TSMC founder Morris Chang

Episode Overview

• This episode features a rare, in-person interview with Dr. Morris Chang, the 93-year-old founder of TSMC, recorded at his office in Taipei, which was secured through an introduction from Nvidia CEO Jensen Huang.
• Dr. Chang recounts pivotal moments in TSMC's history, including his return as CEO in 2009 to resolve a major crisis with Nvidia over the 40nm process node.
• The conversation details the high-stakes, "bet-the-company" decision to partner with Apple on the 20nm iPhone chip, which required a massive investment and temporarily allowed Samsung to take the lead in the subsequent 16nm generation.
• The core philosophies behind TSMC's success are explored, including the strategic power of the pure-play foundry model, the principle of not competing with customers, and the aggressive application of the learning curve theory.

Key Concepts

• Crisis Management: Dr. Chang's hands-on approach to resolving the 40nm yield dispute with Nvidia upon his return as CEO in 2009, highlighting decisive leadership.
• Strategic Bets & Technology Nodes: The immense, company-defining investments made in specific process nodes, particularly the 28nm node which Chang saw as a "tide in the affairs of man," and the 20nm node for Apple.
• The Pure-Play Foundry Model vs. IDMs: A central theme is the inherent conflict of interest for an Integrated Device Manufacturer (IDM) like Intel to act as a foundry for a competitor, which created TSMC's key advantage of customer trust.
• The Power of Personal Relationships: The importance of direct, personal relationships between CEOs, such as Chang's meeting with Jensen Huang at his home, to resolve complex business disputes.
• The Learning Curve Theory: A foundational principle, learned at Texas Instruments, that dictates aggressive pricing and investment to gain market share, thereby accelerating cost reductions and efficiency through increased production volume.
• High-Stakes Trade-offs: The decision to prioritize Apple's 20nm chip secured a landmark customer but delayed the 16nm node, allowing Samsung to temporarily gain a lead, illustrating the intense competitive landscape.
• The Fabless Revolution: Dr. Chang's foresight in anticipating the rise of fabless semiconductor companies, which was the foundational thesis for creating a dedicated foundry service.
• Ecosystem Advantage: The role of Taiwan's Hsinchu Science Park in creating a deeply integrated and difficult-to-replicate ecosystem of suppliers, partners, and academic institutions that supports TSMC.
• Natural Monopoly Dynamics: The concept that the enormous and escalating capital requirements for leading-edge semiconductor manufacturing naturally lead to industry consolidation and a dominant market leader.

