Taiwan Semiconductor Manufacturing (TSM)

Total return at 5 years: 7.7%/year = 6.6% appreciation + 1.1% dividend. The target price ($611) is ex-dividend; the $29 in dividends collected over 5 years are added separately.

By both methods, the value today (DCF $498 · Multiples $516) exceeds the market price ($444).

Today — fairly valued: at $444 trades ~13.9% below its value discounted to today (~$516); the discount is positive but does not reach the margin of safety we require (≥38%).

At 5 years — En valor: the target price ($611) plus dividends yield between the 4% floor and the 10% average return — a reasonable return, though without the margin of a great investment.

The bridge: the return at 5 years exceeds the risk-free rate (4.5%) — but the discount does not reach the required margin of safety (≥38%). To require a 15% annual return, it would need to be bought at ~$323.

Educational / informational. Does not constitute investment advice.

The inverse approach: instead of projecting growth to obtain the value, the market price ($444) is taken as given and it solves for what annual owner-earnings growth would need to hold for 5 years for the present value —at the method's rate (4.5%, no-growth terminal)— to equal that price. It is the disconfirmation test: the expectations the price already pays for, contrasted against the method's projection.

The market discounts less growth (15.1%/year) than we project (18.0%/year) → if the base case holds, there is margin: perception is more pessimistic than the estimated reality.

That growth implies ~NT$3,418.4 bn of owner earnings in year 5 (vs ~NT$3,866.5 bn of our base case). It recalculates if the DCF assumptions are edited.

Year-by-year projection of the selected scenario. From each year, two versions of the flow are derived: growth FCF (operating flow − total capex, the cash surplus) and maintenance FCF (the owner earnings: what the business yields if it only sustains its capacity). Cash accumulates the retained surplus —what is not returned as dividend or buyback—, so that EV falls and multiples compress going forward. The valuation is done on EV/NOPAT. In edit mode, revenue, margins, capex, and exit multiples can be adjusted.

A model in NT$ (Taiwan dollars), the reporting currency: the company bills and reports in NT$ and trades in US dollars, so the entire fundamental apparatus —metric, multiple, cash and debt bridge, owner earnings— runs in NT$ and the conversion happens only once, at the close of the waterfall, at the live exchange rate. The model's price anchor is the New York quotation, in dollars. Everything expressed per share is on a depositary certificate basis, which represents five ordinary shares: the filing confirms it by declaring 5,313,450,863 ordinary shares represented by 1,062,690,167 certificates at 28 February 2026, so the 25,930.6 million diluted ordinary shares equal 5,186.12 million certificates. Year 0 is the fiscal year ended 31 December 2025 and not a rolling twelve-month window: the company files an annual 20-F report and publishes its quarters on 6-K with no tagged data, so manufacturing a rolling period would mix a partial window with full fiscal years (the annual reporter convention). The path starts at 33% for 2026, within the band of the trajectory (−4.5% in 2023, 33.9% in 2024, 31.6% in 2025) and supported by the already-published first half of 2026, which grew 35.6% year over year (NT$2,404.5 billion against NT$1,773.0); the second-quarter earnings release (6-K of 16-Jul-2026, exhibit 99.1) publishes no full-year guidance, but it does for the third quarter of 2026: revenue of between US$44,600 and US$45,800 million, a gross margin of 65% to 67% and an operating margin of 56% to 58%, at an assumed exchange rate of 32 NT$ per dollar. The year-1 capex level comes from the capital investment guidance of between 52,000 and 56,000 million dollars for 2026 declared in the 20-F. The monthly revenue for July 2026 (6-K of 10-Aug-2026) confirms the acceleration: January-July 2026 totaled NT$2,872.06 billion, 37.0% more than the same period of 2025 (NT$2,096.21), above the 35.6% of the first half and the 33% at which the path starts. From there the path decelerates smoothly to 10% in year 5, the durable rate of a foundry growing with high-performance computing demand once the current cycle is normalized. The year-0 operating margin is the reported one (50.8%) and that of year 1 picks up the 59.3% of the first half of 2026, with gradual compression afterward from the dilution brought by the Arizona, Japan and Germany plants and from the depreciation of the 2-nanometre roll-out. Maintenance capex is set at 40% of total capex and not at the shortcut of equating it to depreciation: with revenue growing strongly, the Greenwald criterion assigns most of the investment to expansion, because fixed assets over sales run around one time and the projected increase in sales absorbs close to three quarters of year-1 capex. Cash accumulates because it is real and verifiable surplus: the balance went from NT$2,767.9 billion at the close of 2025 to NT$3,134.2 at 30 June 2026, and the company does not repurchase shares, so the retained fraction comes from the real distribution (dividends paid of NT$466.8 billion against free cash flow of NT$1,002.6). The current declared dividend is NT$7.00 per ordinary share per quarter, or NT$140 a year per certificate, taken from the dividend table in the filing; the market provider's payment series lags the declaration by almost three quarters, which is why the declared figure is used.

