GlobalWafers Deep Value Investment Research

Hunting ten-year five-baggers among great growth stocks — pressing the upside question: "Can it get much bigger?"

• QHow large is its market ceiling? Is it expanding an existing pie, or creating an entirely new market?4/10

  A

  The ceiling is not low, but this is essentially about deepening a large, mature, moderately growing existing market, not creating a new one.

  GlobalWafers sells semiconductor silicon wafers, the substrate material for all chips. This is a mature business that has existed for decades. Demand rises and falls with the overall semiconductor industry; it is not opening a new category from scratch. At the industry level, SEMI data show that in 2025 global silicon wafer shipments grew 5.8% year over year to 12,973 million square inches, while industry revenue fell 1.2% year over year to about US$11.4 billion. An industry with rising volume and falling revenue suggests its ceiling is lifted mainly by semiconductor cycles and gradual process upgrades, rather than by exponential new demand.

  GlobalWafers is already a giant in this market. Citing the company's annual report, the research report says it is the world's third-largest silicon wafer manufacturer, with roughly 15%–20% share, while the top five suppliers together hold more than 90% of the market. This means two things. First, there is almost no room for it to grow simply by taking share from small players in the existing market; its stock-share rivals are similarly entrenched leaders such as Shin-Etsu and SUMCO. Second, for it to keep growing, it mainly needs the entire industry pie to expand, rather than cutting itself a bigger slice.

  The structural forces that can thicken this pie are process and application upgrades: AI, advanced logic, high-bandwidth memory (HBM), power devices, and automotive semiconductors, which drive demand for high-spec 300mm wafers, epitaxial wafers, and SiC/GaN compound semiconductors. The report also states plainly that this structural upgrade is currently concentrated mainly in 300mm and high-end categories, while 200mm and below and mature nodes remain weak. The upgrade dividend will not automatically or evenly land across all product lines.

  Measured against Baillie Gifford's LTGG yardstick: Baillie Gifford prefers companies that create entirely new markets with ceilings that are hard to see. GlobalWafers is a deep cultivator of an existing pie in a highly concentrated mature materials market growing at a single-digit pace. Its ceiling is sufficient to support the long-term operation of a strong company, but it lacks the imaginative room to open a new market an order of magnitude larger over ten years. The upfront conclusion holds: it is expanding an existing pie, not creating a new market.

  Jun 10, 2026
• QCan its revenue at least double over the next five years? Will growth be driven mainly by volume, price, or new businesses?3/10

  A

  Probably not. Doubling revenue over the next five years, or roughly 15% annualized, is a high bar for a strongly cyclical silicon wafer maker. Even if it grows, the growth will mainly come from volume and mix upgrades, not sustained price increases or entirely new businesses.

  Start with the historical base. GlobalWafers' revenue has not moved steadily upward. According to stockanalysis financials, revenue fell from highs of NT$70.29 billion in 2022 and NT$70.65 billion in 2023 to NT$62.63 billion in 2024 and NT$60.60 billion in 2025. In other words, revenue has been stepping down over the past three years, not compounding. To double from the 2025 base of NT$60.6 billion to roughly NT$120.0 billion in five years, it would need multiple consecutive years of double-digit high growth, something it has not achieved once in the past six years.

  The growth drivers break down into three parts, each with a ceiling:

  • Volume: This is the most realistic engine. SEMI shows that global shipments grew 5.8% in 2025 and rose another 13% year over year in the first quarter of 2026, so the industry is indeed recovering. GlobalWafers' large capacity expansion in recent years is a bet on volume growth, but the volume recovery is currently concentrated in 300mm, and the overall industry is still growing from single digits to low double digits. Volume alone is unlikely to support a doubling.
  • Price: Weak. The report notes that even as shipments recovered in 2025, the company's revenue still declined, while gross margin compressed from 43.2% in 2022 to 24.1% in 2025. This shows that even leading suppliers lacked strong pricing power at the cycle bottom, and price recovery lagged volume recovery.
  • New businesses: Compound semiconductors such as SiC and GaN are incremental directions, but they are still small and not yet sufficient to become the main force behind a revenue doubling.

