From: acquiredfm
The Taiwan Semiconductor Manufacturing Company (TSMC) is recognized as one of the most important companies in the global technology landscape, becoming the exclusive manufacturer of leading-edge chips and powering advancements in AI and mobile devices [00:00:32]. By 2025, TSMC has solidified its position as the manufacturing superpower behind the AI era [00:01:07]. Its market capitalization more than doubled from $550 billion in 2021 to over a trillion dollars by 2025, making it one of the few trillion-dollar companies not located on the West Coast of the United States [00:01:37].
Morris Chang: Early Life and Career
TSMC was founded by Morris Chang at age 56 [00:02:55]. Born in Ningbo, China, in July 1931, Chang’s early life was marked by significant upheaval, including fleeing multiple wars before he turned 18 [00:04:53]. In 1949, he moved to the United States and was accepted to Harvard, a dramatic change of fate [00:07:49]. He later transferred to MIT to study mechanical engineering, completing both his undergraduate and master’s degrees in three years [00:10:08].
After failing his PhD qualifying exams at MIT twice [00:12:42], Chang took a job at Sylvania’s new semiconductor division in 1955, choosing it over an offer from Ford Motor Company due to a $1 salary difference [00:13:58]. At Sylvania, he self-taught electrical engineering by studying William Shockley’s textbook, “Electrons and Holes in Semiconductors,” and seeking informal mentorship from a senior colleague [00:15:57]. He left Sylvania after three years due to its internal struggles, famously quoting a senior manager who said, “We at Sylvania cannot make what we can sell and we cannot sell what we can make” [00:18:24].
In 1958, Chang joined Texas Instruments (TI), a leading semiconductor company at the time [00:19:07]. He was assigned to a struggling project with IBM, aiming to produce transistors for the IBM 7090 Mainframe [00:24:49]. Morris quickly increased the yield from 0% to 20% within four months, attracting the attention of TI’s president [00:26:21]. This success led to his first managerial role, leading a germanium transistor development department [00:26:55]. TI sponsored his PhD at Stanford, which he completed in two and a half years, returning to TI in 1964 [00:27:51].
In 1967, as a general manager at TI, Morris Chang developed learning curve pricing [00:33:01]. Contrary to the industry norm of high initial prices for new, capital-intensive fabs, Chang advocated for low initial pricing to drive volume, accelerate yield improvements, and capture market share [00:31:09]. This strategy made TI’s integrated circuits business the largest and most profitable worldwide [00:33:22]. He rose to VP, overseeing the entire semiconductor business by 1972, and was considered a leading candidate for TI’s CEO [00:36:22]. However, he was moved to the consumer products division in 1978 and demoted in 1983, leading to his resignation [00:37:43].
Founding TSMC: The Pure-Play Foundry Model
After short stints at General Instrument and considering a venture capitalist role, Morris Chang’s career appeared to be in decline [00:44:22]. In 1985, he was recruited by KT Lee, known as the “father of Taiwan’s economic miracle,” to head the Industrial Technology Research Institute (ITRI) in Taiwan [00:48:46]. Taiwan at the time was a manufacturing nation with little strength in R&D, circuit design, or intellectual property [00:46:15].
Lee then tasked Chang with starting a new semiconductor company, making an “offer I couldn’t refuse” [00:54:57]. In just three days, Chang conceived the “pure-play foundry” model: a company solely focused on manufacturing chips designed by others, without designing its own [00:56:00]. This idea was highly unconventional in an industry where “real men have Fabs” (Jerry Sanders, AMD co-founder) [00:57:01].
The market for such a service was non-existent. Traditional integrated device manufacturers (IDMs) either produced their own chips or occasionally shared excess capacity, often demanding rights to market products under their own brands [01:14:14]. However, Chang foresaw the rise of “fabulous” (fabless) companies—chip designers who lacked the capital to build their own manufacturing plants [00:59:06].
TSMC was officially established in 1987 [01:03:40]. The Taiwanese government provided 50% of the initial capital ($110 million), while Philips, a Dutch company, invested 28% [01:02:22]. The remaining capital was raised from other Taiwanese business leaders, often under government pressure [01:01:36]. Notably, Morris Chang received no equity in TSMC at its founding, instead investing his own money into the company’s shares as it grew [01:02:20].
Growth, Dominance, and Strategic Importance
Initially, TSMC relied on “drags” — excess or unprofitable manufacturing orders from IDMs [01:14:07]. However, Chang actively evangelized the fabless model, attracting startups like Qualcomm, Broadcom, Marvell, and Nvidia, who benefited from not needing to build their own costly fabs [01:17:15]. Nvidia, for example, only raised $20 million and relied 100% on TSMC for manufacturing [01:17:25].
This created an “insane flywheel” for TSMC: the growth of fabless companies led to increased revenue for TSMC, which, with its high operating margins, could reinvest into more capital expenditure, further advancing its manufacturing technology [01:18:11]. This continuous investment, governed by a version of Moore’s Law (and “Rock’s Law,” which states the cost of a fab doubles every four years), allowed TSMC to pull ahead of competitors [01:50:43].
By the early 2000s, TSMC caught up to the leading edge of technology, and by the late 2010s, only Samsung remained a significant competitor at the most advanced process nodes (e.g., 5 nanometer) [01:19:20]. TSMC is set to launch 3 nanometer process technology in 2022, further solidifying its lead [01:20:03].
Morris Chang returned as CEO in 2009 at age 78, seeing “golden opportunities” in the emerging mobile and cloud computing paradigms [01:32:58]. He secured a pivotal deal with Apple in 2010, investing $9 billion in a dedicated plant to produce chips for iPhones and other Apple devices [01:40:01]. This partnership was a huge risk for both companies but proved immensely successful, with Apple becoming a primary customer and TSMC becoming the sole supplier for Apple’s most advanced chips by the iPhone 6 era [01:40:26].
Current Status and Moat
TSMC’s role in the global semiconductor industry is unparalleled. It holds over 50% of the foundry market share and more than 90% of the market share for leading-edge chips [01:47:46]. The company boasts remarkable financial performance, with 17.4% compound annual revenue growth for 27 years since its 1994 IPO [01:45:05]. In 2020, TSMC reported 48 billion in revenue [01:41:44].
TSMC’s competitive advantages are primarily in scale economies and process power [01:53:53]. The sheer capital required to build and operate advanced fabs (TSMC committed $100 billion in capex over three years starting 2021) [01:42:15] creates an insurmountable barrier for competitors [01:22:04]. Its expertise in manufacturing processes, such as Extreme Ultraviolet (EUV) lithography, is incredibly complex, requiring machines costing hundreds of millions of dollars and precision greater than sending missions to the moon [01:26:08]. This deep expertise, built over decades, cannot be replicated merely by throwing money at the problem [01:29:22].
Despite its dominance, TSMC faces significant geopolitical risks due to its primary manufacturing base in Taiwan, an island with disputed sovereignty [01:59:11]. While diversifying with fabs in Arizona and Japan, TSMC’s core advanced manufacturing remains in Taiwan [02:01:30].
TSMC’s pricing power has also become evident; after years of price reductions, it announced a 20% price increase in 2021 [01:47:00]. This ability to dictate prices in a historically commodity-driven business underscores its impenetrable moat.
Morris Chang, after stepping down as CEO in 2013, fully retired as chairman in 2018 at age 86 [01:41:17]. His vision of the pure-play foundry and the resulting “fabulous” industry has enabled unprecedented innovation and scale in the semiconductor world.