United Microelectronics Corporation (UMC): what the price assumes

In the published model solve dated 2026-Q2, anchored at $19.11, United Microelectronics Corporation (UMC) is priced for today's economics sustained for ~6.2 years. boothcheck publishes no house fair value, target price, or buy/sell rating; individual model outputs and user-controlled scenarios are analytical inputs, not Boothcheck targets. Narrative composed 2026-07-25.

Generated: 2026-08-25 · Source: https://boothcheck.com/report/UMC

Headline

What The Price Assumes (Inversion)

The assumption today's price embeds, recovered by inverting the valuation.

The operating-margin figure is value-band context at year 12: derived from the framework's value band, a separate calculation — not part of the priced-in solve.

Solve inputs: computed at a 9.6% cost of capital; growth searched up to the 25% self-funding ceiling.

How unusual the bet is: within-range (limited comparison data)

Valuation X-Ray

Every valuation family lands below the price. The price therefore sits beyond what those standard frames encode.

How the valuation models price the stock relative to the market price. Price/FV above 1.0 means the market pays more than that lens defends (expensive); at or below 1.0 the lens can defend the price.

Families that call it expensive: Asset, Earnings

The models below discount at their own flat-beta convention rates (cost of equity 9.3%, WACC 8.7%); the inversion above states its own rate.

Per-Model Detail (n=16)

Solvency

Bullet Takeaways

• There is no second business here to absorb a bad year: "The Company only has wafer fabrication operating segment as the single reporting segment", and wafer work is 96% of revenue.
• Profitability went backwards in the last filed year even as sales grew, with the annual report recording that "Our operating margin decreased from 22.2% in 2024 to 18.5% in 2025", against a price that embeds operating profit compounding at the fastest rate the company could self-fund for roughly six years.
• The near-term direction has turned, with June 2026 sales of NT$23.12 billion up 22.85% on a year earlier, and second-quarter results are due July 29, 2026.

Bull Case

A contract chip manufacturer has to be read differently from an ordinary industrial. The fabs are already built and the depreciation runs whether the tools are busy or idle, so almost all of the profit swing comes from how full the plant is rather than from anything management decides in a given quarter. One trailing year therefore tells you where the company sits in a cycle, not what it can earn. The company describes the underlying treadmill plainly: "we have continuously been migrating to increasingly sophisticated technologies to maintain the same level of profitability". That single sentence is the whole industry.

Read that way, the recent turn matters more than the last completed year. Taiwan requires monthly sales disclosure, and June 2026 came in at NT$23.12 billion, up 22.85% against the same month a year earlier. Set that against the 20-F for 2025, which reports operating revenues rising just 2.3% to US$7,573 million for the full year. A plant filling up at a low-twenties percentage rate of growth is a very different machine from one growing at two, because the incremental wafer arrives at close to full contribution once the fixed costs are already covered.

The incremental demand is landing in the right place. In July 2026 the company reached mass production of silicon photonics wafers at its Singapore facility with a named partner already in production. Silicon photonics is specialty work rather than a race for the newest transistor, and specialty is where a foundry with paid-for equipment competes best. The annual report is explicit that scale in existing capacity is the strategic priority, describing how "maintaining our foundry capacity with advanced technology and facilities is critical to maintaining our industry leadership" with a focus on twelve-inch facilities.

The wider foundry market has been repricing hard, which makes the revenue line more impressive than it first looks. GFS, competing in the same specialty and mature-node space, disclosed in its own annual filing "a 10.4% decrease in average selling prices". In a market where a direct competitor's prices fell by a tenth, growing revenue at all requires either volume or mix, and both are earned rather than given.

Underneath it sits a balance sheet built for this industry. Cash and equivalents run to roughly 3.31 billion dollars against gross borrowings near 615 million, leaving a net position of about 2.70 billion dollars. In a business where fixed costs do not flex, that is what allows a company to keep equipping fabs through a trough while thinner competitors defer and fall a technology generation behind.

Bear Case

The fragility in this company is not on the liability side of the balance sheet, which is close to spotless. It is on the asset side, and the annual report says so directly: "Our business and the nature of our industry require us to make substantial capital expenditures leading to a high level of fixed costs", and those outlays are committed early, since "These capital expenditures will be made in advance of any additional sales to be generated by new or upgraded fabs". The filing goes further and acknowledges that the company has incurred, and may again incur, operating losses if revenues fail to cover that spending. Net cash of about 2.70 billion dollars looks reassuring in isolation. Measured against what a modern twelve-inch facility costs, and against the fact that the money leaves before the orders arrive, it buys time rather than safety.

The last filed year shows what that structure does when demand is merely adequate. "Operating income decreased from NT$51,613 million in 2024 to NT$43,949 million (US$1,401 million) in 2025", and the annual report records the operating margin falling by close to four points across the same comparison. Revenue rose. Profit fell. Research and development took 7.5% of operating revenues against 6.7% the prior year, and the annual report attributes most of that increase to depreciation on equipment. The equipment bill grows whether or not the technology migration produces a better price, which is precisely what the company means when it says it migrates to more sophisticated processes in order to hold the same profitability.