  The report's own intrinsic value model also supports caution. Even in its "optimistic scenario", it assumes only 10%–12% growth for the first five years, then 5% for the next five years, which is far from a revenue doubling. The neutral scenario assumes only 7% growth in the first five years. In other words, the report's optimistic assumptions are already below the slope required for a five-year doubling.

  Using Baillie Gifford's LTGG yardstick, "doubling revenue in five years" is one of the hard thresholds for a great growth stock. GlobalWafers clearly does not meet it. It is a cyclical recovery play with operating leverage, not a structurally doubling growth stock. Put plainly: we should not pretend it can double just to fit a growth narrative.

  Jun 10, 2026
• QAfter five years, what will take over as the next growth engine? Does this "second curve" exist today?4/10

  A

  The "second curve" is only embryonic today, and it looks more like a structural upgrade inside the core business than an independent new growth pole.

  Baillie Gifford puts particular weight on whether the next engine that takes over after five years already exists and is growing today. For GlobalWafers, the honest answer is that it does not have a clear, independently viable second curve that can carry growth after the core silicon wafer business peaks. Its "future" is highly tied to upgrades within the same business, not the opening of a new battlefield.

  There are two possible candidates to take over, but neither qualifies as a true second curve:

  • Compound semiconductors (SiC, GaN): The report clearly states that the company's product line already covers compound semiconductors such as SiC and GaN, which aligns with electric vehicle and power device trends. But the report gives no revenue share or standalone growth rate for this business, and does not use it as a main valuation support. That suggests it is still small and not yet enough to become a "successor engine"; it is more like a new branch growing next to the main silicon wafer business.
  • High-spec 300mm / epitaxial wafers / wafers for HBM: This is the real direction pulling demand. SEMI says 2025 growth is coming from AI-driven advanced epitaxial wafers and polished wafers for HBM. But this is essentially the higher-end layer of the same silicon wafer pie. It is a structural upgrade of the core business rather than a second curve. It can improve mix and margins, but it does not change the fact that this is still a silicon wafer business.

  More importantly, GlobalWafers is currently directing almost all of its resources toward expanding capacity in the existing core business, not incubating a new species. The report shows that the company's PPE purchases, or capital expenditure, surged to NT$36.76 billion in 2023. The heavy borrowing and capital increases in recent years, with share count rising from 436,113,725 shares to 478,113,725 shares and dilution of about 9.6%, were all for expanding core production lines. This approach of putting all firepower in years three to seven into extending the main curve is exactly the opposite of Baillie Gifford's preferred pattern of investing early in a second curve.

  Compared with the Baillie Gifford archetype: in a true growth stock, the engine five years out often already has a visible independent form and steep slope today, such as platform extensions or new categories scaling. GlobalWafers does not have such a leg. Its "future" is the current business recovering cyclically plus moving upmarket, with both ceiling and pace locked by the same semiconductor cycle. This is a hard constraint on the company's growth profile and should not be elevated by a one-line appeal to the "SiC/GaN opportunity".

  Jun 10, 2026
• QWhat is its core competitive advantage? Will this moat widen or narrow over the next three to five years?5/10

  A

  Its core advantages are scale, customer qualification, and a global localized supply network. The moat is real but not wide. Over the next three to five years it will probably remain stable, with a slight local boost from geopolitical regionalization, but economic returns have not widened in tandem.

  First define the source of the moat. The report breaks GlobalWafers' barriers into several parts, with clear differences in strength:

