R Refacto StocksThe Verdict

Verdicts / All ratings

CRUSCirrus Logic, Inc.high confidenceFiled Jul 17, 2026

Cirrus Logic, Inc.

Watch · 60/100 · high confidence

Watch
60
Council / 100

Watch · 60/100 · high confidence

CIRRUS LOGIC, INC. (CRUS) — Council Assessment

🟡 WATCH · Score 60/100 · high confidence

High-quality, fortress-balance-sheet fabless chipmaker at a fair-not-cheap price, hamstrung by extreme Apple concentration and near-zero revenue growth.

As of 2026-06-28. 18 lenses weighed in, 0 abstained. Sources: 6 filings, 15 news, 15 discussion, 1 earnings_call.

360 narrative — news & sentiment digest

Cirrus Logic (CRUS) — Investment Briefing

Management Commentary (Q2 FY2026 Earnings Call, Nov 10, 2025)

Headline Results

  • Record September quarter revenue of $561M (toward top end of guidance), up 38% sequentially and 4% YoY, driven by smartphone demand
  • Non-GAAP gross margin: 52.5% (up YoY due to favorable product mix; partially offset by higher inventory reserves)
  • Non-GAAP operating income: $167M (29.8% of revenue); net income $150M ($2.83 EPS)
  • Strong free cash flow: $92.2M operating CF, 16% FCF margin; trailing 12-month FCF margin 27%
  • Zero debt; cash position $896M

Guidance & Outlook

  • Q3 FY2026: revenue $500–$560M; gross margin 51–53%; OpEx $128–$134M
  • Tax rate: 16–18% (benefited from One Big Beautiful Bill Act retroactive R&D expensing in Q2, lowering rate to 14.6%)
  • Management flagged seasonality shift tied to camera content pulling revenue earlier in fiscal year; otherwise expects historical patterns to resume

Three-Pillar Strategy & Segment Commentary

  1. Smartphone Audio (flagship & custom) — Core business remained strong. Latest-generation boosted amplifiers and 22nm smart codec, introduced fall 2024, are shipping in high volume. Management emphasized 5–6 year longevity of custom silicon (predecessor products shipped for 5–6 years), providing visibility and R&D leverage. Android OEM design wins noted (dual boosted amplifiers + haptic driver).

  2. Smartphone Mixed-Signal Expansion — Camera controllers gaining traction as key differentiator in latest devices; multiple R&D programs in battery performance/health believed to represent "excellent opportunity" (no concrete commercialization updates yet, though management expressed confidence in IP).

  3. Non-Smartphone Markets — Most material progress in PC market:

    • Strong design activity across portfolio; SoundWire device-class adoption accelerating
    • Key wins: first mainstream commercial laptop this year, now first mainstream consumer design (revenue expected FY2027)
    • New AI/voice-enabled PC product sampling in December quarter—leveraging smartphone voice/audio IP for ultra-low-power wake and voice features; management sees this as significant SAM expansion
    • Multi-platform vendor support expanding reach
    • General market (professional audio, automotive, industrial, imaging) showing momentum: ADC/DAC/codec design wins; timing products shipping; imaging analog front-end sampling with "positive initial response"

Analyst Q&A Key Points

  • Seasonality: Reaffirmed no material changes to FY2026 smartphone unit outlook despite strong Q2; shape shift (camera pull-forward) expected to persist; no other material seasonality changes anticipated
  • Battery/Power IP: Acknowledged multiple "irons in the fire" with customers; compelling IP for power efficiency, system performance, battery health; no design-win commercialization to announce yet
  • Android softness: Acknowledged weak Android market (widely reported); not a strategic focus area but still valuable contributor
  • Non-flagship customer revenue: Noted softness partly from Android weakness; PC business tracking broadly as expected despite early-stage ramp; company passed "excellent milestones" (especially mainstream consumer win)
  • OpEx timing: Q2 OpEx came below guidance due to project timeline shifts and avoided expenses, not execution misses; CFO noted disciplined approach but willingness to increase spend on high-confidence value-creation opportunities

Management Tone: Confident and measured. Emphasized long-term strategy execution, avoided hyperbole on new markets, acknowledged customer concentration constraints (would not discuss largest customer specifics per policy).