Concentration compounds the problem. "Our largest customer from our wafer fabrication segment accounted for 13.1%, 10.4% and 11.8% of our operating revenues in 2023, 2024 and 2025, respectively", and there is nowhere else in the business to hide, since "The Company only has wafer fabrication operating segment as the single reporting segment". One product, one cycle, and a customer base narrow enough that a single design win moving to a rival shows up in the annual figures.

Which brings the argument to what the price asks for. It embeds operating profit growing at the fastest pace this business could fund out of its own earnings, sustained for roughly six years, and only about 27% of comparable fast growers held a pace at that level for a stretch that long. The last filed year went in the opposite direction. Only the forward-projecting methods reach today's price at all: the asset-value approaches sit roughly two and a half times below it and the earnings-power approaches roughly two and a half times below as well. The company does hold about 1.05 billion dollars of equity investments outside the operating business, a little over 2% of market value, which puts a modest floor under the downside without changing the arithmetic above it. And the horizon is rate-sensitive in a way that cuts fast: each additional percentage point of cost of capital removes roughly 1.8 years from the runway the price can support.

Valuation

What $19.49 pays for, stated as plainly as the arithmetic allows, is a bet on duration. The price embeds operating profit growing at the fastest rate the business could fund from its own earnings, held there for about six years, computed at roughly a 9.5% cost of capital. That is a single pass under fixed assumptions rather than a measurement, so the number to hold onto is the shape: the price is not asking for a new level of profitability, it is asking for an ordinary boom to last an unusually long time.

Two things sit awkwardly against that. The first is the base rate: of comparable fast growers, only about 27% sustained a pace at this level across a stretch of that length. The second is the last completed year, where the direction was wrong. The 20-F reports operating revenues up 2.3% to US$7,573 million for 2025 while "Operating income decreased from NT$51,613 million in 2024 to NT$43,949 million (US$1,401 million) in 2025". A price built on six years of compounding is being asked to reconcile with a filed year in which profit shrank.

The methods split accordingly, and the split is stark rather than ambiguous. Only the forward-growth approaches reach today's price. Peer multiples land about 33% below it, and the asset-value and earnings-power approaches land roughly two and a half times below. The mechanism at the bottom is worth naming: the earnings-power lens averages five years of operating profit and capitalizes the result assuming no growth at all, which for a business this cyclical is a through-cycle floor rather than a considered valuation. What the pattern says is that the price is a durability premium, and durability is the one property a foundry's economics have never reliably offered.

The peer group frames the profitability question usefully. AMKR, doing the packaging and test work that sits downstream of the wafer, earned a 7.6% operating margin on $7.07 billion of revenue while growing revenue 12.7%. At the other end, TXN earned a 35.3% operating margin on $18.44 billion and ADI 32.5% on $12.74 billion, both of which own their designs rather than manufacture other people's. On the basis the annual report uses for 2025, this company's 18.5% sits between those two worlds, which is exactly where a contract manufacturer of commodity capacity should sit. The valuation attached to that middle position is the part that is unusual, not the margin.

Financially there is nothing pressing. Cash and equivalents of roughly 3.31 billion dollars against gross borrowings near 615 million leaves the company able to fund its own capacity plans without asking anyone's permission. That solves the question of whether the next fab gets built. It says nothing about whether the next fab earns the return the price already assumes, and that is the only question in this name.

Catalysts

Second-quarter results are due July 29, 2026, and unusually for a company of this size most of the revenue is already public. Taiwan requires monthly sales disclosure, and June 2026 came in at NT$23.12 billion, a 22.85% increase on the same month a year earlier. What the quarterly report actually adds, therefore, is the margin: whether the higher volume is arriving at prices that improve on the profitability the annual report records for fiscal 2025, which fell against the prior year, or whether the extra wafers are being bought with discounts.

The clearest strategic development landed in mid-July, when the company began mass production of silicon photonics wafers at its Singapore facility, with a partner named as the first customer in volume. Silicon photonics sits in the optical interconnect path that data centre buildouts consume, and it is specialty rather than leading-edge work, which suits a foundry whose competitive position rests on mature and differentiated processes rather than on transistor density. Singapore also diversifies where the capacity physically sits, which for a Taiwan-headquartered manufacturer is a consideration customers increasingly price.

Broker commentary around the mid-July production news pointed to an improving second half. That view and the annual figures are not in conflict; they are measuring different points on the same cycle. The July print is the first opportunity to see whether the monthly sales momentum reaches the operating line, and given how much of a foundry's profit comes from the last few points of plant utilization, it is the margin rather than the revenue that will decide it.

