Source item

The market rumors about TSMC assisting Intel in some form seem likely to persist. TSMC's edge in advanced node (production yields) stems from a complex interplay of factors—employee management, organizational culture, equipment and material procurement, patent and technology roadmap planning, market intelligence, customized services, and more—rather than something easily replicated through equipment purchases or technology transfers. So if TSMC is to help Intel boost its production yields—whether through consulting, equity investment, a joint venture, or otherwise—success depends on TSMC securing management control over Intel and implementing its management model. Without this, it's just a time-consuming effort with a high probability of failure. I’m not implying that TSMC’s management is the only successful approach in the foundry industry. However, TSMC only knows its own management approach, which has already proven successful. Accordingly, if the U.S. government asks TSMC to improve Intel’s production yields, the proposing team must ensure: 1. Can TSMC have management control of Intel? 2. If TSMC can have management control, organizational transformation is never easy. Is a certain level of failure acceptable, and is a waiting period required before seeing results (which might be slower than TSMC's own expansion in the U.S.)? 3. If TSMC can't have management control over Intel, then the failure probability is extremely high. Can the proposing team take responsibility for the failure? Of course, if this is just a formality to show action, with an acceptance that no results may come during Trump’s term, then the above analysis can be disregarded.