Recent Developments

  • Q2 FY2026 results (Nov 10, 2025): $561M revenue, record quarter
  • Tax benefit: One Big Beautiful Bill Act retroactive R&D expensing effective FY2026 start, lowered Q2 tax rate to 14.6%
  • Insider trading: EVP Jeffrey Baumgartner sold 1,170 shares (~$196.8K) on June 24, 2026, under 10b5-1 plan; also exercised options and sold 3,907 shares in June ($~$650K+ total June activity)
  • Analyst coverage: Yahoo Finance comparison piece vs. Skyworks (May 2026); multiple analysts rate Buy/Strong Buy (CFRA, Market Edge, LSEG, Vickers per retail discussion)

Bull Narrative

Optimist perspective (retail + select analyst commentary):

  • High-margin, diversified growth: Smartphone audio still core and stable; camera content expanding; PC market opening with early wins in mainstream consumer (highest volume tier)
  • Long product lifecycle: Custom silicon 5–6 year durability reduces lumpiness and provides R&D leverage
  • AI-PC tailwind emerging: Voice-enabled PC products sampling now; management positioned to capture voice/audio features in AI ecosystem
  • Strong execution: Record Q2 revenue, 52.5% gross margin, 29.8% operating margin, 27% trailing 12-month FCF margin
  • Clean balance sheet: $896M cash, zero debt, $414M remaining share buyback authorization; active capital return
  • Valuation: Retail comments cite "A rating," "8.8–9.4 bullish scores," "mid-range undervalued to fairly valued" positioning; multiple predictions of $200+ target by end of 2026 (though highly speculative)
  • Analyst consensus: Multiple sources (Argus, CFRA, Fidelity, LSEG) list as Buy/Strong Buy

Bear Narrative

Skeptic perspective:

  • Customer concentration risk: Largest customer not disclosed but widely understood to be Apple; company policy refuses to discuss specifics, but earnings rely heavily on one player's new phone cycles
  • Smartphone cyclicality: YoY growth only 4% despite "record" quarter; smartphone unit growth is episodic; subject to OEM inventory/demand swings
  • Unproven adjacencies: PC market acknowledged as "early-stage ramp"; revenue contribution immaterial; voice-PC products only sampling (no revenue yet); battery/health IP nowhere near commercialization
  • Foundry & supply risk: 10-K flagged tighter TSMC/GlobalFoundries capacity driven by AI demand; geopolitical & Middle East logistics risks; onerous wafer commitment with GlobalFoundries through 2026
  • Android headwinds: Management acknowledged softness in Android segment (non-strategic but revenue contributor)
  • Talent/FX volatility: 10-K notes US immigration pressures and FX-driven OpEx swings
  • CHIPS subsidy exclusion: Fabless model excludes company from US CHIPS Act subsidies (disadvantage vs. integrated competitors)
  • Insider selling: EVP liquidating shares in June 2026; while pre-planned (10b5-1), signals confidence not at peak
  • Valuation: GuruFocus (June 2026) rates stock as overvalued despite GF Score of 80/100; Yahoo Finance comparisons suggest stock may be pricing in growth not yet realized

Retail Sentiment

Overall: Bullish, moderate-to-high conviction

  • Dominant tone across forums (Reddit, TradingView, Orbit100, etc.) is bullish buy/hold
  • Common themes: "great price to add," "easy $200 target," "9+ bullish score," "strong long-term hold," "ahead of Reddit crowd"
  • One analyst (@Orbit100, multi-post): "10+ year coverage, A rating, 9.4 bullish score, room to run, hold for retirement"
  • Specific catalysts cited: AAPL conference, earnings beats, diversification narrative
  • Minimal bearish chatter in sample; one contrarian note on foundry risk and subsidy exclusion (d_risk)
  • Options trading activity noted (May calls showing 58% ROI in ~24 days, per BillionerOfKing example)
  • Stock price at time of latest retail posts: ~$166–$174 range (May–June 2026)

Conviction level: Medium-high among retail; however, commentary often generic ("great buy," "$200 target") without deep fundamental depth.


Caveats

  1. Largest customer opacity: Management deliberately obscures Apple relationship per policy; impossible to assess true concentration risk or growth dependency from public material
  2. Unquantified adjacencies: PC, voice-AI, battery/health are strategic pillars but contribute immaterial current revenue; timelines and TAM highly speculative (especially voice-PC sampling only as of call)
  3. Insider selling context unclear: EVP's June liquidation could signal profit-taking after strong run, or routine diversification; 10b5-1 plans are pre-set, so timing not necessarily strategic signal
  4. Analyst targets speculative: Retail "$200 by EOY 2026" claims lack detailed modeling; published price targets in news sample are generic ("can the run continue?")
  5. Seasonality shift not fully modeled: Management flagged camera content pull-forward but acknowledged uncertainty in forward seasonality; guidance only one quarter out
  6. Supply chain risks material but vague: Foundry capacity, GF wafer commitment, geopolitical disruption flagged in 10-K but not quantified; no management Q&A on mitigation
  7. Tax rate normalization risk: Q2 benefited from retroactive R&D expensing; forward rate guidance 16–18% still favorable but less predictable
  8. Valuation debate: GuruFocus overvalued tag vs. retail bullish sentiment; no clear consensus on fair value in material

