Advanced Micro-Fabrication Equipment Inc. China(688012) · AI Semiconductor Equipment

AMEC (688012): China's Domestic Etch Equipment Leader, a Great Company at a Rich Price

Other languages
Quick ReadPlain-language overview · read this first

This is a company that makes high-end chip manufacturing equipment. It is called AMEC, the domestic leader in etching equipment. Etching means using machines to carve chip circuits layer by layer on wafers with high precision. The report's stance is clear: this is a good company, but the current price is too expensive. The rating is Watch, meaning observe first and do not rush to buy.

It mainly earns money by selling this equipment to chip fabs. The advantage is that once its equipment is installed in a customer's critical production line and has passed mass-production validation, switching to another supplier becomes difficult, which creates strong stickiness. Its growth in recent years has indeed been rapid: in 2025, revenue was about RMB 12.385 billion and profit was about RMB 2.111 billion.

The problem lies in both valuation and cash. The current share price is about 265 yuan, and the company's total market value is about RMB 248.7 billion, which means the price has already built in the smoothest possible story for the next ten years. More importantly, accounting profit looks decent, but the cash that truly lands in the pocket is much thinner. Most of the money is being reinvested in R&D and new factory buildings. In other words, reported earnings look attractive, but freely deployable cash is limited.

According to the report's calculations, a more reasonable buy-in range would be 110 to 150 yuan. Above 250 yuan, the stock is clearly expensive, and the current price sits on the expensive side. The key risk to watch is this: if growth or profitability falls even slightly short of expectations, such a high price could pull back sharply at any time. The report does not tell you to buy or sell. It simply reminds you that this is a good company, but not necessarily a good price today. For conservative investors, waiting is more appropriate than chasing.

The above is only a plain-language explanation of this research report and is not investment advice. The stock market involves risk; enter the market with caution.

Lead

AMEC is a leading Chinese high-end semiconductor equipment company, anchored in plasma etch tools and expanding into thin-film deposition, MOCVD, and adjacent platforms. Its long-term case is driven by domestic substitution and advanced-node upgrades, with 2025 revenue of about RMB 12.385 billion and net profit attributable to shareholders of about RMB 2.111 billion, but customer concentration is high and R&D spending is roughly 30% of revenue. Report Rating Watch: a valuable business whose current price already discounts a large share of future success.

Full report

Prices in the article are as of publication; see the valuation band above for the live price.

Conclusion First

Initial rating: Watch. If AMEC is viewed as a business to own for more than ten years, rather than as a short-term trading ticker, my judgment is straightforward: it is a highly valuable company, but at the current price it is not a very attractive entry point. The company operates in a market with high barriers, strong long-term demand, and clear tailwinds from domestic substitution and advanced-node upgrades. Yet it currently looks more like a fast-expanding R&D and delivery machine than a mature cash cow that can steadily produce large amounts of distributable cash flow. As of 2026-06-09, Reuters/LSEG delayed quotes showed its share price at about RMB 265.47 per share and market capitalization at about RMB 248.7 billion, corresponding to P/S of about 18.95x, adjusted P/E of about 90.71x, and P/B of about 10.20x. In 2025, the company generated revenue of RMB 12.385 billion, net profit attributable to shareholders of RMB 2.111 billion, and operating cash flow of RMB 2.295 billion, but recurring net profit was only about RMB 1.550 billion. If cash related to capitalized development expenditure is also treated as an unavoidable operating investment, 2025 was closer to a RMB 421 million “strict owner earnings proxy.” This means the current price already embeds a large amount of future success.

Is there a margin of safety at the current price: not obvious. For conservative and balanced investors, I would rather define AMEC as “an excellent company, but currently a good company paired with a high price.” If you are a long-term growth investor who deeply understands semiconductor equipment and is willing to track orders, contract liabilities, R&D capitalization, and the pace of domestic substitution, it deserves a high-priority watchlist position. If you prefer “Buffett-style” simple businesses, stable free cash flow, and clear valuation anchors, AMEC is not comfortable at the moment.

The key uncertainties are mainly threefold. First, can the company ultimately convert high R&D spending and platform expansion into real cash flow at a pace that keeps up with revenue and profit growth? Second, can thin-film, EPI, metrology/inspection, and the planned wet-process/CMP expansion beyond etch truly become second and third growth curves? Third, has the current valuation already discounted most of the smoothest execution path over the next decade?

The table below condenses my judgment into one line:

Item Judgment
Investment rating Watch
Margin of safety Not obvious
Best suited for Deep-research, long-term growth investors
Less suited for Conservative, cash-flow-focused investors who prefer simple business models
Biggest uncertainty Cash-flow quality, success of platform expansion, excessive valuation

Business Understanding and Industry Structure

From a business-model perspective, AMEC is not hard to understand. It essentially sells high-end semiconductor and pan-semiconductor microfabrication equipment, with core products including etch equipment, MOCVD equipment, and thin-film deposition equipment, while also providing related equipment and services. Reuters/LSEG's business description points in the same direction: the company mainly engages in R&D, manufacturing, and sales of semiconductor equipment, with products covering plasma etch, MOCVD, LPCVD, ALD, and others. The company's own public materials show that the most important growth in 2025 still came from etch equipment. Etch equipment sales in 2025 were about RMB 9.832 billion, up about 35.12% year over year; combined LPCVD/ALD sales within thin-film equipment were about RMB 506 million, up about 224.23% year over year. Looking back, etch equipment sales in 2024 were about RMB 7.277 billion, while thin-film equipment had already secured about RMB 476 million of batch orders and generated about RMB 156 million of sales revenue; in 2023, etch equipment sales were about RMB 4.703 billion, and MOCVD equipment sales were about RMB 462 million. This shows that the company is evolving from an “etch leader plus legacy LED/MOCVD business” into a structure centered on etch, with thin-film and other new platforms beginning to scale.

How does it make money? The core is recognizing revenue from equipment sales, supplemented by follow-on demand from spare parts, services, and installed tools. But this is not a consumer-goods business, nor is it SaaS. Its revenue naturally has the features of large projects, long validation cycles, and customer capex cycles, so it will not be as smooth, recurring, or predictable as Coca-Cola. The positive side is that once equipment enters a customer's critical process and passes batch validation, the customer's willingness to switch drops sharply. The less favorable side is that annual delivery, acceptance, and customer expansion schedules can make revenue and cash flow volatile. By the end of 2025, the company had accumulated more than 7,800 reaction chambers in mass production across more than 170 customer production lines in China and overseas, with cumulative etch equipment shipments exceeding 6,800 units. This means the installed base and relationship stickiness are strengthening. At the same time, by the end of 2025 the company had contract liabilities of about RMB 3.04 billion and inventory of about RMB 7.17 billion, which also shows that the order and delivery cycle is not light.

Who are the customers? They are mainly wafer fabs and pan-semiconductor customers, and customer concentration is quite high. In 2025, the top five customers accounted for about 75.00% of annual sales; in 2024, the share was about 70.22%. This is not surprising for an equipment company still in the domestic-substitution and platform-expansion stage, because leading logic, memory, and specialty-process customers are themselves highly concentrated. But it means AMEC's business does not have the dispersed customer structure typical of consumer goods. It looks more like “a few large customers plus high-intensity co-development.” The supplier side is less concentrated: in 2025, the top five suppliers accounted for about 27.01% of annual procurement, versus about 28.40% in 2024. In sales model, the company mainly uses direct sales; in Europe, where customers are more fragmented, it sells through agents. Overall, customer concentration risk is clearly higher than supplier concentration risk.

From an industry perspective, AMEC sits in the semiconductor equipment industry, where structural growth and strong cyclical volatility coexist. SEMI data show that global semiconductor equipment sales/billings reached about USD 135.0 billion in 2025, up 15% year over year; global equipment billings rose 14% year over year in Q1 2026; and at the end of 2024, SEMI forecast that global semiconductor equipment sales would reach USD 139.0 billion in 2026 and USD 156.0 billion in 2027. More importantly, SEMI's view on foundry/logic applications within wafer fab equipment was that 2026 would grow 15% year over year to about USD 69.3 billion, driven by advanced processes, GAA architecture migration, and capacity expansion. In other words, this is not a declining industry. It is a good industry with very strong long-term demand and very large short-term swings.

