The global semiconductor industry in the mid-1980s operated primarily on a vertically integrated model, a structure that had evolved from the industry’s earliest days. Major players, known as Integrated Device Manufacturers (IDMs), meticulously controlled every aspect of the semiconductor value chain: from initial product concept and design to advanced manufacturing in their proprietary fabrication plants (fabs), and finally, to marketing and sales of their finished semiconductor products. Companies like Intel, Texas Instruments, Motorola, Philips, Toshiba, and NEC exemplified this model, investing heavily to maintain their technological lead across diverse product portfolios, including microprocessors, memory chips, and various logic devices.
This vertically integrated structure necessitated immense and continually escalating capital investment in state-of-the-art fabrication plants, which could cost hundreds of millions of dollars to build and equip even in the 1980s. Furthermore, it required extensive research and development (R&D) in both product design and process technology, along with a comprehensive global sales and distribution network. Such a model presented formidable barriers to entry for smaller design houses or entrepreneurial startups lacking significant financial backing. This effectively limited innovation largely to a few well-capitalized corporations, as the cost and complexity of owning and operating a fab were prohibitive for most.
Simultaneously, the rapidly rising complexity of chip design began to suggest a potential decoupling from manufacturing. Moore's Law, the observation that the number of transistors on integrated circuits doubles approximately every two years, continued its relentless march, leading to increasingly intricate designs and the emergence of application-specific integrated circuits (ASICs). As chip architectures grew more sophisticated, specialized design expertise became paramount. This expertise could theoretically flourish independently if access to advanced, reliable, and cost-effective fabrication facilities were available, creating a growing need for a dedicated manufacturing partner.
Within this evolving landscape, the government of Taiwan, through its Ministry of Economic Affairs and the Industrial Technology Research Institute (ITRI), sought to elevate the nation’s technological standing. Taiwan, having successfully transitioned from an agrarian economy to a manufacturing hub for consumer electronics in the 1970s and early 1980s, recognized the strategic imperative of moving up the value chain into high-tech industries. ITRI, established in 1973, had already made significant strides in semiconductor research and possessed a pilot fabrication facility, primarily through technology transfer programs with companies like RCA. The strategic objective was to foster a high-tech industry that could leverage Taiwan’s growing pool of engineering talent and robust government support, aiming to create a long-term engine for economic growth. The opportunity was identified to create a dedicated manufacturing service that could meet the needs of emerging chip design firms and potentially even established IDMs seeking to outsource specific production stages without relinquishing control over their intellectual property.
At the forefront of this nascent vision was Dr. Morris Chang, an individual whose extensive career at Texas Instruments (TI) had provided him with an unparalleled understanding of the semiconductor industry's intricacies. Having joined TI in 1958, he rose through the ranks to become a Senior Vice President overseeing worldwide semiconductor operations, a role he held during the late 1970s and early 1980s. In this capacity, he was responsible for TI’s global semiconductor manufacturing, R&D, and marketing across a vast portfolio including microprocessors, digital signal processors, and memory. Dr. Chang, an alumnus of MIT and Stanford with a Ph.D. in electrical engineering, recognized a fundamental conflict of interest inherent in the IDM model. These companies, while theoretically capable of offering manufacturing services to external customers, were simultaneously direct competitors in the market for finished semiconductor products. This created a profound trust deficit, as potential clients feared intellectual property leakage or deprioritization of their orders in times of tight manufacturing capacity. Dr. Chang’s motivation was rooted in addressing this structural limitation, envisioning a business model that would completely separate design from manufacturing, thereby offering an unbiased, reliable, and secure production partner to all.
His proposed initial business concept, revolutionary for its time, was the "pure-play foundry." This model stipulated that Taiwan Semiconductor Manufacturing Company (TSMC) would commit exclusively to manufacturing integrated circuits designed by other companies, never designing or marketing its own products. This pure-play approach would assure customers of strict neutrality and confidentiality, alleviating concerns about competition and IP theft. The value proposition was compelling: it would allow nascent fabless companies (companies that design chips but outsource manufacturing) to focus solely on their core competencies—chip design, software development, and marketing—without the enormous capital expenditure, operational complexities, and technological risks associated with building and maintaining a state-of-the-art semiconductor fab. For a startup, this meant a dramatically lower barrier to entry into the competitive semiconductor market, fostering an explosion of innovation. It also offered IDMs a flexible way to offload mature product lines or access specialized processes without incurring massive new investments.
However, the path to establishing this unprecedented model was not without significant challenges. A primary hurdle involved convincing both established IDMs and nascent fabless companies of the viability and reliability of a pure-play model, particularly one originating from a relatively new player in the global high-tech arena. Skepticism needed to be overcome regarding a Taiwanese company’s ability to compete on quality, advanced technology, and production scale with long-established giants in Japan, the United States, and Europe. Taiwan was still perceived by many as primarily a producer of assembled goods rather than a source of leading-edge semiconductor technology. Attracting top-tier engineering talent, especially those with extensive experience in advanced process technology and fab operations, to a novel venture required a compelling vision and competitive compensation, as such talent was scarce globally. Furthermore, securing the necessary initial investments beyond government funding was critical for the capital-intensive nature of semiconductor manufacturing, where the cost of a single advanced fab could easily exceed $250 million even in the late 1980s.
Dr. Chang was recruited by the Taiwanese government in 1985 to head ITRI, with the explicit mandate to develop Taiwan’s semiconductor industry. Under his visionary leadership, the groundwork for TSMC was meticulously laid. The company was conceived as a spin-off from ITRI’s existing semiconductor operations, leveraging the institute's pilot fab facilities, foundational technical expertise, and an initial pool of trained personnel. This strategic move ensured that TSMC did not have to start entirely from scratch, benefiting from years of government-sponsored research and development. Following extensive planning and securing initial commitments, Taiwan Semiconductor Manufacturing Company (TSMC) was officially incorporated in February 1987. The initial founding capital amounted to approximately NT$10 billion (around US$300 million at the time). This capital came predominantly from the Taiwanese government, which held a significant stake of around 48%, alongside a crucial strategic investment from Philips, the Dutch multinational electronics conglomerate. Philips invested approximately 27.5%, not only contributing vital capital but also providing valuable technological expertise through process technology licenses and operational know-how. This international backing from a highly respected industry player like Philips lent significant legitimacy to the untested pure-play model and signaled confidence in TSMC’s potential. This pivotal moment marked the formal establishment of a company with a clear, albeit challenging, vision to redefine semiconductor manufacturing, poised to begin operations with its novel business approach and become the world’s first dedicated semiconductor foundry. TSMC initially focused on mature process technologies, starting with capabilities around 2-micron, 1.5-micron, and 1-micron, aiming to build a reputation for reliability and quality before pursuing the bleeding edge.