  • Scale is relatively strong: It is the world's third-largest silicon wafer manufacturer, with roughly 15%–20% share, and the top five suppliers together hold more than 90%. Third-party industry research also confirms that Shin-Etsu, SUMCO, GlobalWafers, Siltronic, and SK Siltron form a highly concentrated group, with GlobalWafers firmly in third place. In an industry where new entrants can hardly replicate global production lines and customer qualifications, scale itself is a survival threshold.
  • Switching costs and process barriers are moderately strong: Customer qualification, yield matching, equipment introduction, and long-term collaboration create real friction in changing suppliers. Process integration in epitaxy, SiC/GaN, SOI, and other areas requires long accumulation. The report also mentions that the company has long-term contracts with customers, including at least one silicon wafer supply agreement extending to 2027, which provides some revenue visibility.
  • The global localized footprint is relatively strong: Eighteen plants across nine countries, combined with the trend toward geopolitical de-risking, increase the value of nearby delivery. This is the only direction in which the moat may widen slightly over the next three to five years.
  • Network effects and absolute pricing power are clearly weak: The report states directly that having more customers does not make the product more valuable to the next customer, so there is no network effect. Meanwhile, 2025 revenue declined, gross margin compressed from 43.2% in 2022 to 24.1%, and operating margin fell from 35.5% to 14.3% (financials), which proves that even leading suppliers lack through-cycle pricing power.

  The moat trend is the core of this question. My judgment is consistent with the report: overall stable over the next three to five years, with slight local strengthening, but not a widening moat. There are two sides to the reasoning. On one hand, geopolitics makes localized capacity more valuable, which benefits GlobalWafers. On the other hand, ROIC has clearly declined in the past two years. The report's rough operating calculation shows ROIC falling from about 15.8% in 2023 to about 4.5% in 2025. This means the barriers exist, but the economic moat, or the ability to turn barriers into excess returns on capital, is narrowing. Replicating the act of "making wafers" is easy in concept, but replicating "eighteen plants across nine countries plus long-term qualification by leading customers plus a full 3–12 inch product line" requires years and billions of US dollars in capital. The difficult-to-copy nature of the barrier is real.

  Using Baillie Gifford's yardstick: whether a moat will become "wider in the future" is key to whether a growth stock can compound over the long term. GlobalWafers' moat passes the durability test, but not the widening test. It is more like a deep, narrow moat whose returns are being eroded by the cycle and capacity expansion, rather than the kind that self-reinforces with scale and gets wider over time.

  Jun 10, 2026
• QIf its core business is disrupted, does it have the DNA to reinvent itself? How does it handle mistakes and bad news?4/10

  A

  The probability that its core business is "disrupted" is low, because silicon wafers are the irreplaceable physical substrate for chips. But its response to cycles and bad news leans toward heavy-asset endurance and regional diversification. It lacks the fast self-reinvention DNA of a light-capital platform, while its disclosure of mistakes is relatively candid.

  Start with whether it can be disrupted. Silicon wafers are not like software or business models that can be overturned overnight. They are the physical starting point for almost all semiconductor devices, and demand persists as long as the chip industry persists. SEMI data also show that industry volume has been growing moderately over the long term. The real "disruption" risk is not wholesale replacement by some new material, but technical structure migration: demand could rapidly concentrate in 300mm, advanced logic, power devices, and compound semiconductors. If the company's capacity structure or customer introduction speed lags, no amount of capital expenditure will buy returns. The report lists this as one of the risks it cares most about.

  Then look at the "DNA to reinvent itself", which is precisely the implicit premise Baillie Gifford adds to this chained question. GlobalWafers reinvents itself through a heavy-asset path: continuous capacity expansion, a regionalized footprint of eighteen plants across nine countries, and extension into higher-end categories such as SiC/GaN/SOI to adapt to demand migration. This path is real and effective, but it is inherently slow and expensive. The report shows that new plants entering sample introduction and ramp-up first pressure gross margin. In the first quarter of 2026, gross margin had already fallen to 20.8% and operating margin to 10.5% (quarterly financials), and the company stated that the main reasons were new plant ramp-up and rising energy and raw material costs. This kind of "turning by spending money to build capacity" is far from the reinvention ability Baillie Gifford prefers, namely light-capital businesses that can iterate and regenerate quickly. It is more like a heavy freighter that turns slowly.

  One major historical setback says a lot about how it handles mistakes. After the failed acquisition of Siltronic, the company recognized a EUR50 million termination fee in 2021, dragging down that year's earnings per share by about NT$3.5. It later generated fair value volatility from its Siltronic stake, disturbing net profit across multiple years. To its credit, the report's field checks did not find concealment or whitewashing by the company; these setbacks and accounting noise are traceable in public materials. Disclosure of bad news is relatively candid. Conversely, the failure of a major cross-border acquisition also exposed execution risk in its strategic bets.

