AVGO Broadcom Inc.

Dive Nov 16, 2026. Price at dive $332.78. Looks cheap Price at the dive was below our base value.

Researched Cases

Bear
$416.49
Base
$456.06
Bull
$498.82

The base case is the forecast believed most likely, one path. Bear and bull are the same path with the top three swing factors each moved by a researched step, all three in the same direction. Everything else is identical across cases.

Key Assumptions, Stated

Identical in all three cases. Built from SEC filings. Equity risk premium: 3.70% (Damodaran, Oct 8, 2026).

Swing Factors

The three swing factors (KPIs: the few numbers the value hinges on) behind the cases. Each row shows the researched base path and the bear and bull flexes.

Swing FactorBase PathBear FlexBull FlexWhy It Matters
AI semiconductor revenue growth AI revenue $57.6B FY2026 to $108B FY2027 (+87.5%), $195B FY2028 (+80.6%), fading to +3.2% by FY2036 ($425B) AI revenue growth 5% lower each year than base (×0.95): FY2027 +83.1% ($105.5B), FY2028 +76.5% ($186.2B), fading to $391.3B by FY2036 AI revenue growth 5% higher each year than base (×1.05): FY2027 +91.9% ($110.5B), FY2028 +84.6% ($204.0B), rising to $461.1B by FY2036 AI is 55% of FY2026 revenue and essentially all of the growth; the FY2027-28 doubling guidance drives most of the DCF value, so whether it ships is the valuation.
Non-AI revenue growth (semiconductors ex-AI plus infrastructure software) Non-AI revenue $48.3B FY2026 to $57B FY2027 (+18.0%), $60B FY2028 (+5.3%), then +5-8% a year to $100B by FY2036 Non-AI revenue growth 5% lower each year than base (×0.95): FY2027 +17.1% ($56.6B), FY2028 +5.0% ($59.4B), to $96.6B by FY2036 Non-AI revenue growth 5% higher each year than base (×1.05): FY2027 +18.9% ($57.4B), FY2028 +5.5% ($60.6B), to $103.5B by FY2036 The $48.3B non-AI base (VMware software plus broadband, storage and wireless) is the ballast; its growth rate sets the floor under total revenue if AI disappoints.
FCF margin (FCF as % of revenue) 44% of revenue in every forecast year (Q3 FY2026 46%, trailing 44.2%) FCF margin at 43% vs base 44% (1.0pt lower; evidence: quarterly FCF margin ranged 41% in Q1 to 46% in Q2/Q3 FY2026) FCF margin at 45% vs base 44% (1.0pt higher; evidence: quarterly FCF margin ranged 41-46% in FY2026) Every revenue dollar converts at this margin, so one point moves the whole 10-year FCF stream and the terminal value by roughly 2%.

Run Your Own Version

Each workspace has three sliders, one per swing factor. The sliders need JavaScript; the researched values above work without it. Move the sliders past the bear and bull marks, or type any value you believe. The researched cases stay shown above. Your value updates as you go.

See the research

Summary

Base $456.06 rests on three findings: (1) Broadcom's AI doubling path largely delivers, haircut for concentration risk, AI revenue $108B in FY2027 (vs $115B guided) and $195B in FY2028 (vs $230B guided), with signed Anthropic leases ($161B) and the XPV financing platform behind it; (2) the 44% FCF margin holds as AI hardware scales (Q3 FY2026 46%, trailing 44.2%); (3) a 10.37% WACC built on Damodaran's 3.70% headline ERP and an observed 4.95% cost of debt. What would change the call: the Dec 9 Q4/FY2026 report cutting the FY2027 AI guide, an XPV backstop being called, Anthropic/OpenAI funding trouble, or wafer/HBM supply falling short of the doubling.