In terms of competitive structure, the strongest domestic comparable is closer to NAURA Technology Group. Reuters data show that NAURA's product line covers a wider range of core process equipment, including etch, thin-film deposition, thermal processing, wet process, and ion implantation, giving it clearly broader platform width than AMEC. AMEC's strongest long suit is etch, while it is advancing into thin-film, EPI, metrology/inspection, and other areas. China's policy support for domestic equipment substitution is also strengthening. Reuters cited people familiar with the matter in late 2025 as saying that China required newly built semiconductor fabs to use more than 50% domestic equipment. This could be an important external tailwind for AMEC, but it also means part of its growth logic is not purely market-driven and carries some policy dependence. Overall, I rate the industry's attractiveness at 4/5: the runway is long and the ceiling is high, but cycles, technology iterations, and policy variables are all strong.

From a long-term owner's perspective, my answer to “is this a business I can understand?” is: it can be understood at the economic-logic level, but the technical details are not simple. You can understand why it makes money, why it is hard to do, and why customers do not switch easily. But you may not be able to judge like an engineer whether a particular generation of ICP, CCP, LPCVD, or high-selectivity etch route is genuinely leading. For Buffett-style investing, this matters. Therefore, my “business understandability” score is 3/5. If the stock market were closed for five years, I would be willing to hold it only at a lower price and with stronger evidence of cash-flow quality. At the current price, I am not comfortable enough.

Moat and Management

AMEC's moat is not a single brand monopoly. It is an engineering moat formed by the stacking of multiple capabilities. The most important factors are not advertising brand, but process know-how, customer validation cycles, field service capability, installed base, original design, and sustained high R&D investment. By the end of 2025, the company had accumulated more than 7,800 reaction chambers in mass production on customer lines, with cumulative etch equipment shipments exceeding 6,800 units; R&D investment in 2025 was about RMB 3.744 billion, or about 30.23% of revenue; in Q1 2026, R&D expenses still reached RMB 908 million, or 31.14% of revenue. For this type of company, the truly hard-to-replicate part is not “buying a few machine tools and setting up a factory.” It is years of iteration to grind out equipment performance, stability, yield, consistency, and the customer's process window bit by bit. In my view, genuinely replicating AMEC's core capabilities would require at least 5 to more than 10 years of sustained, high-intensity capital and customer co-development. This is an inference, but it rests on solid foundations: very high R&D intensity, long validation cycles, a large installed base, and the pace of advanced-process migration are all visible.

If I judge the ten moat types one by one, my conclusion is closer to “moderately strong, but still under construction.” Brand advantage is medium: the company has already built a fairly strong technical brand in domestic high-end etch, but its global brand power still lags far behind international leaders. Cost advantage is limited: the company relies more on performance, delivery, and local service than on pure low price. Scale advantage is moderately strong: the installed base and number of reaction chambers create a real service and learning curve. Network effects are basically absent. Switching costs are fairly strong, because once customers introduce equipment into critical processes, switching involves revalidation, yield, and production-line risk. Channel advantage is medium: direct sales plus field engineering services are important, but not impossible to replicate. Patent, regulatory, and qualification barriers are fairly strong, although the stronger barrier is really “process qualification.” Data advantage is weak. Corporate culture and operating capability are fairly strong, reflected in original design, sustained high R&D investment, and the pace of new product development. Capital allocation capability is medium: the direction is right, but the return has not yet been fully proven. Taken together, I score AMEC's moat strength at 3.5/5.

Is the moat widening, stable, or narrowing? My answer is: in the core etch business, the moat is probably widening slowly; in the larger proposition of becoming a “platform equipment company,” the moat is still being built. In other words, AMEC is already strong in single-point breakthroughs, but it has not yet grown into an equipment group with platform breadth like NAURA. Reuters' business description of NAURA shows that its product lines cover multiple core steps, including etch, deposition, thermal processing, wet process, and ion implantation. AMEC is advancing toward thin-film, EPI, metrology/inspection, and even wet-process/CMP, but this platform capability has not yet fully converted into stable cash flow and a wider return moat. Put differently, AMEC is “a good company in a good industry,” but it may not yet be “a fully mature great enterprise.”

On management, my basic assessment of the founder and core team is positive. Reuters' list of company leadership shows Gerald Z. Yin as chairman and general manager, and the core executive structure is stable. For years, the company has adhered to a strategy of high R&D, original design, and synchronous iteration with leading customers. The direction is long-term oriented and does not look like a clear pursuit of short-term profit maximization, unlike some equipment companies. The issue is that economic alignment is not especially strong: as of 2025-03-31, Gerald Z. Yin held about 11.20 million shares, or about 1.79%; the 2025 annual report disclosed that the two largest shareholders held about 14.93% and 10.94%, respectively, and the ownership structure is relatively dispersed. In other words, management credibility comes more from industry reputation, technical route, and long-term execution than from direct economic alignment created by very high ownership as in many family-controlled businesses.

Capital allocation is the issue in AMEC that deserves the closest look and is also most easily covered up by a “good story.” The company's cash mainly goes to three things: R&D, capacity and base expansion, and platform acquisitions/layout. R&D investment in 2025 was about RMB 3.744 billion. In the first half of 2025, the company had disclosed that production and R&D bases of about 140,000 square meters in Nanchang and about 180,000 square meters in Shanghai Lingang had been put into use, and it planned to build new production and R&D bases in Guangzhou Zengcheng and Chengdu High-tech Zone. In addition, the company is advancing a transaction to acquire control of Hangzhou Zhongsi through issuing shares and paying cash. Strategically, these moves all make sense, because semiconductor equipment is a classic reinvestment-driven industry. But for shareholders, the question is whether these investments will turn into per-share intrinsic value in the future, rather than merely producing a larger scale and a better story.

In shareholder returns, the company is more “reinvestment-oriented” and does not rely on high dividends or buybacks to prove capital discipline. Public dividend records show that the company paid RMB 2.0 per 10 shares for the 2023 interim period, RMB 3.0 per 10 shares at year-end 2023 and year-end 2024, and RMB 3.5 per 10 shares at year-end 2025, while also converting capital reserve into 4.9 shares per 10 shares. On Reuters/LSEG's basis, the dividend yield was only about 0.09%. Separately, in May 2026, the company added about 1.9231 million listed shares due to equity-incentive vesting, increasing share capital from 626.9 million shares to 628.8 million shares. After the 4.9-for-10 bonus issue, Reuters/LSEG shows current issued shares under the float basis at about 936.97 million shares. The bonus issue is not economic dilution, but the equity incentive does mean continued small dilution. My conclusion is: the direction of capital allocation is not bad, but constraints on per-share value are not strong enough, and dividends and buybacks are not major highlights. I score “management and capital allocation” at 3.5/5.

Financial Quality and Owner Earnings

First, a note on method. Because SSE PDFs are not stable in search retrieval, the key figures below are mainly cross-checked from three types of sources: Reuters/LSEG structured financial data, search snippets from company preliminary results/quarterly reports/annual reports, and mirrored company announcements. Where an item cannot be stably verified, I explicitly write “additional information needed” rather than forcing in a number.

In terms of growth, AMEC has been very impressive over the past few years. Revenue in 2019 was about RMB 1.947 billion, and net profit attributable to shareholders was about RMB 189 million; in 2020, they were about RMB 2.273 billion and RMB 492 million; in 2021, about RMB 3.108 billion and RMB 1.011 billion; in 2022, about RMB 4.740 billion and RMB 1.170 billion; in 2023, about RMB 6.264 billion and RMB 1.786 billion; in 2024, about RMB 9.065 billion and RMB 1.616 billion; and in 2025, about RMB 12.385 billion and RMB 2.111 billion. On this basis, 2019-2025 revenue CAGR was about 36%, and 2021-2025 revenue CAGR was about 41%. This is an extremely rare high-growth curve.

The other side of the growth curve is that profit quality and cash-flow quality are not as easy as the stock-price narrative suggests. In 2023-2025, on Reuters/LSEG's basis, the company's gross profit was about RMB 2.859 billion, RMB 3.603 billion, and RMB 4.772 billion, respectively, corresponding to gross margins of about 45.6%, 39.8%, and 38.5%. Net margins attributable to shareholders were about 28.5%, 17.8%, and 17.0%. Looking at recurring net profit, which better represents operating reality, 2023 was about RMB 1.19 billion, 2024 about RMB 1.388 billion, and 2025 about RMB 1.550 billion, implying a recurring net margin of only about 12.5% in 2025. In particular, the company disclosed equity investment income of about RMB 607 million included in non-recurring gains and losses in 2025, which means reported profit is not the same as operating cash profit.