Today's elevated multiple is the price of growth: if the business grows, the entry point cheapens on its own going forward (EV falls as cash increases). The exit multiple at 3 years is higher than the terminal at 5 years —at 3 years there is more growth still ahead—, so the value curve shows whether value creation is concentrated in the early or the later years. The required return is applied to the base scenario.

They are valued separately, with their own rationale, and are not incorporated into the base or the verdict (they are excess return). When assigning them value — in Editmode —, the total with optionalities updates live, without moving the base.

The verdict, the base CAGR, and the margin of safety are always calculated on the base; optionalities do not alter them (with optionalities at $0 they do not move).

Each card fixes a required annual return and answers: if the business is worth $611 in 5 years, what is the maximum that can be paid today to obtain that return? Since it now trades at $444, the margin of safety is how much cheaper the market is than that maximum. The three thresholds: 4% covers inflation (the floor), 10% is the long-term average return, and 15% is the level of a great investment.

With a target price of $611 in 5 years plus $29 of dividends collected (the dividend adds to the return, not to the price) and a required return of 4.5% annually, the maximum to pay today is $516. Against the current market price ($444), the margin of safety is 13.9% (trades below the maximum → there is margin) and the total return at that price would be 7.7% annually.

• Entry multiple. 41× on after-tax operating profit for fiscal 2025, a year that still does not capture the 59.3% margin from the first half of 2026.
• Exit multiple. 23 times, within the band for the semiconductor archetype and toward the high end on a wide moat and elevated return on capital.
• Estimated total return. +8% annually over five years at $444, against a value per depositary receipt of $611.
• Dividend contribution. 1.0% of current yield, with the declared dividend rising from NT$5.00 to NT$6.00 per ordinary share per quarter during 2025 and to NT$7.00 in the two already-approved quarters of 2026.
• Second lens. The present value discounted at the risk-free rate with a floor provides the alternative reference; the terminal value is computed with no perpetual growth.

Each margin over sales, year by year: historical (solid line) → projection (dotted).

• ✓ ROIC exceeds the cost of capital (~10%)
• ✓ CFROIC backs up the ROIC (105%, cash vs. accruals)
• ✓ Healthy balance sheet (low corporate debt)
• ✓ Durable competitive moat (multiple advantages)

• Return on capital. 43.9% on the balance sheet's actual invested capital, well above the 10% bar; 24.7% without netting the excess cash.
• Cash generation. Operating cash flow of NT$2,275.0 billion in 2025 against net income of NT$1,697.6 billion.
• Conversion to free cash. Free cash flow was NT$1,002.6 billion after investment of NT$1,272.4 billion: available cash is depressed by voluntary expansion, not weakness.
• Stock-based compensation. Irrelevant, NT$0.2 billion in the first half of 2026, so there is no hidden cost in the metric.
• Capital intensity. Investment equals roughly a third of revenue: the business compounds, but demands feeding it every year.

Annual levels from the official filing (10-K); the % over each bar is the year-over-year (YoY) growth vs the prior year.