  Overall judgment: the low risk of wholesale disruption is the safety cushion of this business. But "self-reinvention" relies on heavy-capital endurance, not agile regeneration. The pace is slow, the cost is high, and returns fluctuate with the cycle. This item cannot receive a high score. It has resilience, but not the Baillie Gifford-style DNA that can quickly grow a new form after the core is knocked out.

  Jun 10, 2026
• QDoes management, especially the founder, have a long-term view, and are its interests deeply aligned with the company? Is it willing to sacrifice current profit for five to ten years out?4/10

  A

  Management has strong industry vision and execution and has been deeply involved in the industry for a long time, but the point about "deep alignment with all minority shareholders" is weak. This is a structure led by a controlling parent company, and in recent years it has been more willing to sacrifice per-share value for expansion than to sacrifice current profit for five to ten years out in a way that rewards shareholders.

  Start with the positives. Chairwoman Doris Hsu has decades of experience in the semiconductor industry and has long led parent company SAS. Her industry understanding and execution are both solid. Public reporting also confirms that she is the chairwoman of both GlobalWafers and SAS, and led major strategic moves such as the US$4.5 billion acquisition bid for Siltronic. The report also notes that governance is not crude: independent directors account for 50% of the board, the compensation committee consists of four independent directors, and audit, risk, and information disclosure are relatively standardized. This is a management team with industry depth and the courage to make large strategic decisions.

  But on the question of "alignment", GlobalWafers does not answer well. There are two hard constraints:

  • It is not a company where the founder or management owns a high personal stake and is fully aligned with minority shareholders. The report states that as of March 26, 2026, parent company SAS, represented by Doris Hsu, held 46.64% of the shares, forming a clear controlling structure. Multiple reports also confirm that SAS holds about 47% and that Doris Hsu leads both companies. This means management is first accountable to the controlling parent company, and may not be fully aligned with all minority shareholders. That is fundamentally different from the Baillie Gifford-preferred pattern in which a founder has their net worth tied to the company and sits in the same boat as small shareholders.
  • The main axis of recent capital allocation has been expansion rather than shareholder returns. The report notes that the company's share count rose from 436,113,725 shares (April 2024) to 478,113,725 shares, dilution of about 9.6%, mainly from the Crystalwise share swap and subsequent cash capital increase. The main axis has been capacity expansion, borrowing, and capital raising, not buybacks or improving intrinsic value per share.

  As for whether it is "willing to sacrifice current profit for five to ten years out", the Baillie Gifford trait that matters most, GlobalWafers presents a distorted version of "sacrificing the present". It is indeed investing heavily in future capacity and depressing gross margin in the short term (new plant ramp-up drove first-quarter 2026 gross margin down to 20.8%), so from that perspective it is willing to invest for the long term. The problem is that the beneficiaries of this sacrifice tilt more toward "expanding enterprise scale and the controlling group's territory" than toward "expanding per-share value". The report repeatedly stresses that growth in enterprise value is not excellent capital allocation for long-term value investors if it does not also translate into growth in per-share value.

  Honest conclusion: management capability is credible and there are no obvious integrity problems, but under Baillie Gifford's core standard of "deep alignment with all shareholders and putting per-share value first", GlobalWafers is only neutral. Capability scores high; alignment is moderately weak.

  Jun 10, 2026
• QIf it disappeared tomorrow, how much would customers miss it? Is its growth model sustainable and not dependent on harming society or regulation?6/10

  A

  If it disappeared tomorrow, leading chip customers would miss it a lot. As the world's third-largest silicon wafer source with 15%–20% of supply, it would be hard to replace seamlessly in the short term. Its growth model is also healthy and does not depend on harming society or challenging regulation. Instead, it benefits from the policy direction of supply chain security. Both points, indispensability and social sustainability, stand up.