Assumptions, Stated

InputResearched ValueDecisionEvidence
Risk-free rate 5.10% kept FRED DGS10 5.10% on Nov 12, 2026, up from 4.98% Nov 10 and 4.84% Nov 5; cross-checked against Yahoo Finance 10-year yield index at 5.125%. Kept. Source
Equity risk premium 3.70% (Damodaran current headline implied ERP, Oct 8, 2026) kept Damodaran's current headline implied ERP is 3.70% as of Oct 8, 2026 (expected return on stocks 8.99% less the then 10-year Treasury 5.29%), on the trailing-12-month cash-flow basis. The 4.20% figure circulating is his adjusted-payout variant as of Oct 1, 2026, not the headline, and does not outrank it. The tool payload already carries the correct 3.70% headline; no stale fallback was present to correct. A November monthly update could not be verified from his page this run, so the Oct 8 headline is the latest verified figure in hand. Source
Beta 1.49 kept StockAnalysis 1.51, MarketBeat 1.49, FI Handbook 1.51, managed market-data channel 1.46, self-computed 5-year monthly regression vs S&P 500 1.48. Dispersion is only 0.04-0.05 across vendors, so the median 1.49 is stable; beta 1.46-1.51 moves the value only $451.71-$463.70. Source
Cost of debt 4.95% pre-tax, observed (3.71% after tax at 25%) replaced Observed recent issuance beats the formula: Broadcom's Jan 6, 2026 pricing term sheet for $4.5B senior notes, 4.30% due 2031, 4.60% due 2033, 4.95% due 2036, 5.70% due 2056, with yields to maturity 4.32%/4.64%/4.96%/5.75%; Moody's A3 positive. The 10-year point (2036 notes, 4.95% coupon, 4.96% YTM) is used. The synthetic 5.85% overstates the observed rate by 90bp, but debt is only 3.6% of capital, so WACC barely moves. Source
Discount rate (WACC) 10.37% adjusted CAPM cost of equity 10.61% = 5.10% + 1.49 x 3.70%. Observed pre-tax cost of debt 4.95%, after tax 3.71% at the 25% marginal rate. Market-value weights at $332.78: equity about $1,588.5B (4,773.6M shares outstanding) vs $59.42B debt, so 96.4% equity / 3.6% debt. WACC = 10.61% x 0.964 + 3.71% x 0.036 = 10.37%.
Revenue and FCF path Revenue-split FCF path: FY2027 revenue $165B (AI $108B + non-AI $57B), FY2028 $255B (AI $195B + non-AI $60B), then growth fading 22%/15%/11%/9%/7%/5%/5%/4% to $525B FY2036; FCF = revenue x 44% replaced Filed actuals: Q1 FY2026 revenue $19.31B (AI $8.4B), Q2 $22.19B (AI $10.8B), Q3 $29.59B (AI $16.7B, +221%), Q4 guide $34.8B (AI $21.7B), FY2026 about $105.9B with AI about $57.6B. Management guides AI revenue about $115B in FY2027 and about $230B in FY2028 with supply secured, and FY2028 EPS above $30; consensus (48 analysts) is FY2027 revenue $175.5B / EPS $19.47 and FY2028 revenue about $262.5B / EPS $30.37. The base haircuts the AI guide (to $108B and $195B) for customer concentration and financing risk, landing modestly below consensus. The tool's EPS-conversion path is rejected: it implies 121x trailing earnings and a $1,538.89 DCF. Source
Margins (FCF margin) FCF margin 44% of revenue, flat across the forecast adjusted Trailing FCF margin 44.2% (FCF $39.40B on TTM revenue $89.1B). Quarterly FY2026: Q1 FCF $8.01B on $19.31B (41%), Q2 $10.26B on $22.19B (46%), Q3 $13.67B on $29.59B (46%). Non-GAAP operating margin Q3 67.9% and gross margin 75.0% show the mix is holding even as lower-margin AI hardware scales. 44% sits at the trailing level, not at the best quarter. Source
Terminal growth 3.0% adjusted The AI runway is long (XPV platform sized above 20GW through 2028, Anthropic 5GW in FY2027 plus 10GW in FY2028, OpenAI 5GW in FY2028), which supports a healthy terminal rate, but the business is hardware-cyclical, concentrated in six XPU customers, and the terminal year follows a doubling that cannot compound forever. 3.0% is below the tool default and well under the 5% cap; terminal value is still 55.9% of enterprise value.
One-time items No adjustment; no material one-time item in the trailing period kept Q3 FY2026 10-Q: restructuring charges $12M in the quarter and $35M over nine months (immaterial); restructuring and other charges $103M in Q3. Amortization of acquisition-related intangibles was $2,006M in Q3 ($1,499M in cost of revenue plus $507M in operating expenses), a recurring non-cash charge, excluded from non-GAAP earnings but not a one-time item and not added back in the FCF path. No termination or breakup fees, settlements or disposal gains in the quarter. FY2024 GAAP earnings were depressed by VMware-related costs, which is why the tool's normalized bridge adds back amortization; the FCF path used here is unaffected. Source
Share count 4,887M diluted shares kept 10-Q for the quarter ended Aug 2, 2026: weighted-average diluted shares 4,887M in Q3 and 4,884M over nine months; shares outstanding 4,773,629,865 as of Aug 28, 2026. Buybacks were $8,450M over nine months ($7.8B in Q1 alone) under a $10B authorization through Dec 31, 2026, but stock-based compensation of $6,287M over nine months offsets much of it, so the diluted count is essentially flat vs 4,860M a year ago. Source
Net debt $35.44B kept 10-Q balance sheet Aug 2, 2026: short-term debt $2,252M plus long-term debt $57,167M = $59,419M; cash and equivalents $23,975M; net debt $35,444M. The company also repurchased $5,641M of senior notes in Q3. NOT added to net debt: the AI-rack residual-value backstop (maximum about $29B on the first XPV tranche, fair value not material, nothing paid) and the second $63B Anthropic financing announced Oct 26 with about $43B of residual-value support, both are contingent, off-balance-sheet, and loss depends on a customer default plus rack resale below 85% of outstanding lease amounts, so they are flagged as the key downside risk rather than counted as debt. Customer convertible notes of up to $42B are issuable under the lease arrangements; none had been issued as of Aug 2, 2026. Source
Stock-based compensation About $8,482M trailing twelve months adjusted 10-Q: SBC $2,019M in Q3 FY2026 and $6,287M over nine months vs $5,373M in the prior nine months. Trailing SBC is $7,568M - $5,373M + $6,287M = $8,482M, about 21% of trailing FCF of $39.40B. Reported FCF already reflects SBC as a non-cash add-back inside operating cash flow, so owner FCF is lower; the SBC lens (FCF net of SBC growing 3% a year) gives $431.27 vs the $456.06 base and is carried in the sensitivity span, not deducted in the base. Source
World context (M&A, customers, competition) AI demand real and contracted, but concentrated and financed; priced as a haircut on guidance, not a separate case kept Six XPU customers led by Google (TPU co-development), Anthropic (largest customer in FY2027) and OpenAI (second-largest in FY2028), plus Meta. Anthropic's S-1 shows $161B of five-year equipment leases with Broadcom and about $895B of total purchase and lease commitments. AI XPV platform (June 2026) enables 20GW+ through 2028; first $35B tranche (Apollo/Blackstone) carries the $29B backstop; second $63B tranche (BofA, Blackstone, Citi, Morgan Stanley, Oct 26) carries about $43B residual support. Redburn downgraded to Neutral Nov 13, cutting its target to $398 from $595, questioning earnings durability and customer funding; the Fed's Nov 13 Financial Stability Report named AI the top risk to US financial stability. Marvell agreed to buy Lumentum for $115B (optical consolidation, closes H1 2028). Next checkpoint: Q4/FY2026 results Dec 9, 2026. Source