The table below summarizes the financial-quality metrics I consider most important. Revenue, gross profit, net profit attributable to shareholders, total assets, total liabilities, and operating cash flow come from Reuters/LSEG; recurring net profit, cash outlays for long-term assets such as fixed assets/intangible assets, and cash related to capitalized development expenditure come from company annual-report/quarterly-report search snippets; gross margin, net margin, debt-to-asset ratio, standard free cash flow, and the “strict owner earnings proxy” are calculated from them.

Year Revenue YoY Gross margin Net profit attributable to shareholders Recurring net profit Operating cash flow Standard FCF Strict OE proxy Total assets Total liabilities Debt-to-asset ratio
2023 RMB 6.264 billion 32.2% 45.6% RMB 1.786 billion About RMB 1.19 billion -RMB 977 million -RMB 1.847 billion Additional information needed RMB 21.526 billion RMB 3.699 billion 17.2%
2024 RMB 9.065 billion 44.7% 39.8% RMB 1.616 billion RMB 1.388 billion RMB 1.458 billion RMB 563 million -RMB 281 million RMB 26.218 billion RMB 6.481 billion 24.7%
2025 RMB 12.385 billion 36.6% 38.5% RMB 2.111 billion RMB 1.550 billion RMB 2.295 billion RMB 1.440 billion RMB 421 million RMB 29.846 billion RMB 7.151 billion 24.0%

For long-term investors, three points in this table matter most. First, revenue growth is very strong, but gross margin is declining, which shows the company has not automatically gained stronger and stronger pricing power from domestic substitution, at least not yet. Second, operating cash flow is improving, but free cash flow is not stable: if you treat only cash outlays for fixed assets/intangible assets as capex, then standard FCF did turn positive in 2024 and 2025; but if you also treat “cash paid for projects directly related to development expenditure meeting capitalization conditions” as necessary operating investment, then 2024 was still negative, and 2025 was only RMB 421 million. Third, the balance sheet itself is robust, but that does not automatically mean the equity has a margin of safety, because a solid balance sheet and a cheap share price are different things.

Next, working capital. At the end of 2023, inventory was about RMB 4.26 billion, and contract liabilities were about RMB 770 million. By the end of 2025, sell-side research organized from the annual report showed inventory of about RMB 7.17 billion and contract liabilities of about RMB 3.04 billion. In Q1 2026, contract liabilities fell back to about RMB 2.853 billion, while accounts payable increased from about RMB 1.856 billion at the end of 2025 to about RMB 2.077 billion. This data set sends a typical equipment-company signal: backlog and delivery schedules are heavy, and revenue recognition and cash collection are not naturally smooth. This is not a bad thing, but it reminds you not to evaluate AMEC through the lens of a consumer-goods company.

The complete year-by-year balances of accounts receivable, notes receivable, and contract assets could not be stably extracted under a unified basis from the public snippets directly retrieved this time, so I am not willing to force the numbers. But it can be confirmed that the company applies the lifetime expected credit loss model to notes receivable, accounts receivable, and contract assets in its annual report. Given the extremely high concentration of the top five customers, what needs continuous tracking in the future is not an abstract “bad-debt ratio,” but payment rhythm from major customers, changes in contract assets, and whether revenue recognition and cash collection are synchronized. This area requires supplementary direct annual-report statements for a more rigorous special review.

On ROE, ROA, leverage, and survival capacity, my conclusion is: AMEC is not financially fragile, but it is also not a high-cash-return company. Based on a rough estimate at the end of 2025, ROE was around 9%, and ROA around 7%; Reuters/LSEG's current key metrics page also gives ROE around 9.16%. Total debt at the end of 2025 was about RMB 754 million, while Yahoo Finance's latest key statistics showed total cash of about RMB 8.53 billion, implying net cash of about RMB 7.78 billion. Therefore, net debt/EBITDA is significantly negative, and debt-service risk is very low. The exact interest coverage ratio requires additional information in the structured data currently retrieved, but given the company's net-cash position, financial leverage is not the main issue.

On accounting risk, I have not seen direct evidence of financial fraud in the public materials retrieved, nor have I seen abnormal debt pressure or an unexplained large cash gap. But I would put R&D capitalization on the long-term watchlist. The balance of capitalized development expenditure was about RMB 1.248 billion at the end of 2024 and about RMB 1.535 billion at the end of 2025; by Q1 2026, development expenditure had further reached about RMB 1.902 billion. This does not mean the company has a problem, but it means net profit cannot be directly equated with cash profit freely distributable to shareholders. If commercialization of new products falls short, future amortization and impairment pressure will deserve more attention.

Using an “Owner Earnings” approach to estimate true earning power, I would present two measures. Loose measure: operating cash flow minus cash outlays for fixed assets/intangible assets and other long-term assets, about RMB 1.440 billion in 2025. More conservative measure: further subtract cash related to capitalized development expenditure, about RMB 421 million in 2025. Considering that current heavy R&D and platform expansion include both maintenance and growth components, I think a more prudent “normalized owner earnings range” is roughly RMB 800 million to RMB 1.6 billion per year. This is still materially below the profit level implied by the market's pricing of the equity. At the current market capitalization of about RMB 248.7 billion, AMEC trades at about 160x recurring net profit, 173x loose FCF, and 591x strict OE proxy. From a long-term owner's perspective, that is not an easy number.

Intrinsic Value and Margin of Safety

I will give the conclusion first, then the process. At the current price, AMEC looks more like an expensive option waiting for excellent results to materialize than an asset priced clearly below value. For a relatively conservative investor with a holding period of more than 10 years, I would not define it as a buying opportunity with a margin of safety.

Owner Earnings DCF

This is the hardest part, because the valuation disagreement on AMEC depends almost entirely on how much of today's R&D, capitalized development expenditure, and new-base investment you believe is maintenance spending and how much is growth spending. If we strictly use 2025 cash figures, owner earnings were only about RMB 421 million, making the company obviously very expensive. If you believe most of that spending is front-loaded investment for platform expansion over the next decade, then a higher “normalized starting Owner Earnings” can be justified. I therefore use three scenarios and explicitly state the assumptions. Based on 2025 data, current share count, and conservative treatment of net cash, my estimates are as follows:

Dimension Bear Base Bull
Starting Owner Earnings assumption RMB 800 million to RMB 1.0 billion RMB 1.5 billion to RMB 2.1 billion RMB 2.5 billion to RMB 3.0 billion
Growth rate in the first five years 20% 25%–27% 30%
Growth rate in the next five years 10% 12% 15%
Discount rate 11%–12% 9%–9.5% 9%
Terminal growth rate 3% 3.5%–4% 4%
Implied valuation About RMB 35–60/share About RMB 95–170/share About RMB 245–292/share

The most important message in this table is not a single point estimate. It is that the current market price of RMB 265.47 is already close to, and partly above, the valuation range of the “bull scenario.” What does the bull scenario require? It requires the company's current normalized owner earnings to have already broadly reached or exceeded reported net profit, with the first five years of the next decade maintaining high growth of about 30%, the following five years still achieving high growth of about 15%, a discount rate of only 9%, and terminal growth still at 4%. This is not impossible, but it is no longer a conservative investment assumption. It is an assumption of almost perfect execution.

Relative Valuation

Relative valuation happens to validate the same intuition. As of 2026-06-09, Reuters/LSEG showed AMEC at about P/S 18.95x, adjusted P/E 90.71x, P/B 10.20x, and ROE 9.16%. Compared with its strongest domestic competitor, NAURA Technology Group, at about P/S 10.24x, adjusted P/E 75.98x, and P/B 10.80x, and ACM Research (Shanghai) at about P/S 17.04x, adjusted P/E 87.97x, and P/B 8.64x, with Reuters also showing negative annual free cash flow for ACM Research (Shanghai), AMEC is not valued at a clear discount to peers. It is not cheap on sales and earnings multiples, while NAURA also has broader platform width. If AMEC is assigned a more prudent but still not harsh 12–16x P/S range, the corresponding equity value is about RMB 148.6 billion to RMB 198.2 billion, or roughly RMB 159–211 per share.

Asset and Liquidation Value

This method is usually not the primary valuation method for a high-R&D equipment company, but it is useful for showing “how thick the downside cushion is.” At the end of 2025, on Reuters/LSEG's basis, the company had total assets of about RMB 29.846 billion and total liabilities of about RMB 7.151 billion, corresponding to book shareholders' equity of about RMB 22.695 billion. Based on the current share count, book value per share is only around RMB 24/share. Looking at cash, Yahoo Finance's latest key statistics showed total cash of about RMB 8.53 billion, compared with Reuters/LSEG 2025 year-end total debt of about RMB 754 million, implying net cash of about RMB 7.78 billion, or only about RMB 8/share. This reminds us that asset value provides very little protection for today's share price. Buying AMEC is essentially buying the realization of competitiveness and cash flow over the next decade, not buying assets at a discount.