  Start with indispensability. Silicon wafers are the physical starting point of chip manufacturing, and qualified suppliers are scarce. Citing the company's annual report, the research report says the top five suppliers together control more than 90% of the global market, while GlobalWafers is firmly third with roughly 15%–20% share. Industry research also confirms this highly concentrated structure. Silicon wafers used by chipmakers go through lengthy qualification, yield matching, and equipment introduction, and cannot be switched on short notice. The report notes that customers prefer to maintain long-term relationships with reliable suppliers, and the company also has long-term contracts extending to 2027. So if it suddenly disappeared, customers would face real supply gaps and the cost of redoing qualification. The "degree of being missed" would be high. This is GlobalWafers' true strength in this question.

  But it is important to distinguish "being missed" from "irreplaceable". It is one of a few indispensable players, not the only one. Shin-Etsu and SUMCO are larger and could theoretically absorb part of the gap over time, though that would require time and capacity expansion. The report also points out that being global top three does not automatically mean strong pricing power. The company's 2025 revenue decline and gross margin compression from 43.2% to 24.1% (financials) prove this. In other words, its "indispensability" lies in supply security and qualification stickiness, not the extreme dependence where customers cannot leave it and it can raise prices at will.

  Then look at social and regulatory sustainability, the second premise Baillie Gifford adds to this question. GlobalWafers' growth model is quite clean. It earns money by selling materials, expanding capacity, and completing qualifications. It does not harm users, drain social trust, or operate in regulatory gray zones. On the contrary, against the global backdrop of semiconductor supply chain de-risking and localization, its footprint of eighteen plants across nine countries and nearby supply for regional customers align with, rather than fight, regulation and industrial policy. Governments in multiple countries instead encourage local plants through subsidies (the report mentions a one-off subsidy that lifted margins in the fourth quarter of 2025). The larger its business becomes, the more positive its contribution to the resilience of the global chip supply chain.

  Overall judgment: indispensability is strong. It is one of a few qualified suppliers, with qualification and long-term contracts creating real switching friction, though it is not unique and does not have unlimited pricing power. Social/regulatory sustainability is strong. Its growth model is healthy and aligned with the policy direction of supply chain security. This is one of the few dimensions where GlobalWafers earns a relatively high evaluation under the Baillie Gifford framework.

  Jun 10, 2026
• QWhat are the unit economics of this business, in gross margin and incremental returns? Do they improve or deteriorate with scale? Where does the money it earns go?4/10

  A

  The unit economics are a heavy-asset model that looks very attractive at the cycle top and weakens clearly at the bottom. Larger scale has not produced sustained improvement in unit economics; instead, expansion and higher depreciation have made them worse in recent years. The money earned has mainly gone to capacity expansion and debt repayment, not shareholders.

  Start with the large swings in gross margin and profitability, the core of judging unit economics. According to stockanalysis financials, GlobalWafers' gross margin fell from a 2022 peak of 43.2% to 37.4% in 2023, 31.6% in 2024, and 24.1% in 2025. Operating margin over the same period fell from 35.5% to 14.3%. By the first quarter of 2026, reported gross margin had fallen further to 20.8% and operating margin to 10.5%. For a healthy growth stock, unit economics should stabilize or improve as scale expands. GlobalWafers is the opposite: scale is expanding while unit economics are retreating.

  Then look at incremental returns (ROIC), the most damaging evidence for this question. Using a rough operating basis, the report estimates that the company's ROIC fell from about 15.8% in 2023 to about 9.4% in 2024, then to about 4.5% in 2025. In other words, the returns on the large amounts of incremental capital deployed in recent years are decaying rapidly. This is the typical feature of a heavy-asset, strongly cyclical business: new plants consume profit first during sample introduction and ramp-up, and only potentially release returns after utilization rises. The report states clearly that one of the main reasons for gross margin decline was new plant ramp-up plus rising energy and raw material costs. Do unit economics improve or deteriorate after scale increases? At this stage, the answer is deteriorate.

