Meta Platforms (META)

Total return at 5 years: 9.6%/year = 9.3% appreciation + 0.3% dividend. The target price ($1,159) is ex-dividend; the $14 in dividends collected over 5 years are added separately.

By both methods, the value today (DCF $1,061 · Multiples $942) exceeds the market price ($743).

Today — fairly valued: at $743 trades ~21.2% below its value discounted to today (~$942); the discount is positive but does not reach the margin of safety we require (≥38%).

At 5 years — En valor: the target price ($1,159) 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 ~$585.

Educational / informational. Does not constitute investment advice.

The inverse approach: instead of projecting growth to obtain the value, the market price ($743) 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 (3.7%/year) than we project (12.0%/year) → if the base case holds, there is margin: perception is more pessimistic than the estimated reality.

That growth implies ~$86.7 bn of owner earnings in year 5 (vs ~$127.2 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). The flow is returned almost in full (dividend + buyback) or redeployed into the operation, so that EV stays roughly flat and multiples compress because the metric grows, not because of cash accumulation. The valuation is done on EV/NOPAT (owner earnings). In edit mode, revenue, margins, capex, and exit multiples can be adjusted.

Rolling grid: the base year is the real TTM (revenue $228.2bn, operating cash flow $130.3bn to Jun-2026 — FY2025 10-K plus the first half of 2026 from the earnings release) and the projection is 12-month windows (+1y…+5y). The revenue path starts at +21.5%, anchored to the company's own guidance for the following quarter ($61-64bn, that is +19% to +25%) and not to the +27.6% of the TTM. The underlying change is in the margin: second-quarter operating income fell 8% year over year and the margin went from 43% to 31%, with R&D up 67%; total expense guidance for 2026 ($165-169bn) implies an operating margin on the order of 34% for the calendar year, against 41.4% in 2025. That is why the base-year margin (38.1%) falls to 34% in year 1 and recovers only partially, to 38% by year 5 — the depreciation of artificial-intelligence capex is a cost that does not go away. The valuation metric is NOPAT (EBIT after tax), which expenses stock-based compensation ($25.1bn, ~11% of revenue — a real cost that reported free cash flow adds back). Maintenance capex is split by Greenwald (fixed assets over sales × change in sales), which leaves ~45% of total capex as maintenance rather than the shortcut of setting it equal to depreciation. Buybacks are out of the model: they were zero in the first half of 2026 (TTM $3.3bn against $26.3bn in the prior year), so the share count stays flat and cash does not accumulate — the flow is being redeployed into capex, which guidance takes to $130-145bn in 2026. Watch: incremental return on capital, which in this TTM fell to 8.8%, below the 10% bar.

Today's elevated multiple is the price of growth: if the business grows, the entry point cheapens on its own going forward (the metric grows while EV stays roughly flat). 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 $1,159 in 5 years, what is the maximum that can be paid today to obtain that return? Since it now trades at $743, 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 $1,159 in 5 years plus $14 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 $942. Against the current market price ($743), the margin of safety is 21.2% (trades below the maximum → there is margin) and the total return at that price would be 9.6% annually.

• One business, one multiple. Family of Apps drives the value and is valued on owner earnings (EV/NOPAT, which expenses stock-based compensation ~$22bn — reported FCF adds it back). Reality Labs is treated separately as optionality — this isn't a sum of positive parts.
• Today's multiple is misleading, cheapens forward. Reported FCF, depressed by capex and gross of stock-based compensation, is the wrong lens; on owner earnings (≈NOPAT, ~$73.5bn TTM) and as the capex pays off, the multiple cheapens (26× → 12×).
• The DCF converges. The DCF (at the method's risk-free rate, zero-growth terminal) yields a value today well above the price: it confirms the undervaluation through a different lens.
• Margin of safety at the end. At market price the base CAGR ~17% exceeds the 15% hurdle; the margin lives in the required hurdle, not in each assumption.
• Optionalities kept separate. Reality Labs, Meta AI, WhatsApp and Threads are valued separately (default $0) → they don't contaminate the base case; they're upside.
• Sensitivity — the capex. The value is sensitive to the return on AI capex: if it's malinvestment, FCF stays depressed and the multiple re-rates. It's the risk of the case.