Peer Cohorts (Per Segment, With Filing Citations)

Core business (reported)

• GFS (GLOBALFOUNDRIES Inc.)
• FY2025 20-F: …and our ability to scale our operations in a way that adequately responds to our long-term strategy. Our ability to compete successfully also depends on factors partially outside of our control, including component supply, intellectual property, including cell libraries that our customers embed in their product…
• FY2025 20-F: …earnings and margins. If we are unable to compete effectively with other sophisticated players in the highly competitive foundry segment of the semiconductor industry, we may lose customers and our profit margins and earnings may decrease. • Our competitors and Integrated Device Manufacturers ("IDMs") have commenced…
• TSEM (TOWER SEMICONDUCTOR LTD.)
• FY2025 20-F: …foundry service providers that also provide specialty technologies, such as Taiwan Semiconductor Manufacturing Corporation (TSMC), United Microelectronics Corporation (UMC), and Semiconductor Manufacturing International Corp. (SMIC). Although these three pure-play semiconductor foundries primarily compete against one…
• FY2025 20-F: …new ones. The semiconductor market is characterized by rapid change, including rapid technological developments, evolving industry standards, changes in customer and end-user requirements, frequent new product introductions and enhancements, and short product life cycles with declining prices as products mature. Our…
• ASX (ASE Technology Holding Co., Ltd.)
• FY2025 20-F: …may negatively impact our business. Increasing competition may lead to declines in product prices and profitability and could have a material adverse effect on our business, financial condition, results of operations, and future prospects. Our profitability depends on our ability to respond to rapid technological…
• FY2025 20-F: …the end-use applications of various products, such as communications, computing, and consumer electronics products. Any deterioration of conditions in the markets for the end-use applications would reduce demand for our services and would likely have a material adverse effect on our financial condition and results of…
• AMKR (AMKOR TECHNOLOGY, INC.)
• FY2025 10-K: …functions, such as processors, sensors and connectivity devices, into small form factors, which requires innovation in advanced packaging. We have a strong position across multiple device functionalities within premium and high-tier smartphones. We are collaborating with industry leaders as smartphones transition to…
• FY2025 10-K: …the use of shorter, thinner gold wire and migration to copper wire. However, we typically do not have long-term contracts that permit us to impose price adjustments, and market conditions may limit our ability to do so. Significant price increases may materially and adversely impact our gross margin in future periods…
• SKYT (SkyWater Technology, Inc.)
• FY2025 10-K: …price erosion, wide fluctuations in product supply and demand, evolving technical standards, and short product life-cycles for semiconductors and the end-user products in which they are used. In addition, changes in general economic conditions also can cause significant upturns and downturns in the semiconductor…
• FY2025 10-K: …competition in the future. Many of our current and potential competitors have longer operating histories, greater name recognition, access to larger customer bases, and significantly greater financial, sales and marketing, manufacturing, distribution, technical, and other resources than us. As a result, they may be…
• MU (Micron Technology, Inc.)
• FY2025 10-K: …competition and DRAM and NAND oversupply due to significant investment in the semiconductor industry, including by the Chinese government and various state-owned or affiliated entities, such as CXMT and YMTC. In addition, the May 2023 decision by China's Cyberspace Administration (the "CAC") that critical information…
• FY2025 10-K: …concentrations and customers, see Part II, Item 8. Financial Statements and Supplementary Data, Notes to Consolidated Financial Statements, Note 28. Certain Concentrations. Competitive Conditions We face intense competition in the semiconductor memory and storage markets from a number of companies, including Samsung…
• TXN (TEXAS INSTRUMENTS INCORPORATED)
• FY2025 10-K: …storage Communications equipment Wireless infrastructure (3% of TI revenue) Wired networking Broadband fixed line access In addition, we sell calculators, which was about 1% of our revenue. Market characteristics Competitive landscape Despite consolidation, the analog and embedded processing markets remain highly…
• FY2025 10-K: • Our ability to compete in products and prices in an intensely competitive industry; • Market demand for semiconductors, particularly in the industrial and automotive markets, and customer demand that differs from forecasts; • Losses or curtailments of purchases from key customers or the timing and amount of customer…
• ADI (ANALOG DEVICES INC)
• FY2025 10-K: …industry, and we expect this competition to increase in the future, including from companies located outside of the United States. Competition is generally based on innovation, design, quality and reliability of products, product performance, features and functionality, product pricing, availability and capacity,…
• FY2025 10-K: …can benefit from such government incentives and we cannot, it could strengthen our competitors' relative position and have a material adverse effect on our reputation and business. Existing or new competitors may develop products or technologies that more effectively address the demands of our customers and markets…

Methodology Note

• Priced-in inversion: the valuation is inverted on the current price to recover the operating-income growth, duration, and steady-state margin the price embeds (ROE for financials, FFO growth for REITs).
• Valuation x-ray: the valuation models, grouped into four families (asset, earnings, relative, growth). Each model is expressed as a price/FV ratio (distance from price), not a point fair-value estimate. The spread across families is the disagreement.
• Solvency: net cash/debt, net-debt-to-NOPAT, interest coverage, and share-count CAGR from EDGAR financials (net debt / FFO and fixed-charge coverage for REITs; regulatory-capital framing for financials).
• Peer cohorts: per-segment comparables with deep-linkable SEC filing citations.

Fundamentals sourced from SEC EDGAR filings. Current price from Databento. The priced-in inversion and valuation x-ray are computed by the boothcheck engine; narrative composed by AI from the structured data.