  One point of credit is worth adding: operating cash flow has always been positive. Core operating profit is real cash, not accounting puffiness. But free cash flow is extremely uneven. The report shows operating cash flow falling from NT$37.57 billion in 2022 (the peak, with verified FCF of about NT$25.2 billion) to NT$12.74 billion in 2025, while in 2023 capacity-expansion capex surged to NT$36.76 billion, driving FCF directly negative to about -NT$18.2 billion. Net profit also contains fair value noise from the Siltronic stake (valuation loss of NT$10.13 billion in 2022, valuation gain of NT$3.00 billion in 2023, and first-quarter 2026 net profit up 30% year over year partly due to the rebound in Siltronic's share price). To assess unit economics, this portion must be stripped out, with focus kept on operating cash flow.

  Where does the money earned go? The answer is clear: capacity expansion and debt repayment come before shareholder returns. The report states that by the end of 2025 the company had short-term borrowings of NT$31.01 billion and long-term borrowings of NT$43.24 billion, with net debt/EBITDA of about 3.2 times. In recent years it also increased share capital from 436,113,725 shares to 478,113,725 shares through capital raising, dilution of about 9.6%. Capital has gone to production lines and the balance sheet, not dividends or buybacks.

  Using Baillie Gifford's yardstick: the unit economics of a great growth stock should get better with scale, generate high incremental returns, and leave room to give back. GlobalWafers today is a case of unit economics weakening with the cycle and expansion, incremental returns decaying, and cash being absorbed by capacity. This item is clearly weak and should not be lifted just because it once made a lot of money at the top.

  Jun 10, 2026
• QWhat conditions must hold simultaneously for it to rise fivefold in ten years? Are those conditions realistic? What expectations are embedded in today's share price?2/10

  A

  For the stock to rise fivefold in ten years, "earnings up fivefold + no major valuation compression" must both happen. Yet today it trades at about 50 times earnings while revenue is declining and ROIC has slid to about 4.5% at the cycle bottom. The chance that both conditions are met is low. Today's share price embeds optimistic expectations that new plants ramp smoothly to full utilization, gross margin returns above 30%, and the market remains willing to give a cyclical stock a high multiple for a long time. It does not embed a margin of safety.

  First anchor today's price. According to the stockanalysis real-time snapshot, 6488 last traded around NT$752 (intraday on 2026-06-10), corresponding to a market capitalization of about NT$383.9 billion and a trailing P/E of about 50 times. The report uses the June 9 closing price of NT$803 and 2025 EPS of about NT$15.29, implying a static P/E of about 52.5 times and P/B of about 4.1 times. Note that its 52-week range is NT$278.50–1,040 (same source), with huge volatility. That itself suggests the current price sits in a relatively elevated zone of cycle and sentiment.

  What conditions must hold simultaneously for a fivefold gain in ten years? At least four items are needed, and they must stack together:

  1. Earnings themselves must approach fivefold growth: Starting from a high 50 times multiple, if the exit P/E falls back to a normal cyclical-stock range such as 15–20 times, then a fivefold share price gain would almost require net profit to rise 12–15 times. That is unrealistic for a company with 2025 net profit of only NT$7.3 billion and declining revenue.
  2. Gross margin must recover structurally and hold: The report notes that after excluding major expansion bases and the Siltronic fair value impact, management's simulated basis can reach a gross margin of 30.9%, operating margin of 21.4%, and EPS of 4.96. But this is a simulation, not cash flow in hand. Actual reported gross margin in the first quarter of 2026 was only 20.8% (quarterly financials).
  3. No continued dilution + deleveraging: Net debt/EBITDA needs to fall from about 3.2 times to below 1.5 times, and without relying on refinancing. Yet the company has already diluted its share capital by about 9.6% in recent years.
  4. The market must remain willing to assign cyclical materials stocks high multiples for the long term: This is the least controllable condition.

  What expectations are embedded in today's share price? The report's reverse calculation is the clearest. Its own DCF gives a reasonable intrinsic value range of NT$240–360, with even the optimistic range only at NT$450–650, and an ideal buy range of NT$200–280. Even the third-party Simply Wall St fair value estimate is only about NT$613, all below the current price around 752. This means the current price has already pulled forward the realization of the optimistic scenario. The report's reverse expected annualized returns at the current price also tell the story: about -12% in the conservative case, about -6% in the neutral case, and only +2% to +4% in the optimistic case.