Combining the three methods, I give the following valuation ranges:

Valuation Range My range Explanation
Conservative intrinsic value RMB 25–90/share Close to the lower bound under strict Owner Earnings and the asset method
Fair intrinsic value RMB 130–210/share Balances relative valuation and partial Owner Earnings normalization
Bull intrinsic value RMB 240–300/share Requires long-lasting high growth and near-perfect profit and cash-flow realization

At the current RMB 265.47/share, AMEC is roughly above the fair value range and close to the middle of the bull value range. For conservative investors, I think the required margin of safety should be at least a 20%–30% discount to the lower bound of fair value, so the ideal buy range is roughly RMB 110–150/share. If you have already held it for a long time and confirm that platform realization is proceeding smoothly, RMB 150–220/share can be understood as a “barely acceptable holding price range.” Above RMB 250/share, I would view it as clearly overvalued.

The margin-of-safety question can also be stated more directly: If AMEC only achieves revenue CAGR of 12%, a net margin of 14%, and an exit valuation of 35x P/E over the next decade, annualized returns from buying today are likely negative or close to zero. If it achieves revenue CAGR of 15%, a net margin of 16%, and an exit valuation of 35x P/E, annualized returns are only in the low single digits. Only under strong assumptions of revenue CAGR of 18%–20%, a net margin of 18%, and an exit valuation still at 40–45x P/E could annualized returns enter the 7%–10% range. For balanced, relatively conservative capital, those odds are not attractive. So the answer is clear: the current price does not provide a sufficient margin of safety.

Risks, Bear Case, and Failure Conditions

AMEC's most important risk is not share-price volatility, but permanent capital loss. The first risk is excessive valuation. Today's share price already discounts very strong growth, very high execution quality, and a long runway dividend. If growth or margins over the next 2–3 years are even merely “not that good,” valuation multiple compression could cause significant capital loss. Based on the current 18.95x P/S and 90.71x adjusted P/E, if the market simply re-rates it to a still not cheap 10–12x P/S or 40–50x recurring earnings range, a 40%–60% share-price drawdown would not be exaggerated.

The second risk is competition and technology-cycle risk. AMEC is already strong in etch, but the industry is not static. SEMI clearly states that important drivers of future foundry/logic equipment demand include advanced nodes, GAA architecture migration, and new device structures. This means industry demand may not disappear, but if AMEC falls behind in key technology generations such as high-selectivity etch, next-generation ICP, and advanced thin films, demand may remain in the industry but not necessarily remain with the company. NAURA's broader platform capability and the deep accumulation of international leaders in certain process steps mean AMEC's moat is not unchallengeable.

The third risk is customer concentration and policy dependence. In 2025, the top five customers contributed 75% of revenue, which binds AMEC's growth to the expansion cycles, yield ramps, capex schedules, and domestic-equipment adoption pace of a small number of major customers. At the same time, Reuters cited people familiar with the matter as saying that newly built Chinese fabs were required to use more than 50% domestic equipment. This is a potential positive for AMEC, but conversely, it also shows that part of the growth logic is strongly policy-driven. If policy support weakens at the margin, local subsidy timing changes, customer expansion is delayed, or budgets shrink, the company's orders and deliveries may come under pressure.

The fourth risk is mismatch between cash flow and accounting measures. Reported profit looked good in 2025, but recurring net profit was only RMB 1.550 billion, and the balance of capitalized development expenditure rose from RMB 1.248 billion at the end of 2024 to RMB 1.535 billion at the end of 2025, then further to about RMB 1.902 billion in Q1 2026. If over the next few years you continue to see “nice net profit, but thin strict Owner Earnings,” that would indicate AMEC may be more suitable as a growth-story stock than as a long-term cash-return asset. This is not saying the company has a problem. It is saying the return logic of your purchase must match the company's real cash economics.

The fifth risk is capital allocation and acquisition integration. The company is clearly in reinvestment mode: R&D, new bases, and platform expansion all require capital. At the same time, the transaction around acquiring control of Hangzhou Zhongsi means there will be new variables in integration, valuation, culture, and execution. For a company that is already very expensive, one capital-allocation mistake in a high-priced acquisition would sharply reduce the market's tolerance.

If I were to write the strongest bear case, I would put it this way: “AMEC is not a bad company. It is a good company with a bad price. The market is pricing it with a near-perfect script: domestic substitution advances steadily, etch share continues to rise, thin-film/EPI/metrology and inspection ramp smoothly, potential acquisitions integrate well, high R&D spending is not wasted, capitalized development expenditure all commercializes successfully, and ten years from now the market is still willing to give it a high valuation multiple. If even two or three of these items fail to happen together, shareholder returns may be very ordinary.” This bear case is not extreme. It is quite powerful within my framework.

What facts would overturn the current “keep watching rather than buy” judgment? I would track the following: First, strict owner earnings improve significantly over the next two to three years and stabilize above RMB 1.5–2.0 billion per year; second, new platforms beyond etch generate substantive revenue and cash flow rather than just R&D stories; third, gross margin and recurring net margin stop declining and even recover, without relying on large non-recurring gains; fourth, customer concentration declines, or at least major-customer expansion becomes more sustainable; fifth, the share price falls back to a more reasonable range and materially improves the odds. Conversely, if the following facts appear, I would admit I was wrong and avoid it more firmly: contract liabilities decline materially, inventory stays high while delivery stalls, gross margin falls below 35%, capitalized development expenditure keeps growing rapidly but new-product commercialization fails, acquisition integration goes poorly, or cash flow remains weaker than profit for a long time.

Comparisons, Checklist, and Final Judgment

First, a horizontal comparison. The valuation and key metrics for AMEC, NAURA, and ACM Research (Shanghai) in the table below come from Reuters/LSEG; the CSI 300 valuation comes from third-party compilations based on China Securities Index data; the China 10-year government bond yield comes from the China bond yield curve/CFETS.

Object Current approximate valuation/yield Implication
AMEC P/S 18.95x; adjusted P/E 90.71x; P/B 10.20x; ROE 9.16% High growth expectations are already heavily priced in
NAURA Technology Group P/S 10.24x; adjusted P/E 75.98x; P/B 10.80x Broader platform, yet lower valuation
ACM Research (Shanghai) P/S 17.04x; adjusted P/E 87.97x; P/B 8.64x The industry is generally not cheap, but AMEC has no obvious discount
CSI 300 TTM P/E about 14.23x Buying AMEC means giving up a lower-valuation index alternative
China 10-year government bond About 1.73%–1.74% Risk-free yield is low, but AMEC should still require a higher risk premium

On the question of “is it clearly better than buying the index,” my answer is: not obvious at the current price. The CSI 300's P/E is only a fraction of AMEC's. Even if you acknowledge that the index lacks growth, it wins on cheapness, diversification, and tolerance for error. To prove it deserves to significantly outperform the index, AMEC must continue to deliver above-industry growth and cash realization over the next decade. From today's odds, there is not much room for mistakes. Relative to the 1.73%–1.74% China 10-year government bond yield, AMEC certainly has a higher long-term return ceiling, but that ceiling depends on very strong execution, not on cash flow already in hand. If I could hold only 5 assets, I do not think AMEC has a strong enough claim to enter the portfolio at the current price.

Below is the long-term investment checklist you requested. I use “Pass / Fail / Uncertain” where possible and explain the reason when uncertain.

Checklist Conclusion Explanation
Can I understand this business? Pass The economic logic is understandable, but technical details are complex
Does it have long-term stable demand? Pass Advanced processes and domestic substitution support long-term demand
Does it have a durable moat? Uncertain The etch moat is fairly strong, but the platform moat is still being built
Does it have pricing power? Uncertain It has some bargaining power in critical processes, but customer concentration is a major constraint
Can it generate stable free cash flow? Fail Cash flow improved in 2023-2025, but the strict measure remains unstable
Is its return on capital excellent? Uncertain ROE is around 9%, which is not impressive
Is management trustworthy? Pass Long-term and R&D orientation are relatively clear
Is capital allocation rational? Uncertain Direction is reasonable, but return verification is still pending
Is the balance sheet robust? Pass Net cash, low debt, low leverage
Is valuation below intrinsic value? Fail Current price is above the fair range and close to the bull range
Is the margin of safety sufficient? Fail The current price leaves little room for error
Would I be comfortable holding it long term? Uncertain It depends on whether you can continuously track technology and cash flow
What key facts would make me sell? See below Deterioration in cash flow, gross margin, capitalized R&D, and order quality
Am I interested only because of price action or sentiment? Probably yes The current price is more easily driven by sector narrative

Final rating: Watch. One-sentence investment thesis: AMEC is an excellent and highly promising Chinese high-end semiconductor equipment company, but at the current price around RMB 265, the market has already paid too much in advance for a fairly smooth growth script over the next decade.