Summary for Investment Council

Cirrus Logic delivered a strong Q2 FY2026 with record $561M revenue and healthy 52.5% gross margin, driven by smartphone demand and latest-generation audio/camera product adoption. Three-pillar strategy (smartphone audio dominance, mixed-signal expansion, non-phone markets) is progressing, with notable early wins in PC mainstream consumer segment and emerging AI/voice PC opportunity. Balance sheet is pristine ($896M cash, zero debt, strong FCF), and management's tone is disciplined and realistic about adjacency timelines.

However, material uncertainties temper upside: Largest customer (widely presumed Apple) remains opaque; non-smartphone revenue is nascent and unquantified; supply chain risks (foundry capacity, GF commitments, geopolitics) are real but unquantified. Retail sentiment is bullish but often shallow; insider selling in June offers mixed signals. Valuation debate is live (GuruFocus overvalued tag vs. bull case "easy $200").

Recommend: Deep-dive on customer concentration (if able to access management confidentially), quantified PC/voice/battery roadmap timelines, and supply-chain mitigation plans before committing material capital. Stock may deserve premium for longevity and margin quality, but near-term upside depends heavily on Apple cycle and PC ramp credibility.

Bull case

CRUS is a genuinely elegant fabless franchise: 31.9% FCF margin, 17.3% ROIC vs ~10.3% WACC (700bps spread), zero debt, $778M net cash, and only $14M capex on $2B revenue. Operating cash flow ($651M) exceeds net income by 57% — the forensic short-seller gives it a clean bill of health, and Greenblatt (78) sees a genuine Magic Formula buy at 11.6x P/FCF. Mauboussin (74) and the disruption referee (72) argue the market anchors on the misleading 1.7% revenue CAGR while ignoring the far steeper FCF trajectory and real optionality in AI PC/voice adjacencies where Cirrus is already sampling product. Base-case DCF of $177.83 implies ~21% upside; buybacks and a pristine balance sheet backstop the story.

Bear case

The single most-cited fact across all lenses is Apple concentration (presumed >80%), which management refuses to quantify — a moat that belongs partly to the customer, with Apple's demonstrated pattern of in-sourcing silicon (modems, displays) as an existential tail risk (Munger, Greenwald, Fisher, Smith). Revenue CAGR is only 1.7% and PEG is 10.4x — this is a mature, cyclical cash cow, not a compounder. Greenwald's EPV work is the sharpest dissent: normalized earnings power yields just $89-110/share, meaning the $146.50 price is a 33-65% premium to earnings power and pays for unproven growth. The valuation referee shows a mild stress test (WACC +100bps, normalized $500M FCF) drops intrinsic value below the current price, erasing the margin of safety. 69% of DCF value sits in terminal value, and FY2025's 31.9% FCF margin is a 5-year peak that may not be sustainable.

Dissent — where the council disagrees

The council is sharply split. Schloss (AVOID 28) is the harshest — 3.5x book, at 52-week high, no asset floor for a deep-value buyer. Greenwald (WATCH 58) is the most analytically damning bull-killer: EPV of $89-110 says the market is paying a large premium to earnings power, not investing on a margin of safety. Against them, Greenblatt (PASS 78), Mauboussin (74), and both referees (72) argue the FCF quality and ROIC-WACC spread justify a buy and that the PEG/revenue-CAGR bears are anchoring on the wrong numerator. The tension that resolves to Watch: the growth, value-purist, and risk lenses all cluster at 52-62 flagging concentration and thin margin of safety, while the quality/formula bulls concede they are paying a fair-not-gift price. When the valuation referee's own stress test erases the margin of safety and EPV sits far below price, a clean Pass is unwarranted.

Key risks

  • Apple concentration (~80%+, undisclosed) — a single design-loss or in-sourcing decision collapses the revenue base
  • FY2025 31.9% FCF margin is a 5-year peak; normalization to ~22-25% cuts intrinsic value below current price
  • 69% of DCF value in terminal value; highly sensitive to WACC/terminal-growth in a higher-for-longer regime
  • GlobalFoundries wafer commitment through 2026 flagged as potentially onerous — write-down/take-or-pay risk
  • Non-smartphone adjacencies (PC, voice-AI, battery) are pre-revenue; growth story unproven vs. a 1.7% CAGR reality
  • Fabless model excludes CHIPS Act support; foundry capacity competition from AI demand

Catalysts

  • Mainstream consumer PC design win generating FY2027 revenue
  • AI/voice ultra-low-power PC product converting from sampling to material revenue
  • Continued buybacks reducing share count
  • Apple content-per-device gains in next iPhone cycle
  • Any disclosure improving visibility on customer concentration

DCF valuation (finance-expert model)

two-stage DCF, Gordon terminal value, CAPM-weighted WACC.