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

• ✓ ROIC exceeds the cost of capital (~10%)
• ✕ High incremental ROIC (compounding)
• ! CFROIC backs up the ROIC (75%, cash vs. accruals)
• ✓ Healthy balance sheet (low corporate debt)
• ✓ Durable competitive moat (multiple advantages)

• Strong, predictable cash generation. Reported FCF ($46.1bn in FY25) is depressed by growth capex; maintenance FCF (~$86bn TTM) shows the real cash the core throws off.
• Sustainably high ROIC. Yes — ~32% at the level, ~35% incremental. CFROIC (~24%) backs a good share of ROIC with cash (less accrual than a lender).
• Creates economic value (EVA). CFROIC (~24%) comfortably clears the 10% bar → strong value creation per dollar of capital.
• Reinvestment runway. The AI capex (~$70bn) is the runway — the malinvestment test (is incremental ROIC > cost of capital?) is the central theme of the case; Q1'26 suggests it's paying off so far.
• The SBC trap. High SBC (~$22bn, ~11% of revenue): reported FCF and Adj-EBITDA add it back → value on owner earnings/NOPAT (which expense it), never on reported EV/FCF.
• Accounting distortions to watch. (1) growth capex depresses FCF; (2) high SBC inflates the headline FCF; (3) Reality Labs (−$19bn) masks the core's real profitability.

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

Advertising revenue is volume times price: daily active users times revenue per person. The series show completed fiscal years, which is the comparable basis the filings publish. What the second quarter of 2026 adds is a change in the mix of the engine: daily users reached 3,600 million but are growing just 3% year over year, while revenue rises 28% — meaning practically all the growth comes from monetizing the existing base better (impressions +14% and price per ad +12%), not from adding people. Capital expenditure, which was $89.3bn over the trailing twelve months and is guided to $130-145bn for 2026, sustains the infrastructure that makes that monetization possible.

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 (30-jun-2026): 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).

• Real growth, not nominal. Yes — +22% FY25 (+33% in Q1'26) in USD, driven by volume (impressions +12%) and price (+9%), not FX or an easy comp.
• Where the engine comes from. Broad-based by geography: US/Canada (39% of revenue, +25%), Asia-Pacific (27%, +20%), Europe (23%, +21%). Not dependent on a single region.
• Driven by volume and price. Impressions +12% × price per ad +9%: AI lifts both (more engagement, better targeting) — quality growth, not just pricing.
• Realistic deceleration. The base starts from the real trajectory (~+22% FY24/FY25, TTM +26%) and decelerates +21% → +12% by year 5; it doesn't extrapolate the Q1'26 peak (+33%), which is left as the bull scenario.
• Reinvestment runway. Wide: monetization of WhatsApp/messaging and Threads is just starting, and AI opens new surfaces (Meta AI, generative ads).
• Capital cycle / competition. It competes for attention (TikTok) and for AI talent/compute; the $70bn capex is both defense and offensive bet.

• Relevant in 10 years / recurrence. Yes — social networks and messaging are daily habits for 3,600 million people; attention is recurring and hard to dislodge.
• Market share / cost. Dominant in social and messaging outside China; scale keeps cost per user very low and makes the ads more effective than an entrant's.
• Pricing power. Real: price per ad rose +9% in 2025 (AI improves ad value) without losing impressions (+12%).
• Operating leverage. High: revenue grows faster than core costs (Q1'26: revenue +33%, net income +61%), though AI capex and Reality Labs mask it.
• Supply-side / capital cycle. AI is drawing massive capital; META is responding with its own capex ($70bn) — the risk is a spending race that compresses returns if it doesn't monetize.
• Recession behavior. Advertising is cyclical (advertisers cut first), but scale and performance advertising (measurable ROI) make it more defensive than traditional advertising.

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 $6.6 bn
• ✓Interest coverage (EBIT / interest)86.9x
• ✓Liquidity (current ratio)2.23×
• !Cash quality (CFROIC vs ROIC)CFROIC backs 75% of ROIC
• ✓Value creation (ROIC − 10% bar)+18pp
• ✕Malinvestment test (capex vs incremental ROIC)Investment mode 3.9x with incremental ROIC < cost of capital
• –Float / working capitalNeutral WC
• !Dilution (SBC % of revenue + shares)SBC 11.0% 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.

• Controlled leverage. Debt ($59bn in senior notes) nearly doubled due to AI capex, but it's low against ~$62bn of annual owner earnings and a strong cash position → positive net cash.
• Ample coverage. Interest expense (~$1bn) is trivial against operating income ($83bn) — coverage of dozens of times.
• Capex / lease commitments. Beyond debt, ~$104bn of lease obligations not yet commenced (data centers): the commitment behind the AI bet is enormous; the risk is one of return.
• No maturity wall. Senior notes are spread out (nothing in 2026; ~$4bn in 2027-28; the bulk, ~$49bn, after 2030) and cash generation covers them comfortably.
• Investment phase, not deterioration. The jump in debt and capex is voluntary AI investment, not a balance sheet in trouble: solvency is solid; the risk is one of investment return.