The core bull reasons are mainly fourfold. First, the industry itself is good enough: SEMI data show the global semiconductor equipment market reached about USD 135.0 billion in 2025 and remains on a growth track in 2026-2027. Second, the core business is strong enough: etch equipment remains a high-barrier core area, and the installed base is already large. Third, new platforms are starting to scale: thin-film equipment has moved from validation to revenue contribution. Fourth, the balance sheet is robust: clear net cash gives the company ample ammunition for continued R&D and expansion.

The core bear reasons are also substantial. First, the current valuation is too high and already approaches bull-scenario pricing. Second, truly distributable cash flow is far less impressive than reported profit. Third, customer concentration is high, policy variables are strong, and technology routes iterate quickly. Fourth, capitalized development expenditure and platform expansion mean future amortization/impairment and execution risk cannot be ignored.

If the key assumptions fail, the investment logic weakens. At minimum, I think the following must be satisfied: etch equipment continues to scale in advanced logic and memory critical steps; thin-film, EPI, metrology/inspection, and potential wet-process/CMP platforms truly form cash returns; capitalized R&D projects convert at high quality rather than accumulating accounting assets; and during the high-growth phase, the company can gradually raise strict owner earnings closer to recurring profit.

Fair buy price range: RMB 110–150/share. This range roughly corresponds to the requirement of retaining a 20%–30% margin of safety below the lower bound of fair intrinsic value, and better fits your “balanced, relatively conservative” risk preference. Acceptable holding price range: RMB 150–220/share. Clearly overvalued range: above RMB 250/share. Based on the 2026-06-09 price of about RMB 265.47, I tend to place it on the “clearly overvalued” side.

Target holding period: more than 10 years, but only if you are willing to track fundamentals frequently. This is not a company you buy and put in a drawer for ten years without looking. It is better suited to a “long-term holding plus continuous recalculation” framework.

Expected annualized return under relatively prudent ten-year scenarios would be as follows: The bear case is about -3% to 0%; the base case about 1% to 4%; and the bull case about 7% to 10%. In other words, buying at the current price becomes attractive only if the company maintains very strong growth over the next decade and the market remains willing to assign a high valuation. For conservative capital, the odds are not thick enough.

The maximum loss risk is not business failure, but “excellent company, valuation reversion.” If revenue and profit growth disappoint market expectations over the next 3–5 years while valuation compresses from nearly 19x P/S to 10–12x, or from nearly 90x adjusted P/E to 40–50x, a 40%–60% permanent capital loss in the share price would not be surprising. If technology misjudgment, acquisition mistakes, or policy/order headwinds are added, the loss could be larger.

Key indicators to track going forward should include at least these eight items: Revenue growth; etch equipment revenue mix and growth; revenue from new platforms such as thin-film/EPI/metrology and inspection; gross margin and recurring net margin; operating cash flow and strict Owner Earnings; the balance of capitalized development expenditure and its share of revenue; changes in contract liabilities and inventory; top-five customer concentration and breakthroughs with new key customers.

Signals that should trigger reassessment include: Gross margin declining clearly for two to three consecutive reporting periods; contract liabilities weakening while inventory keeps rising; capitalized R&D continuing to grow rapidly while new products fail to convert into revenue; etch growth significantly lagging the industry and core peers; integration or realization problems after the Hangzhou Zhongsi acquisition; and strict Owner Earnings failing to rise for a long time.

Final recommendation: If your goal is to own one of the few Chinese semiconductor equipment companies with genuine global competitive potential, AMEC deserves long-term tracking and even deep research for a future cheaper opportunity. But if your goal is to buy a good business today at a reasonable price that is likely to make real money ten years from now, I would stay disciplined: first admit it is a good company, then admit it may not be a good price right now. For conservative capital, the most rational action is not to chase, but to wait.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.

002371688082

Semiconductor EquipmentEtch EquipmentDomestic SubstitutionAMECBuffett FrameworkValue Investing
Reader Q&A10

Baillie Framework · Ten Questions for Growth Investing

10

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

Baillie Framework · Ten Questions for Growth Investing — score profile: 48/100 total Ceiling 6/10 · Revenue 2x 6/10 · Next engine 5/10 · Moat 6/10 · Reinvention 5/10 · Management 5/10 · Customer need 6/10 · Unit economics 5/10 · 5x path 2/10 · Blind spot 2/10 0510 How large is its market ceiling? Is it expanding an existing pie, or creating an entirely new market? — 6/10 Ceiling 6 Can its revenue at least double over the next five years? Will growth mainly be driven by volume, price, or new businesses? — 6/10 Revenue 2x 6 After five years, what will take over as the next growth engine? Does this "second curve" exist today? — 5/10 Next engine 5 What is its core competitive advantage? Will this moat widen or narrow over the next three to five years? — 6/10 Moat 6 If its core business is disrupted, does it have the DNA to reinvent itself? How does it deal with mistakes and bad news? — 5/10 Reinvention 5 Does management, especially the founder, have a long-term perspective and deep alignment with the company? Is it willing to sacrifice current profits for five to ten years out? — 5/10 Management 5 If it disappeared tomorrow, how much would customers miss it? Is its growth sustainable and not dependent on harming society or regulators? — 6/10 Customer need 6 What are the unit economics of this business, including gross margin and incremental returns? Do they improve or deteriorate as scale increases? Where does the money it earns go? — 5/10 Unit economics 5 What conditions must all hold for it to rise fivefold over ten years? Are those conditions realistic? What expectations are implied by today's share price? — 2/10 5x path 2 Why has the market not realized all this yet? Is it because investors do not understand it, look down on it, or cannot see far enough? What will become the "narrative inflection point"? — 2/10 Blind spot 2
  • How large is its market ceiling? Is it expanding an existing pie, or creating an entirely new market?6/10

    Conclusion: AMEC's market ceiling is high, but it is mainly expanding and redistributing an existing semiconductor equipment pie, rather than creating an entirely new market. SEMI disclosed that global semiconductor manufacturing equipment sales reached USD 135.1 billion in 2025, up 15% year over year, driven by capacity expansion in advanced logic, memory, and AI-related applications; SEMI's subsequent forecast also shows that global semiconductor equipment sales are expected to reach USD 145.0 billion in 2026 and USD 156.0 billion in 2027, with wafer fab equipment expected to reach USD 135.2 billion by 2027. This gives a high-end etch and thin-film equipment supplier like AMEC a long enough runway.

    AMEC can capture this pie not simply because the industry is large, but because it has moved from "domestic substitution in concept" to actual delivery. The company's 2025 annual report disclosed 2025 revenue of RMB 12.385 billion, up 36.62% year over year; etch equipment sales of about RMB 9.832 billion, up 35.12%; and thin-film equipment sales, including LPCVD and ALD, of about RMB 506 million, up 224.23%; its 2026 first-quarter report further showed quarterly revenue of RMB 2.915 billion, up 34.13% year over year. Its ceiling therefore comes from three layers: rising share for etch tools in advanced logic and memory, expansion into adjacent platforms such as thin film, EPI, metrology and inspection, and MOCVD, and the continued adoption of local supply chains by Chinese fabs.

    This is still not a "new market creation" story. AI, HBM, GAA, and advanced packaging will lift equipment intensity, but AMEC still sells process equipment within fab capital expenditure; customers buy etch, deposition, and related tools to manufacture existing or next-generation chips, not an end demand newly defined by AMEC. The annual report also states that sales to the top five customers accounted for 75.00% of annual sales, showing that its growth is highly tied to the expansion, qualification pace, and localization procurement of a small number of fabs, rather than to an independent demand curve it creates by itself like consumer internet or new drugs.

    From a Baillie Gifford perspective, Q1 supports a moderately positive conclusion: the market is large enough, and AMEC still has substantial room to gain share relative to its RMB 12.385 billion revenue base; but the nature of this room is "expansion of an existing high-end equipment market + domestic substitution + platform-based share gains." Using the unified valuation anchor of about RMB 267.0 billion in market capitalization for this round, a fivefold gain in ten years would require a market value of about RMB 1.34 trillion. That cannot rely only on a large industry ceiling. Etch share must keep rising, second curves such as thin film must form independent profit pools, cash-flow quality must keep pace with revenue growth, and valuation multiples cannot compress materially.

