The First Post-Founder Year: TSMC and the Limits of Celestial
Morris Chang founded TSMC in 1987. He retired as CEO in 2018. In 2026, the founder’s active influence is essentially zero. The framework reads TSMC as Celestial Grade A. The reading is correct. What the reading cannot capture is whether the governance quality it measures is institutional or residual.
The Gold Standard
TSMC has appeared throughout this Foundation Series as the Celestial anchor. Note 1 opened with TSMC’s classification. The framework’s reading of TSMC is consistent, well-grounded, and correct.[1]
The T-axis reading falls in the high band. The B-axis reading falls in the high band. The board comprises eight directors, of whom five are independent — a ratio exceeding 50%. No single controlling shareholder exists. The largest shareholder — the National Development Fund, a government entity — holds approximately 6.4%. The R-axis reading falls in the medium-high band. The composite reading: Celestial, Grade A.
The Transition
Morris Chang’s succession architecture was deliberate and extended.[2] The 2005 CEO appointment of Rick Tsai — and Tsai’s subsequent departure in 2009 — demonstrated that the first transition attempt did not hold. The 2013-2014 installation of a dual-CEO structure, with Mark Liu and C.C. Wei sharing operational leadership, was the second attempt.
The 2018 retirement completed the founder’s operational exit. Mark Liu assumed the chairmanship. Wei assumed the sole CEO role. In June 2024, Mark Liu stepped down as chair. Wei Che-chia assumed both roles — chairman and CEO — reconsolidating the positions that Chang’s succession architecture had separated.
Samsung Chameleon, TSMC Celestial
Samsung: published Sample Scorecard (L1 public). TSMC: qualitative bands only (non-Sample issuer, exact scores NDA).
Archetype difference is firm-specific (R-axis governance history), not industry-driven.
Samsung Electronics reads as Chameleon, Grade B, with published scores of T 95, B 78, R 63, and a composite of 78.6. TSMC reads as Celestial, Grade A.[3]
The two companies operate in the same semiconductor industry. The archetype split is not industry-driven. It is firm-specific. The difference concentrates on the R-axis. Samsung’s R-axis score of 63 falls below the framework’s structural threshold of 70, reflecting Samsung’s governance history — the Lee family succession trail, the Samsung C&T–Cheil Industries merger of 2015, and the related-party transaction structures. TSMC carries none of this.
What Celestial Does Not Mean
Celestial is a structural classification. It indicates that a company’s governance profile clears the framework’s threshold on all three measured axes. It does not indicate that the company is safe, that governance failure is impossible, or that the current structural reading will persist.
For TSMC specifically, the counter-narrative is not that the Celestial reading is wrong. It is that the reading captures the governance architecture Morris Chang built, and the question the framework cannot answer is whether that architecture is self-sustaining without the founder’s presence — or whether the first post-founder year is the beginning of a trajectory the next production refresh will start to read.
The framework also does not read geopolitical exposure, supply-chain concentration, customer-revenue dependency, or regulatory risk across jurisdictions. A Celestial governance reading coexists with exposures on every other dimension of risk.
The Foundation Series Argument
This Case Study closes the Taiwan Foundation Series — six Notes and three Cases that map what the Apex G-Score framework reads in the Taiwanese listed market and where it stops reading.
Note 1 surfaced T-axis saturation. Note 2 found one regime on two boards. Note 3 traced cluster pathology. Note 4 documented the compliance wall. Notes 5 and 6 mapped the framework’s boundaries at capital return and sustainability disclosure. Case 1 backward-validated against the 2007 Rebar collapse. Case 2 traced the before and after of Tatung. Case 3 reads TSMC as the gold standard and discloses what the gold standard’s reading cannot tell us.
The series argues that Taiwanese governance is measurable, structurally uneven across axes, and not reducible to a single number. The framework reads with clarity because the disclosure regime is strong. The reading is honest because the framework discloses what it cannot see.
Apex Governance LLC · Taiwan Foundation Series · Case 3 of 3. Production data: GS-TWN v06 (scored 2026-04-27). TSMC (2330) reading: Celestial Grade A. TSMC is not in the published 3-firm Sample Scorecard (Samsung Electronics / Toyota Motor / Reliance Industries); axis-level scores reported in qualitative bands only. Taiwan’s annual report filing deadline falls at end of Q2; FY2025 financials were not universe-complete at the April 2026 scoring date. FY2024 is the latest fiscal year with full coverage across all 1,965 issuers. Korea reading: FY2025 cross-section. Taiwan reading: FY2024 cross-section. The one-year offset reflects FSC filing-deadline calendar differences (Taiwan Q2 vs Korea Q1). Per Apex G-Score IP guardrails: per-firm exact scores for non-Sample issuers are NDA.
Notes
- TSMC (2330) Apex G-Score v2.0 production reading: Celestial, Grade A. T-axis in the high band. B-axis in the high band. R-axis in the medium-high band. B-03 independent director ratio at the above-mandate tier (>50%). No active Kill Switch triggers. Specific axis scores are NDA per the framework’s non-Sample-Scorecard disclosure policy. ↩
- TSMC governance transition timeline compiled from TWSE public disclosure, annual reports, and published reporting. Key transitions: 2005 Rick Tsai CEO appointment; 2009 Morris Chang CEO return; 2013-2014 dual-CEO installation (Mark Liu, C.C. Wei); 2018 Morris Chang retirement (Liu chair, Wei CEO); 2024 Liu retirement (Wei assumes chair + CEO). ↩
- Samsung Electronics (005930 KS) Sample Scorecard: T 95, B 78, R 63, Composite 78.6, Grade B, Chameleon (R-weak). Published as one of the Apex G-Score framework’s three designated Sample Scorecard benchmarks. The archetype difference is driven by the R-axis: Samsung’s R 63 falls below the structural threshold of 70; TSMC’s R-axis reading clears the threshold. ↩
Apex Governance LLC (2026). The First Post-Founder Year: TSMC and the Limits of Celestial. Apex G-Score Taiwan Foundation Series, Case Study No. 3.https://apexgscore.com/research/taiwan/case-studies/tsmc-governance-transition
This public note summarizes selected market-level findings. Issuer-level T/B/R scores, archetype classifications, weak-axis tags, Kill Switch flags, monthly refresh history, and portfolio-level risk overlays are available only under institutional license.
This research is published by Apex Governance LLC as part of the Apex G-Score™ Taiwan Foundation Series. The Apex G-Score framework, TBR architecture, indicator design, and analytical conclusions are the work of Apex Governance LLC, led by Yunjung (Michelle) You, Ph.D., Founder & Chief Architect. Technical advisory support was provided by Wonsang You, Ph.D. (Dongduk Women’s University, LUNA Lab). AI tools supported code implementation, data structuring, drafting assistance, and editorial polish; they did not replace governance judgment or final analytical review.