What Would Have to Be True

For the $332.78 price to be right on the base FCF path, the discount rate would have to be 12.84% instead of the researched 10.37%, or terminal growth would have to be -2.9% instead of 3.0%. In other words, the market would have to be right that the AI doubling path largely fails to convert into durable free cash flow, or that the XPV backstop and customer-concentration risks deserve roughly a 2.5-point risk premium over the CAPM rate.

Sensitivity Span (Detail Only)

$374.52 to $564.89. Full variant span as detail only: low is the stacked bear extreme (bear KPI flex plus WACC +0.5pt and terminal growth 2.5%); high is the stacked bull extreme (bull KPI flex plus WACC -0.5pt and terminal growth 3.5%). Single-factor moves: WACC +/-0.5pt gives $424.82-$491.92; terminal growth +/-0.5pt gives $438.45-$476.24; revenue growth +/-2pt gives $397.32-$523.55; beta 1.46-1.51 gives $451.71-$463.70; the SBC lens (FCF net of about $8.5B trailing stock-based compensation) gives $431.27. The memo's headline range is bear-to-bull, $416.49-$498.82. The span is detail only, never the headline range.

Cross-Checks (Tool Only)

Tool DCF $1,538.89 is rejected as a headline input: it rides the 65.9% analyst EPS CAGR through a 1.17x FCF conversion and a linear fade, implying 121x trailing earnings; the researched DCF ($456.06) replaces it. Graham $656.76 uses the same aggressive EPS path. Multiples $221.55 uses trailing GAAP EPS, which is pre-ramp and understates forward earnings power. Street context: consensus target averages about $528-533 across 51 analysts (Strong Buy), Redburn $398 after its Nov 13 downgrade to Neutral, TD Cowen $425.

Smell Tests

Educational only. Not financial advice.