    Jun 9, 2026
  • Can its revenue at least double over the next five years? Will growth mainly be driven by volume, price, or new businesses?6/10

    Conclusion: AMEC has a realistic chance of at least doubling revenue over the next five years, but this should not be treated as a risk-free baseline. Starting from 2025 revenue of about RMB 12.385 billion, up 36.62% year over year, doubling to about RMB 24.77 billion would require only about a 14.9% compound annual growth rate; 2026Q1 revenue of RMB 2.915 billion, up 34.13% year over year, and recurring net profit attributable to the parent up 60.09% year over year also show that near-term momentum has not broken. Still, equipment companies are driven by large projects, long qualification cycles, and customer capital expenditure, rather than linear subscription revenue. A higher base, top-five customer concentration of about 75%, and the timing of contract liabilities and inventory will all make the path to doubling uneven.

    The main driver should be volume, followed by new businesses, while price is not the core. Etch equipment remains the base business: 2025 etch equipment sales were about RMB 9.832 billion, up 35.12% year over year; the Q1 report also disclosed a significant increase in new shipments of high-end key etch products for advanced logic and memory, with more than 300 reactors for ultra-high aspect ratio etchers already in stable, large-scale mass production. This looks more like shipment and share gains from domestic substitution, advanced process and memory capacity expansion, and customer adoption, rather than sharp price increases per tool. Conversely, customer concentration and the absence of a sustained rise in gross margin in recent years both show that it is not strong enough to double primarily through "price."

    New businesses will provide a second layer of acceleration, but they are not the sole pillar yet. The Q1 report said thin-film equipment revenue grew sharply by about 224.23% year over year in 2025, while new shipments of LPCVD, ALD, EPI, and other products maintained rapid growth. This indicates that the platformization direction has begun to scale; however, relative to total revenue of RMB 12.385 billion, thin-film-related revenue remains small and still needs to turn from qualification and new shipments into stable batch revenue over the next several years. The external environment provides room: SEMI said global semiconductor equipment sales were USD 135.1 billion in 2025, up 15% year over year, with China's equipment spending of USD 49.3 billion still near a record, and global equipment billings were USD 36.55 billion in 2026Q1, up 14% year over year. My judgment is therefore: if Chinese fab capital expenditure continues, etch share keeps rising, and thin film, EPI, and MOCVD deliver, AMEC can double revenue in five years; if customer expansion slows or new-product qualification stretches out, it may still grow, but doubling would be pushed back.

    Jun 9, 2026
  • After five years, what will take over as the next growth engine? Does this "second curve" exist today?5/10

    Conclusion: AMEC's second curve already exists today, but it exists as "product and customer adoption," not yet as an "independent profit pool." Five years from now, the most likely successor is not MOCVD, but the thin-film deposition platform, especially LPCVD/ALD, followed later by EPI, metrology and inspection, and wet-process/CMP capabilities that may be added through acquisitions and internal R&D. The reason is straightforward: etch remains the main revenue axis, with 2025 etch equipment sales of about RMB 9.832 billion, up 35.12% year over year; in the same year, thin-film equipment sales including LPCVD/ALD were about RMB 506 million, up 224.23%. The growth rate is high, but the scale is only about one-twentieth of etch, indicating that the annual report already contains revenue evidence for a second curve, though it is still far from taking over.

    The external demand behind this curve is not imaginary. SEMI expects global semiconductor manufacturing equipment sales to rise from USD 145.0 billion in 2026 to USD 156.0 billion in 2027, with growth mainly coming from AI-related advanced logic, memory, HBM, and advanced packaging; rising process complexity in these areas will also increase the value of deposition, epitaxy, etch, and metrology and inspection steps. If AMEC relies only on domestic substitution in its core etch business, growth will be constrained by customer capex cycles and the ceiling of a single category; if LPCVD/ALD, EPI, and metrology and inspection can achieve batch qualification among advanced logic and memory customers, AMEC has a chance to move up one level from "etch leader" to "front-end platform equipment company."

    But it cannot yet be described as having completed the handoff. The company recorded 2026 first-quarter revenue of RMB 2.915 billion, up 34.13% year over year, and R&D expense of RMB 908 million, equal to 31.14% of revenue, showing that the company is still using high R&D intensity to exchange for future productization; this is a necessary condition for the second curve, and it also means short-term profit and free cash flow will continue to be absorbed by R&D, qualification, inventory, and delivery cycles. The real inflection point is not the mere existence of "new products," but evidence over the next three to five years that non-etch revenue keeps stepping up, gross margin is not dragged down by new products, and contract liabilities and cash flow improve at the same time.

    My judgment is therefore: the second curve exists, but the evidence is early to mid-stage. The most credible successor is thin-film deposition; EPI, metrology and inspection, and wet process/CMP are an option pool; MOCVD is more of a historical business and a compound-semiconductor supplement. They may become a second growth engine after five years, but only if they grow from "new-product revenue at the RMB 500 million level" into "platform businesses at the several-billion-RMB level that can contribute cash profit." Before that, AMEC is still primarily a strong etch company, rather than a proven full-stack front-end equipment platform.

    Jun 9, 2026
  • What is its core competitive advantage? Will this moat widen or narrow over the next three to five years?6/10

    Conclusion: AMEC's core competitive advantage is an "engineering moat" in etch equipment, rather than a consumer brand, network effect, or monopoly over a single patent. Its barriers come from the combination of process know-how, customer qualification cycles, switching costs after line adoption, field service capability, installed base, and sustained R&D investment: once equipment enters a customer's key etch process, replacement is not just buying another machine; it requires rerunning the process window, yield, stability, and production-line rhythm. Hard evidence supports this view as well: 2025 etch equipment sales were about RMB 9.832 billion, up about 35.12% year over year, while R&D investment was about RMB 3.744 billion, about 30.23% of revenue; by the first quarter of 2026, the company further disclosed more than 8,300 reactors cumulatively in mass production across more than 180 customer chip and LED production lines in China and overseas, with R&D expense of RMB 908 million, equal to 31.14% of revenue. These data show that AMEC's advantage is no longer just a "domestic substitution story"; it is backed by real customer adoption, running tools, and R&D iteration.

    Over the next three to five years, I lean toward its moat in the core etch business continuing to widen. The reason is that industry demand is still tilting toward advanced logic, advanced memory, AI, and high-bandwidth-memory-related capacity. SEMI disclosed that global semiconductor equipment sales were about USD 135.1 billion in 2025, up 15% year over year, while China's equipment spending remained near historical highs, and global equipment billings rose another 14% year over year in the first quarter of 2026. The more demand moves toward complex processes such as high-aspect-ratio etch, GAA, and 3D DRAM, the more customers value the stability and engineering support of qualified equipment; the larger AMEC's installed base becomes, the easier it is to form a cycle of "more mass-production feedback - more process experience - stronger customer trust - more follow-on orders."

    This moat is not unlimited, and it should not be described as a full-platform monopoly on the level of the global leaders. First, Lam Research, Applied Materials, Tokyo Electron, and others still have deep global accumulated expertise, and domestic peer NAURA has greater platform breadth; second, AMEC's platform businesses outside etch, such as thin film, EPI, and metrology and inspection, are still in early delivery, with 2025 LPCVD/ALD sales of about RMB 506 million, still not in the same league as the core etch business; third, the top five customers accounted for about 75% of sales in 2025, showing that while customer stickiness is strong, bargaining power and order timing are also affected by a small number of large customers. My judgment is therefore: over the next three to five years, the core etch moat is likely to widen; whether the company's overall moat can upgrade from a "strong single point" to a "broad platform" still depends on whether thin film, EPI, and metrology and inspection can form sustainable revenue, gross margin, and cash flow, rather than remaining at the stage of R&D and qualification progress.

    Jun 9, 2026
  • If its core business is disrupted, does it have the DNA to reinvent itself? How does it deal with mistakes and bad news?5/10

    Conclusion first: AMEC has the DNA for self-reinvention, but it has not yet passed the ultimate stress test of its "core business being disrupted." It is not a company merely living off its legacy etch equipment franchise: from MOCVD/LED to etch, and then to platform directions such as LPCVD/ALD, EPI, PVD, PECVD, and metrology and inspection, the organization has been migrating its underlying plasma, vacuum, thin-film, and customer qualification capabilities into new products. The latest official quarterly report disclosed that in 2026Q1 the company had R&D expense of RMB 908 million, equal to 31.14% of revenue, and was advancing R&D on six major categories and more than 20 new equipment products, while thin-film equipment revenue in 2025 also grew sharply by about 224.23% year over year; this shows that it is willing to use high R&D intensity to reconstruct its product portfolio in advance, rather than waiting until the etch business loses speed before taking remedial action.