This company's indicators are physical and mix-based, not platform-based: wafers shipped measure volume, and the share of advanced technology measures average price, the lever that explained close to half of 2025's growth. Gross margin tracks capacity utilization. The dividend is expressed per depositary receipt, equivalent to five ordinary shares: NT$110 declared in 2025 and NT$140 at the current quarterly pace of NT$7.00 per ordinary share, which is the current annualized rate and not a historical declared figure.

The % are the annual (year-over-year) growth: each year —historical and projected— vs the prior one; the TTM (trailing 12m) vs the TTM of a year ago, to avoid overlapping windows. The historicals are exact figures from the official filings; the projected years come from the year-by-year model (the intermediate years without their own series are interpolated between the anchors). The projected columns (+1y…+5y) are 12-month windows counted from the TTM close (31-dic-2025): the projection starts from the most recently reported data, not the fiscal year. The projected base is realistic and unbiased — the risk discount is applied at the end, via the required return. The rationale for each metric is in the (i).

• Recent trajectory. 33.9% in 2024 and 31.6% in 2025, with the first half of 2026 at 35.6% year over year.
• Volume lever. Shipments went from 12,910 to 15,022 thousand 12-inch equivalent wafers.
• Price lever. 7-nanometer or below technology rose from 69% to 74% of wafer revenue.
• Reinvestment runway. Guided 2026 investment is $52,000 to $56,000 million, with 2-nanometer capacity and advanced packaging both expanding.
• Dependence on the computing cycle. High-performance computing became 58% of revenue: acceleration depends on that cycle continuing.

• Ten-year relevance. Advanced semiconductor manufacturing underpins high-performance computing, phones and automotive; the leading edge is becoming scarcer, not less relevant.
• Differentiated product. Not a generic product: at the leading edge there is a single supplier able to manufacture in volume, and that shows up in the average price.
• Pricing power. In 2025 almost half of growth came from price and mix, with gross margin rising from 56.1% to 59.9%.
• Operating leverage. Mostly fixed cost from owning the fabs: it amplifies margin when utilization rises and compresses it when it falls.
• Behavior in downturns. In 2023 revenue fell 4.5% and operating margin dropped to 42.6%: the cycle exists and is felt.

Not all debt is equal: only the structural needs refinancing; the rest is operational (self-liquidating).

Structural debt is what is exposed to the contraction phase of the cycle; operational debt (leases, matched funding) self-liquidates with the business.

• ✓Leverage (net debt / EBITDA)Net cash $1,735 bn
• ✓Interest coverage (EBIT / interest)156.4x
• ✓Liquidity (current ratio)2.51×
• ✓Cash quality (CFROIC vs ROIC)CFROIC backs 105% of ROIC
• ✓Value creation (ROIC − 10% bar)+34pp
• ✓Malinvestment test (capex vs incremental ROIC)Capex/D&A 1.8x
• !Float / working capitalConsumes cash $96.8 bn (positive WC)
• ✓Dilution (SBC % of revenue + shares)SBC 0.0% of revenue

A traffic-light interpreted by the method (not generic): float (negative WC) adds up, capex is judged by incremental ROIC (malinvestment test), and a lender is not subjected to corporate solvency. The (i) shows the derivation of each number.

• Net cash. NT$1,735.0 billion of net cash: NT$2,767.9 billion of cash against NT$1,032.9 billion of gross debt.
• Coverage. Operating profit covers financial expense more than one hundred fifty times over.
• Current liquidity. Current assets of NT$3,817.1 billion against current liabilities of NT$1,522.4 billion (20-F).
• Funding the expansion. 2025 investment was covered by the year's operating cash flow, with bond issuance as a complement.
• Commitments tied to state incentives. Funds received in the United States and Germany are subject to compliance conditions with clawback risk.

How cash comes in and how it is deployed. In green, the business's own cash (the owner-FCF it generates and reinvests); in gray, the float and credit — customer and funding money, which is not the shareholder's.