Quotes

• At 1:11 - "We threw the Hail Mary. We asked friend of the show Jensen Huang if he would ask Dr. Morris Chang, the 93-year-old founder of TSMC, if he would be open to an interview with us." - Ben Gilbert explaining the extraordinary circumstances that led to securing the interview.
• At 28:25 - "The patents are controlled by just a few companies... the few companies that control the patents of LED would not let up at all." - Morris Chang explaining why TSMC’s venture into LED technology failed.
• At 29:34 - "I probably spent almost half of the time on how to resolve the problem with Nvidia." - Morris Chang highlighting his immediate focus upon his return as CEO in 2009.
• At 30:08 - "40-nanometer was a very important node in the progression of Moore's Law... if we do the 40 well, can we do the 28." - Dr. Chang on the strategic necessity of mastering the 40nm process.
• At 32:01 - "I sent an email to Jensen. I said, 'I'm coming to Silicon Valley next week on this date. I will be at your home at 6:00. Let's have just salad and pizza.'" - Dr. Chang describing his direct approach to meeting with Nvidia's CEO to resolve their dispute.
• At 34:16 - "Our offer is effective for 48 hours... If you don't accept the offer within 48 hours, we will have to go to an arbitrator." - Dr. Chang recounting the firm, take-it-or-leave-it terms of the settlement he offered to Nvidia.
• At 41:26 - "So I said, 'Let's pick 8%.' 8% regardless of whether there's a recession or not." - Dr. Chang on his decision to set a stable R&D budget to empower his team.
• At 46:34 - "There's a tide in the affairs of man which, taken at its flood, leads on to fortune. I decided that 28-nanometer was going to be our tide." - Dr. Chang quoting Shakespeare to explain his conviction about investing massively in the 28nm node.
• At 64:04 - "Actually, back in my mind, I was thinking of... the time when Kissinger was Nixon's national security advisor... And who had more power, you know? Kissinger." - Morris Chang's analogy for why a small business development team held immense strategic influence.
• At 78:32 - "A half-step is a detour... you would have to spend effort on the 20, which would of course help us on the natural next node, which was 16." - Dr. Chang explaining the technical challenge Apple presented by requesting a 20nm process.
• At 86:23 - "We will even confiscate the deposit if the time comes for him to take the wafers and he doesn't." - Dr. Chang recalling a strict policy from the dot-com bust to ensure customer commitment.
• At 96:08 - "This is a bet-the-company move. You're taking on a bunch of debt to go build the fabs to make this happen." - The interviewer characterizing the massive risk of the Apple deal, to which Dr. Chang replied, "I didn't think I would lose."
• At 98:28 - "Intel was no longer a name that you would, when you hear it, you would stand up and bow." - Morris Chang, explaining that by 2011, Intel's once-unassailable prestige had diminished.
• At 100:39 - "On customer trust, we thought that our customers trusted us more than Intel's customers trusted Intel." - Dr. Chang highlighting the core strategic advantage of the pure-play foundry model.
• At 103:57 - "Intel just does not know how to be a foundry." - Morris Chang, quoting what Tim Cook told him, which validated TSMC's business model and cemented the Apple partnership.
• At 111:28 - "...we delayed our 16-node development. And then, Samsung... they went ahead of us in the 16-nanometer development." - Dr. Chang explaining the significant trade-off TSMC made by focusing on Apple's 20nm needs.
• At 114:17 - "As soon as you're ready with your 16, we'll buy from you. We'll buy all of the needs from you." - Dr. Chang recounting Jeff Williams's promise to shift all of Apple's 16nm business to TSMC once their process was ready.
• At 134:40 - "Unlikely in your opinion." - Dr. Chang's immediate response when the host characterized TSMC's success as "unlikely," showcasing his deep-seated confidence.
• At 135:33 - "You've just been talking about learning curve. You know that... how could we plan to stop at two fabs?" - Dr. Chang explaining why the learning curve principle made the initial plan for only two fabs instantly obsolete.
• At 142:04 - "This idea that is genius in hindsight of not competing with your customers... at the time, I get the sense it was actually much more about what can we win at." - Ben Gilbert analyzing TSMC's core strategy as being born from pragmatic necessity.

Takeaways

• To resolve critical customer disputes, leaders should engage directly and personally, cutting through bureaucracy to preserve strategic partnerships.
• Build your business model on a foundation of customer trust; a commitment to not competing with your customers can be a more powerful moat than technology alone.
• Identify generational technological shifts and be willing to make massive, "bet-the-company" investments to secure a leadership position for the long term.
• Leverage the learning curve principle by pricing to win market share, as the resulting volume is the fastest path to sustainable cost leadership.
• In negotiations, presenting a fair but firm, time-bound offer can create the urgency needed to break a stalemate and reach a resolution.
• Provide R&D teams with stable, predictable budgets (e.g., a fixed percentage of revenue) to free them from annual negotiations and empower long-range planning.
• Recognize that focusing all resources on one strategic customer may create a temporary opening for competitors; have a clear plan to regain the lead.
• When planning succession, value strategic influence over the number of direct reports, and frame roles based on their potential impact on the company's future.
• In capital-intensive industries, continuous and aggressive reinvestment is not just for growth but is a crucial defensive strategy that creates a natural monopoly.
• Do not hesitate to exit ventures, even after investment, if the market structure (e.g., patent control by competitors) makes success improbable.
• When securing a massive new customer, be creative with financing (like using convertible bonds) to fund expansion without alarming the market or existing shareholders.
• Understand that a company's greatest competitive advantage may lie in the surrounding ecosystem of partners, suppliers, and talent, which is nearly impossible for others to replicate.