    The real stress test is whether AMEC can proactively weaken an old profit pool and redirect resources to next-generation equipment if the process route, customer architecture, or competitive landscape of etch itself changes sharply. The industry environment will indeed force continuous reinvention. SEMI expects 300mm fab equipment spending to continue double-digit growth in 2026 and 2027, driven by AI, sub-2nm advanced nodes, HBM, and memory technology transitions, meaning equipment companies cannot rely only on the previous generation of etch platforms. AMEC's positive evidence is high R&D, the rapid launch of new thin-film/EPI products, and entry into customer qualification; its constraints are that in 2025 etch equipment still contributed about RMB 9.832 billion, LPCVD/ALD about RMB 506 million, and top-five customers about 75% of revenue, so platformization has not fully become an independent cash-flow moat.

    On its attitude toward bad news, I would give the judgment "relatively honest disclosure, but the cultural evidence is still not hard enough." In its Q1 report, the company did not discuss only the +197.20% growth in net profit attributable to the parent; it also separated recurring net profit growth of +60.09%, negative operating cash flow, and after-tax net income of about RMB 397 million from the sale of Piotech shares during the period. This disclosure at least allows investors to see profit quality. The higher-level Baillie Gifford requirement is whether management can publicly acknowledge mistaken projects and cut low-return R&D and acquisitions in time, rather than hiding capitalized development expenditure, inventory, and failed customer qualifications inside a "long-term investment" narrative. Current public materials are still insufficient to prove that AMEC has formed such a strong correction culture. The conclusion is therefore: it has technology migration and early self-reinvention capability, but still needs to prove itself through cash flow from non-etch businesses, realization of capitalized R&D, and discipline in exiting failed projects.

    Jun 9, 2026
  • Does management, especially the founder, have a long-term perspective and deep alignment with the company? Is it willing to sacrifice current profits for five to ten years out?5/10

    Conclusion: management has a long-term perspective and is willing to sacrifice current profits for a technology platform five to ten years out; however, interest alignment is only moderate and should not receive full marks as a high-ownership founder-owner case. Gerald Yin remains chairman, general manager, and core technical personnel. The 2025 annual report's director and senior management table shows that his term runs through 2028 and that he still directly held about 4.16 million shares at the end of 2025; the company also disclosed that the largest shareholder, Shanghai Venture Capital, held 14.93%, while the second-largest shareholder, Xinxin Investment, held 10.94%, with no concerted action among major shareholders, and the company has no controlling shareholder and no actual controller. Its governance therefore looks more like founder-led professionalism under dispersed ownership, rather than an owner-operator structure in which the founder controls the company through absolute equity ownership.

    The evidence for long-term tradeoffs is fairly solid: the company had 2025 R&D investment of RMB 3.744 billion, equal to 30.23% of revenue and up 52.65% year over year, and the annual report said ongoing R&D covered six equipment categories and more than 20 new equipment products; in 2026Q1, it again maintained R&D expense of RMB 908 million, equal to 31.14% of revenue. This shows that management is not cutting R&D to protect short-term margins, but using current profit and cash flow to pursue platform opportunities beyond etch, including thin film, EPI, and metrology and inspection. The report also points out that 2025 recurring net profit grew only 11.64%, materially below revenue growth of 36.62%, partly because of high R&D expense, which fits the feature of "sacrificing current profits for five to ten years out."

    But "deep alignment" needs a discount: official shareholding reduction announcements show that before his January 2026 reduction, Gerald Yin directly held 4,159,436 shares, or 0.664%; by February, he had cumulatively reduced 208,430 shares due to tax needs related to restoration of Chinese nationality, bringing current holdings down to 3,951,006 shares, or 0.631%; a May announcement further showed that he directly held 4,022,536 shares, or 0.640%, and planned to reduce holdings by no more than 100,000 shares for equity-incentive exercise funds and tax payment needs. The reductions are not large and have tax/exercise explanations, so they do not mean long-term motivation has deteriorated; but from a Baillie Gifford perspective, this is indeed below the strongest category in which a founder has most of their net worth tied to the company and would rather not sell for the long term. Succession, equity incentives, and whether capital allocation continues to center on long-term per-share value still need observation.

    Jun 9, 2026
  • If it disappeared tomorrow, how much would customers miss it? Is its growth sustainable and not dependent on harming society or regulators?6/10

    Conclusion: if AMEC disappeared tomorrow, core customers would clearly miss it, especially fabs that have already adopted its etch, LPCVD/ALD, and other equipment in advanced logic and memory production lines; but that pain would be more like the sudden loss of a qualified key supplier, rather than an absolute monopoly where "no second company in the world can do this." The annual report disclosed 2025 revenue of RMB 12.385 billion, etch equipment sales of about RMB 9.832 billion, R&D investment of RMB 3.744 billion equal to 30.23% of revenue, and more than 7,800 reactors cumulatively in mass production on customer production lines by year-end, with more than 6,800 cumulative shipments of etch equipment; these figures show that it is already embedded in customer processes and service systems, rather than remaining at the prototype or policy-concept stage. For fabs, replacing a key etch tool is not just repurchasing equipment; it also involves process windows, yield, certification cycles, and field engineering support, so a short-term disappearance would create real friction.

    This indispensability is strengthening, but it still has a ceiling. The 2026 first-quarter report disclosed that the company's key middle-of-line etch processes for advanced logic and ultra-high aspect ratio etch processes for advanced memory devices had achieved mass production, while LPCVD, ALD, and other thin-film equipment had also entered the market smoothly and maintained rapid new shipments, which will increase customer stickiness; at the same time, both the research report and the annual report point to about 75% revenue concentration among the top five customers, showing that AMEC grows with a small number of major customers and shares concentration risk with them. Fabs naturally maintain multiple-supplier backups. Globally, there are strong competitors such as Lam Research, Applied Materials, and Tokyo Electron, and domestically there are platform equipment makers such as NAURA; customers would miss AMEC, but if it disappeared, they would more likely incur requalification and delivery-delay costs than stop expanding capacity.

    The growth model is generally sustainable, but not unconditionally so. AMEC's growth comes from semiconductor manufacturing equipment demand, domestic supply-chain resilience, and upgrades in advanced logic and memory processes. SEMI disclosed that global semiconductor manufacturing equipment sales rose to USD 135.1 billion in 2025, up 15% year over year, mainly driven by capacity expansion in advanced logic, memory, and AI-related areas; global equipment billings rose another 14% year over year to USD 36.55 billion in the first quarter of 2026. This growth does not rely on addiction, predatory finance, or regulatory arbitrage, and its social function is broadly positive; the real constraints are equipment cycles, customer concentration, export controls, the pace of industrial policy, and whether high R&D investment can keep converting into cash returns. The answer to Q7 is therefore: AMEC has an engineering moat that "customers would miss," and its growth path is not premised on harming society; however, regulatory and geopolitical variables are not noise, but core variables that must be built into a long-term valuation discount for this business.

    Jun 9, 2026
  • What are the unit economics of this business, including gross margin and incremental returns? Do they improve or deteriorate as scale increases? Where does the money it earns go?5/10

    Conclusion: AMEC's unit economics are "engineering-driven high gross margin + high reinvestment," rather than an asset-light, high cash-return model. The annual report shows 2025 revenue of RMB 12.385 billion and a gross margin of 39.17% for semiconductor equipment revenue, with gross margin down 1.89 percentage points year over year; revenue increased by RMB 3.319 billion and gross profit increased by RMB 1.128 billion, but the same annual report also disclosed that R&D expense increased by RMB 1.058 billion, while recurring net profit rose by only RMB 162 million to RMB 1.550 billion. In other words, the first layer of gross profit on incremental revenue is decent, but the second layer of true incremental return reaching shareholders is temporarily thin. In the first quarter of 2026, the company continued to report revenue of RMB 2.915 billion, up 34.13% year over year; recurring net profit of about RMB 478 million, up 60.09% year over year; and R&D expense of RMB 908 million, equal to 31.14% of revenue, showing that core-business profit leverage is improving but still being absorbed by sustained high R&D.