The split is calculated on 2025 owner earnings: expansion investment, that is, total capex of NT$1,272.4 billion less the maintenance portion, absorbs the larger share, and cash dividends paid totaled NT$466.8 billion. There is no buyback or material acquisitions. The retained fraction, close to half of free cash flow, accumulates on the balance sheet: cash went from NT$2,767.9 billion at year-end 2025 to NT$3,134.2 billion at June 30, 2026.

Who owns the shares — the alignment and whether there is a controlling shareholder.

• Capital allocation. Reinvestment at a return well above the cost of opportunity, a growing dividend, and no buyback or acquisitions to distract.
• Financing. Expansion is funded with internal cash flow and corporate bonds, with financial expense of NT$12.4 billion per year.
• Alignment. Directors and executives hold 0.23% of the capital; there is no controlling shareholder, and the state fund holds 6.38% with one designated director.
• Transparency. The filing breaks down mix by node, by platform and by customer headquarters, and publishes next fiscal year's investment guidance.

• Jurisdiction-risk discount: most manufacturing is in Taiwan, and the filing itself acknowledges that markets use episodes of tension with mainland China to depress the price of Taiwanese securities.
• Missing buyers via the vehicle: foreign investors access the stock through a depositary receipt rather than the ordinary share, and part of the institutional capital has mandates that limit exposure to the region.
• Regulatory noise with a known date: tariffs on advanced chips, trade investigations and an annual export license for the Nanjing fab that renews with no guarantee.

The price is not depressed by doubts about the business, which grew 31.6% in 2025 and 35.6% in the first half of 2026 with operating margin rising to 59.3%. The gap between perception and reality lies in the jurisdiction discount: the market charges for a binary and hard-to-bound risk on a company whose process moat widened over the same period. Subsequent event that should be disclosed: the first half of 2026 includes a NT$63,202 million gain from the sale of equity-method investments, which is non-recurring and is not projected.

The annual return (CAGR at 5 years) in each scenario, with the total period return below — the margin of safety made visual: upside range wide, downside range narrow.

Even in the bear scenario, the return holds at -9%/year (-38% total): the margin of safety protects the downside. The bull (+17%/year, +123% total) exceeds it comfortably — a favorable asymmetry, with a narrow downside range and a wide upside range.

• Geopolitics and jurisdiction. Most production and leadership are in Taiwan; the filing acknowledges that episodes of tension with mainland China depress the price of Taiwanese securities.
• Tariffs and export controls. 25% tariff on certain advanced computing chips following the Section 232 investigation, and Section 301 investigations aimed in part at Taiwan.
• Nanjing fab license. The validated end-user authorization expired in December 2025, and the fab operates under an annual license with no guarantee of renewal.
• Customer concentration. The top ten went from 70% of revenue in 2023 to 78% in 2025, and the second-largest customer rose from 11% to 17%.
• Execution outside Taiwan. Fabs in Arizona, Japan and Germany with higher costs, scarce labor and different regulatory regimes.
• Cycle with fixed cost. The filing describes prolonged periods of downturn and overcapacity in the industry, with operating cost mostly fixed.

Each thinker's analytical framework applied to our data.

• Buffett / Graham Quality + margin of safety

  A wide moat and ROIC 44% above the 10% bar, but the margin is limited (+14%) → excellent business at a fair price.

• Peter Lynch Growth at a reasonable price (GARP)

  A fast grower growing 18% at a multiple/growth of 2.3 → expensive for its growth.

• Joel Greenblatt Cheap and high-return (Magic Formula)

  Earnings yield 3% (EBIT/EV) + ROIC 44% → falls outside the Magic Formula.

• Howard Marks Perception vs reality + cycle

  The price discounts 15% vs our 18%: perception is more pessimistic than reality, with an identified source of the discount.

• Seth Klarman Capital protection (bear scenario)

  Bear-scenario floor -11%/yr over 5y (material loss) → risk of permanent capital loss.

• Pat Dorsey Moat strength (Five Rules)

  A wide moat, widening; sources: intangibles, efficient scale, switching costs, cost advantage, network effects → passes the Five Rules.

• Aswath Damodaran Expectations implied by the price

  Justifying the price requires discounting 15%, within what we project (18%) — the story squares with the numbers.