  Honest conclusion: a fivefold gain in ten years requires a high starting valuation and cycle-bottom earnings to both reverse upward and remain sustained, which is not realistic. Today's price around 752 is not leaving room for fivefold upside; it is paying for "everything going right". That runs directly against Baillie Gifford's standard of finding ten-year fivefold opportunities with blue-sky room. GlobalWafers lacks both the earnings engine for fivefold growth and a valuation that leaves room for upside.

  Jun 10, 2026
• QWhy has the market not realized all this yet? Is it too hard to understand, too overlooked, or too long-term? What will become the "narrative inflection point"?2/10

  A

  The market has not failed to recognize it. More precisely, the market already understands it fully, and perhaps optimistically. This is not an obscure stock that is misunderstood or ignored, but a high-quality cyclical asset that has been chased again as sentiment recovered, with valuation running ahead of fundamentals. The real "narrative inflection point" is not when the market discovers its strengths, but when new-plant profits and the cycle direction are confirmed or falsified.

  First correct the premise embedded in the question. When Baillie Gifford asks why the market has not realized it yet, the implied setup is an undervalued stock with a positive perception gap. GlobalWafers is in the opposite situation. The evidence is that its price and valuation are already expensive: current price around NT$752, trailing P/E about 50 times (stockanalysis), P/B about 4.1 times, and the stock has risen over the past year from a 52-week low of NT$278.50 toward a high near NT$1,040. This is not "ignored"; it is "enthusiastically priced". The report's judgment is direct: at the current level, it looks more like a high-quality cyclical asset re-embraced by the market as sentiment warmed, rather than a business obviously below intrinsic value. The perception gap may even be negative.

  So which of "too hard to understand / too overlooked / too long-term" applies? Strictly speaking, none of the three really applies:

  • Not too hard to understand: Silicon wafers are a mature, transparent industry with ample institutional coverage. SEMI industry data, company financials, and sell-side models are all readily available, and the market understands its cyclical nature clearly.
  • Not overlooked: A 50 times P/E itself shows the market has given it respect and a premium far above ordinary cyclical stocks.
  • The only slightly relevant idea is "too long-term", but in the opposite direction: The market may be looking too far ahead and too optimistically, discounting the long-term simulation of "new plants at full utilization and gross margin back to 30%+" into today's share price. The report repeatedly stresses that management's attractive simulated basis, with gross margin of 30.9% after excluding expansion and Siltronic effects, is an upside option, not currently distributable cash flow. Actual reported gross margin in the first quarter of 2026 was only 20.8% (quarterly financials).

  What will become the narrative inflection point? This is the key Baillie Gifford addition to the question. For a stock already priced optimistically, the inflection point is a two-way confirmation or falsification:

  • Confirming the upside narrative: Over the next 6–8 quarters, new plant ramp-up ends, reported gross margin stabilizes around 30%, operating cash flow recovers to the NT$18.0–22.0 billion range, net debt/EBITDA falls below 1.5 times without refinancing, and EPS growth comes from utilization rather than Siltronic revaluation. If these appear consecutively, the "expensiveness" of the current price will be redefined.
  • Breaking the upside narrative: Recovery falls short of expectations, returns on expansion remain low, another large capital increase dilutes shareholders, or customer introduction is delayed. The report warns that if the market reprices it at a mid-cycle valuation of 1.5–2.0 times PB, a share price returning to NT$280–390 is not unimaginable, with a 50%–70% drawdown from the current price in an extreme scenario.

  Honest conclusion: this question needs to be read in reverse. The market has not underestimated it; it may have overestimated it. The narrative inflection point is not "the market finally discovers its value", but whether new-plant profits and the cycle direction confirm or falsify today's optimistic expectations. At a price around ~752, betting on confirmation means accepting asymmetric downside risk from narrative falsification and valuation normalization.

  Jun 10, 2026