also developed a 22-inch TFT-LCD screen, one inch larger than the world’s previous biggest LCD (produced by Sharp). This signaled that Korean companies had become innovators in TFT-LCD technology. As a result, Korean firms are now attractive candidates for strategic alliances with Japanese firms. LG Electronics, for instance, established a 50:50 joint venture R&D firm with Alps Electric (Japan) to expedite the development of the next generation TFT-LCD, such as plasma processing and LCD panel processing.

In 1999, Korea is the second largest TFT-LCD producer after Japan, accounting for about 30 percent of the global market. According to International Data Corp, Korea is expected to expand its market share to 40 percent by the end of the year.

In the early years of Korea’s industrialization drive most small and medium-sized enterprises (SMEs) were barely surviving as traditional enterprises, largely neglected by government policies. Especially in the 1960s and 1970s, SMEs suffered from a disproportionate allocation of financial, technical, and human resources to the chaebols. As a result, SMEs in Korea accounted for less than half the shares of manufacturing employment and value added of SMEs in Japan and Taiwan. It was only in the early 1980s that the government belatedly recognized the importance of SMEs and began to support their growth. Consequently, the share of SMEs in manufacturing employment rose from 37.6% in 1976 to 51.2% in 1988, and their share of manufacturing value added from 23.7% to 34.9%. The case of one technology-based small firm, Medison, illustrates the increasing potential for innovative growth for Korean SMEs.

Medison is one of many technology-based small firms spun off from KAIST, the research-oriented graduate school of applied science, and one of the most successful new venture companies in Korea. The founder, Min-Hwa Lee, with a Ph.D. in electronics engineering, and his four co-founders were all graduate students under Professor Song-Bae Park. Professor Park directed a research project for two years (1984-85) on ultrasonic scanner technology, funded jointly by the government and a local medical equipment manufacturer. When the company decided to pull out of the project, the laboratory research team searched for an alternative industry sponsor (national R&D projects required an industry partner who would be willing to commercialize research outcomes). Failing to find another supporter, the research team, led by Dr. Lee, decided to spin off from KAIST to form a new venture in July 1985, becoming the industrial sponsor of the project. A person experienced in medical equipment marketing joined the five researchers from KAIST. The research team had already published eight research articles at the time of founding and obtained four patents.

The government played three important roles in Medison’s success: R&D supporter, venture business financier, and market creator. First, the government funded the initial R&D project for more than four years before and after the establishment of Medison, leading to substantial progress in mission-oriented research on ultrasonic technology at a KAIST laboratory. Second, the government assisted Medison indirectly through venture capital financing: during its first year, Medison received crucial venture capital investment and a loan from the Korea Technology Development Corporation, a venture capital company established by the government in 1981 to promote the emergence of new ventures (Kim, 1997). Third, the government created a market for Medison by restricting imports of foreign ultrasonic scanners, gradually liberalizing the market as Medison established its position.

The company’s first product was a technical and a commercial failure. The team established a very ambitious goal: to complete the development of a prototype within two months so as to exhibit it at the Korea International Medical Equipment Show in September 1985. The research team rented a small room in an inn near the KAIST campus, virtually living in the room and working around the clock for two months to translate their highly innovative patent (eight point continuous dynamic focusing technology) into a working model. They met the deadline, and the government bestowed on the founder an Industrial Achievement Award for one of the most innovative products introduced that year. However, although they placed and improved working models in two university teaching hospitals before they began marketing the product in February 1986, the 30 or so rural hospitals that purchased the product found that its image was blurry and often faded away. The scanners also broke down completely two or three times a month, forcing the team members to travel constantly to service their products.

Facing problems in commercializing their innovative technology, the team adopted a simpler proven technology to produce more reliable scanners. Once again, driven this time in part by financial desperation, they worked around the clock for another four months. This time, their experience provided enough capability to develop a second and more reliable model, which incorporated a unique SCD design developed by the team. This model was commercially successful, and over 100 units were sold in the first year, including exports to Turkey, Pakistan, Italy, Hong Kong, India, and Mexico.

Sensing a potentially large market not only in Korea but also abroad, Medison attempted to develop an inexpensive portable model. Again, their first effort at a product was a disaster, but provided the platform for a more successful model in 1988, which found an export market in developing countries. By 1990 it received U.S. FDA approval and became the best selling portable model in many developing countries.

Not content with these markets, however, Medison continued to push its technological capabilities, through its own R&D and through joint projects with KAIST. In 1988 it developed a model that incorporated patented uniform ladder algorithm (ULA) into sector display technology, but its quality was noticeably inferior to comparable foreign models. Again, however, it proved the platform for an improved product in 1989, a model that came to dominate the domestic market even in the face of increased foreign competition in the liberalizing Korean market. Its next generation product, incorporating doppler technology, became a hot selling model abroad, and Medison doubled its sales almost every year.

Medison has been aggressive at diversifying into related businesses. It established Meridian, a subsidiary to launch its bio-energy medical equipment business, integrating oriental medicine with modern bio-energy technology developed in Russia. A research team from Medison spun off to form Meridian. Medison also established Medidas to develop medical information systems such as its medical image display archiving system, tele-radiology, and a picture archiving communication system (PACS). Medison also invested in Korea Multimedia Communication, Byte Computer, and Taeha Mechatronics for research in medical information systems and medical automation.

It has reached beyond Korea to establish overseas marketing subsidiaries in the US, Europe, Russia, China, Japan, and Singapore. In production, it has entered a licensing agreement with an Indian firm to assemble Medison’s scanners on a CKD (Complete Knock-Down -- that is, assembly of pre-packaged components) basis. Shanghai Medison has also begun assembling Medison models for the Chinese market. And in R&D, Medison acquired Kretztechnik of Austria, a leading ultrasonic equipment developer, in 1996 and Acoma X-ray of Japan in 1997. Medison also entered strategic alliances with Advanced Technology Laboratory (ATL) of U.S. and NEU-Alpine of China. It has 15 domestic subsidiaries. In 1998, Medison developed the world’s first and only 4D (real-time digital 3D) color ultrasound device, making a major milestone in ultrasound technology.

Medison is a rapidly growing company of a new generation (about 300 workers at the headquarters and over 1,000 around the globe), moving rapidly along changes in frontier technology and aggressively seeking out opportunities on a global basis. It became a leading force in 13 medical equipment sectors. It has a vision to become the third largest medical equipment producer in the world by 2005.
4. THE ASIAN CRISIS AND KOREA’S BUSINESS SYSTEM IN TRANSITION

After three decades of phenomenal growth, Korea has recently plunged into a major economic crisis in the late 1997. Unlike past crises, which were evoked largely by external shocks, the current crisis stems largely from its structural weaknesses in technology and business systems. On one hand, the economic crisis has resulted in a major blow to the Korean economy, but on the other hand it can be a blessing in disguise, providing a rare opportunity for Korea to fix the structural weaknesses. This section discusses the impact of the crisis on Korea’s business systems and its transition into the 21^st century.