    As scale expands, the technology and customer sides are improving, but the cash economics have not yet moved into the top tier at the same pace. SEMI disclosed global semiconductor manufacturing equipment sales of USD 135.1 billion in 2025, up 15% year over year, and this kind of industry expansion can help AMEC spread service, qualification, and manufacturing systems over a larger base; however, AMEC's own gross margin declined in 2025, and operating cash flow was RMB -159 million in the first quarter of 2026, with inventory of RMB 7.259 billion and contract liabilities of RMB 2.853 billion, showing that equipment delivery, acceptance, and stocking remain heavy. In other words, greater scale brings a larger installed base and learning curve, but customer concentration, discounts, R&D, and working capital consume part of the scale benefit.

    The money it earns mainly goes to three places. First is R&D and development expenditure: 2025 R&D investment was RMB 3.744 billion, the capitalized development expenditure balance was RMB 1.535 billion, and cash expenditure directly related to capitalized development expenditure during the year was RMB 1.020 billion, while the development expenditure balance further rose to RMB 1.902 billion in the first quarter of 2026; second is capacity and delivery, with RMB 855 million paid in cash in 2025 for the purchase and construction of fixed assets, intangible assets, and other long-term assets, and inventory remaining above RMB 7.0 billion; third is platform expansion, including thin film, EPI, metrology and inspection, wet process/CMP product lines, and acquisition layouts. Under the strict report methodology, 2025 operating cash flow was RMB 2.295 billion; after deducting RMB 855 million for long-term asset purchases and RMB 1.020 billion of capitalized development cash, strict owner earnings were only about RMB 420 million. Meanwhile, StockAnalysis's valuation page still shows about 90 times trailing PE and about 140 times P/OCF, meaning the market is buying future realization of scale effects, not cash returns already mature today.

    Baillie Gifford-style judgment: this is a business with a real engineering barrier and a meaningful gross-profit pool, but the evidence that "bigger is better" remains at the operating level and has not fully translated into per-share cash returns. If non-etch revenue scales up in the future, gross margin stabilizes around 40%, R&D intensity gradually declines from 30%+, and strict owner earnings catch up with recurring profit, scale will improve the business; if revenue keeps growing rapidly while capitalized R&D, inventory, and acceptance cycles expand in parallel, scale will only make the company larger, not necessarily make unit economics materially better.

    Jun 9, 2026
  • What conditions must all hold for it to rise fivefold over ten years? Are those conditions realistic? What expectations are implied by today's share price?2/10

    Conclusion: for AMEC to rise fivefold from today, "domestic substitution + a strong industry cycle" is not enough. It would need to grow from about RMB 267.0 billion in market value to about RMB 1.34 trillion, while its valuation multiple ten years from now does not collapse materially; this is closer to a low-probability blue-sky scenario than a baseline case.Using the unified anchor for this round, the share price is RMB 285.04; StockAnalysis showed the same-day price of RMB 285.04, trailing PE of 91.34x, and forward PE of 64.19x. If, ten years from now, the market gives a mature high-quality equipment company only 40-50x PE, AMEC would need annual net profit of about RMB 27.0-33.5 billion; even at 70x, it would still need about RMB 19.0 billion in net profit. Compared with 2025 revenue of RMB 12.385 billion, net profit attributable to the parent of RMB 2.111 billion, recurring net profit attributable to the parent of RMB 1.550 billion, and R&D investment equal to 30.23% of revenue, this is not a simple doubling of revenue over five years, but a roughly tenfold profit increase over ten years, alongside a clear improvement in cash-flow quality.

    At least the following conditions must all hold. First, the global equipment market must continue expanding, and AI/HBM/GAA, advanced logic, and memory investment must be more than one capex peak; SEMI forecasts total equipment sales rising from $133B in 2025 to $156B in 2027, with WFE reaching $135.2B by 2027, while China's capacity expansion and localization share must remain strong. Second, AMEC's core etch business must keep gaining share in key processes and convert its installed reactor base into service, repeat purchases, and process learning curves. Third, second curves such as thin-film LPCVD/ALD, EPI, metrology and inspection, and wet process/CMP must move from "high-growth small bases" into multi-ten-billion-RMB revenue/profit pools. Fourth, R&D investment must remain high for the long term without consuming shareholder cash flow, and top-five customer concentration, equipment cycles, geopolitical restrictions, and potential dilution from equity incentives or acquisitions must not deteriorate materially. Near-term facts support "growth is still present": the company reported 2026Q1 revenue of RMB 2.915 billion, up +34.13% year over year; recurring profit of RMB 478 million, up +60.09%; and R&D investment equal to 31.14% of revenue, while SEMI also showed 2026Q1 global equipment billings of $36.55B, up +14% year over year. These only show that tailwinds exist, not that ten-year compounding is locked in.

    I would split realism into two layers: the industrial direction is realistic, but the conditions for shareholders to earn a fivefold return are demanding. AMEC has real strengths in etch leadership, domestic substitution, high R&D, and customer qualification cycles, but the global equipment market itself will not naturally expand fivefold over ten years. The company therefore needs simultaneous share gains, product-line expansion, margin improvement, and sustained high valuation to achieve a fivefold outcome; if any one of these fails, ten-year returns will be meaningfully weakened. Today's valuation is already around 90x trailing PE and 60x+ forward PE, leaving very little room for error: if in ten years it is merely an excellent but still cyclical Chinese equipment platform, and its valuation returns from a high-growth narrative to 35-50x, operating growth could first be offset by multiple compression.

    Today's share price therefore implies expectations that are more demanding than the mild idea that "AMEC continues to be excellent." It implies "etch share keeps rising, thin film/EPI/metrology and inspection form a second profit pool, capitalized R&D and new base investments convert at high quality, domestic equipment adoption is not interrupted, free cash flow gradually approaches recurring profit, and the market remains willing to pay a high multiple for a high-growth equipment leader over the long term." This set of expectations is not absurd, but it is already full. The real signal that could make a ten-year fivefold gain realistic is not more discussion of domestic substitution, but several consecutive years of high revenue growth, simultaneous improvement in recurring profit margin and operating cash flow, scaling of second-curve revenue, and no dilution of per-share value from financing and equity incentives.

    Jun 9, 2026
  • Why has the market not realized all this yet? Is it because investors do not understand it, look down on it, or cannot see far enough? What will become the "narrative inflection point"?2/10

    Conclusion: the market does not fail to understand AMEC, nor does it look down on it. AMEC's identity as a "domestic semiconductor equipment leader + core etch asset" has already been highly recognized; what has not been fully priced is whether it can move from a strong etch company into a multi-platform equipment group and convert high R&D investment into distributable cash flow. Using the unified price anchor for this round of about RMB 285/share and a market value of about RMB 267.0 billion, StockAnalysis still showed the valuation on the same day at more than 90 times trailing PE and about 19 times P/S, so this is not a case of "the market has not discovered a bargain." The market has already paid a high price for a very smooth future.

    The disagreement remains because both sides have hard evidence. On the optimistic side, the industry market is indeed expanding: SEMI said global semiconductor equipment sales reached USD 135.1 billion in 2025, up 15% year over year, with AI, advanced logic, memory, and HBM as the main drivers; AMEC itself is also delivering, with the official 2026 Q1 disclosure showing revenue of RMB 2.915 billion, up +34.13% year over year; recurring profit of about RMB 478 million, up +60.09% year over year; and R&D investment of about RMB 908 million, equal to 31.14% of revenue, while emphasizing that 2025 thin-film equipment revenue grew sharply by about 224.23% year over year. The cautious side is also reasonable: thin film, EPI, metrology and inspection, and CMP have not yet proven themselves as independent profit pools, while customer concentration, equipment cycles, capitalized R&D, and cash-flow quality will still determine whether a "good story" ultimately turns into per-share value.

    The narrative inflection point will not be another slogan of "domestic substitution," but verifiable operating evidence. First, thin film/EPI/metrology and inspection/CMP outside etch must move from qualification and small-scale revenue into billion-RMB-level and then several-billion-RMB-level revenue contributions, without dragging down gross margin or collections. Second, advanced logic and memory customers must keep repurchasing, and contract liabilities, inventory turnover, and revenue recognition must form a healthy loop. Third, free cash flow or owner earnings under a strict definition must gradually approach recurring profit, rather than being consumed for a long time by capitalized R&D and capacity expansion. At that point, the market narrative may shift from "expensive domestic equipment leader" to "Chinese equipment compounder forming a platform moat"; conversely, if orders slow, inventory stays elevated, gross margin falls, or cash flow remains weaker than profit, the narrative inflection point may also occur downward.

    Jun 9, 2026
Ask about this report

Members can ask about this report; once answered it appears under "Reader Q&A" on this page. You can also highlight a passage in the text to ask about it directly.