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.

Capital allocation changed shape in a single quarter. Of trailing twelve-month operating cash flow ($130.3bn), almost all is reinvested: $89.3bn of capital expenditure, with guidance taking 2026 spending to $130-145bn. Buybacks —$26.3bn in 2025— stopped entirely in the first half of 2026, and to fund the pace the company issued $24.9bn of debt. Only the dividend is maintained ($5.4bn annually). The malinvestment test is what decides whether this creates or destroys value, and over the trailing twelve months it produced an 8.8% incremental return on capital: below the 10% bar.

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

• Capital allocation (ROIC). Reinvests aggressively in AI (~$70bn) at a historically high ROIC, and returns capital (buybacks $26bn + dividend $5.3bn). The test is whether the incremental capex sustains ROIC.
• Alignment / skin in the game. Founder-led (Zuckerberg since 2004) with ~13% of the capital. Maximum incentive alignment, but with ~60% voting control (Class B) that leaves the minority shareholder with no real check — a double-edged sword.
• Buybacks at a good price. Repurchased $26.3bn (40M shares) in 2025; the challenge is that high SBC (~$22bn) means a good share of the buyback only offsets dilution rather than reducing net shares.
• Integrity / communication. Communicates the capex and Reality Labs losses transparently (quantifies them), though the metaverse/AI rhetoric is ambitious.

• The market is discounting the investment cycle: capital expenditure guided to $130-145bn left second-quarter free cash flow at $784 million, and a reading that does not distinguish maintenance from growth sees there a business that stopped generating cash. The distinction matters: maintenance capital expenditure, split by the Greenwald criterion, is ~$40bn of the $89.3bn spent over the year.
• The second-quarter charges —$2.40bn of legal contingencies and $1.18bn of severance— sink reported earnings without touching the value of the business; this is the kind of non-recurring item that punishes the headline more than the reality.
• Trailing twelve-month net income carries a tax rate distorted in both directions: fiscal year 2025 had non-recurring tax charges that pushed the effective rate to 29.6%, and the first quarter of 2026 a $5.0bn benefit. The company guides 15-17% going forward.
• Reality Labs loses $19.1bn a year and contaminates consolidated earnings; many participants penalize the loss instead of separating it as the option it is.

Market perception and business reality separated in opposite directions and should not be confused. On the side where the market overreacts: the non-recurring charges and the reading of cash flow without distinguishing maintenance from growth. On the side where the market may be right: the incremental return on capital fell to 8.8%, below the bar, and that is not optical. The opportunity depends on that compression being a matter of timing rather than structural — which is exactly what the adverse scenario tests.

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 -6%/year (-28% total): the margin of safety protects the downside. The bull (+21%/year, +157% total) exceeds it comfortably — a favorable asymmetry, with a narrow downside range and a wide upside range.

• AI capex malinvestment. $69.7bn (guidance ~$100bn), partly debt-financed; if incremental ROIC falls below the cost of capital, FCF stays depressed and the multiple re-rates toward cash. It's the central risk.
• Reality Labs drains cash. Loses ~$19bn/year and the loss keeps growing; if it never reaches a viable platform, it's sustained value destruction.
• Regulation / antitrust. The EU (DMA/DSA) and the FTC press on the advertising model (privacy, targeting) and even the structure (possible divestitures of Instagram/WhatsApp).
• Competition for attention. TikTok and short-form video compete for users' time; a loss of engagement hits impressions.
• Dual-class governance. Zuckerberg controls ~60% of the vote with ~13% of the capital: the minority shareholder has no real check over major decisions (capex, acquisitions, the metaverse).

Each thinker's analytical framework applied to our data.

• Buffett / Graham Quality + margin of safety

  A wide moat and ROIC 28% above the 10% bar, but the margin is limited (+21%) → excellent business at a fair price.

• Peter Lynch Growth at a reasonable price (GARP)

  A fast grower growing 12% at a multiple/growth of 2.2 → expensive for its growth.

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

  Earnings yield 5% (EBIT/EV) + ROIC 28% → falls outside the Magic Formula.

• Howard Marks Perception vs reality + cycle

  The price discounts 4% vs our 12%: perception is more pessimistic than reality, with an identified source of the discount.

• Seth Klarman Capital protection (bear scenario)

  Bear-scenario floor -7%/yr, bull-scenario ceiling +21%/yr over 5y: reasonable asymmetry, without an ample cushion.

• Pat Dorsey Moat strength (Five Rules)

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

• Aswath Damodaran Expectations implied by the price

  Justifying the price requires discounting 4%, within what we project (12%) — the story squares with the numbers.