Intrinsic value: $177.83/share vs price $146.5 → +21% (bear $159.36 · base $177.83 · bull $218.46).

Step Value
Base free cash flow $637M
FCF growth (yrs 1-5) 2.0% (revenue CAGR)
WACC (β 1.195) 10.3%
Terminal growth 2.5%
PV of explicit FCF $2.5B
PV of terminal (residual) value $5.7B (69% of EV)
Enterprise value $8.2B
less Net debt $-778M
= Equity value $9.0B
/ Shares (51M) = intrinsic/share $177.83

Short-sell evaluation

🚫 AVOID SHORTING

Despite genuine overvaluation-vs-EPV concerns and a stagnant top line, CRUS is a poor short. The forensic short-seller (72) explicitly finds no accounting red flags — cash flow exceeds earnings by 57%, the balance sheet is fortress-class with zero debt and $778M net cash, and there is no distress, no debt wall, and no accrual manipulation to catalyze a decline. The valuation is fair-to-full, not egregiously bubbly (11.6x P/FCF, 17.8x P/E), so there is no obvious de-rating trigger. The bear case rests on Apple concentration and narrative valuation — real risks, but unverifiable and lacking a near-term catalyst. Shorting a debt-free, FCF-gushing, buyback-active quality name near fair value against unlimited downside is exactly the trade that gets squeezed.

Pros (the short could work)

  • Price sits 33-65% above Greenwald's EPV of $89-110 — paying for unproven growth
  • Revenue CAGR of 1.7% and PEG of 10.4x while adjacencies remain immaterial
  • FY2025 FCF margin at a 5-year peak; reversion to ~22% cuts fair value below price
  • 69% terminal-value dependence makes valuation fragile to any WACC or demand shock
  • Apple in-sourcing (base-rate behavior) would be a revenue-cliff catalyst
  • GlobalFoundries wafer commitment write-down risk if demand softens

Cons (what kills the short)

  • Clean forensic profile: OCF exceeds net income by 57%, no accrual or accounting red flags
  • Fortress balance sheet — zero debt, $778M net cash, 7.65x current ratio; no distress catalyst
  • Genuinely cheap on cash flow (11.6x P/FCF) — limited de-rating room, valuation is not a bubble
  • Active buybacks and rising FCF provide support and squeeze fuel
  • Durable switching-cost moat with 5-6 year custom-silicon lifecycles
  • AI PC/voice is a demand tailwind, not a disruption threat, per the disruption referee
  • Unlimited downside on a short against a quality name with positive momentum optionality

Council scorecard

Lens School Stance Score Conf
Joel Greenblatt value 🟢 pass 78 high
Michael Mauboussin quality 🟢 pass 74 medium
Forensic Short-Seller (Chanos/Einhorn-style) referee 🟢 pass 72 medium
AI & Disruption Referee (Christensen-style) referee 🟢 pass 72 medium
Chuck Akre quality 🟡 watch 62 medium
Warren Buffett quality 🟡 watch 62 medium
Valuation Referee (Damodaran-style) referee 🟡 watch 62 medium
Charlie Munger quality 🟡 watch 62 medium
Terry Smith (Fundsmith) quality 🟡 watch 62 medium
Philip Fisher growth 🟡 watch 58 medium
Bruce Greenwald value 🟡 watch 58 medium
Ray Dalio risk 🟡 watch 52 medium
Stanley Druckenmiller risk 🟡 watch 52 medium
Benjamin Graham value 🟡 watch 52 medium
Seth Klarman value 🟡 watch 52 medium
Peter Lynch growth 🟡 watch 52 medium
Howard Marks risk 🟡 watch 52 medium
Walter Schloss value 🔴 avoid 28 high

Member reasoning

Joel Greenblatt — 🟢 pass · 78/100 · high confidence

Cirrus Logic scores well on both legs of the Magic Formula. On ROIC: EBIT of $460M against a tangible capital base that is extraordinarily lean — net working capital (current assets $1,409M minus current liabilities $184M = $1,225M) plus net fixed assets (capex of only $14M implies minimal net PP&E as a fabless company; using total assets minus current assets minus intangibles, net fixed assets are likely sub-$200M) yields an EBIT/tangible capital ratio that is very high, consistent with the reported ROIC of 17.3% on a book basis. On earnings yield: EV = market cap $7,391M + zero debt - excess cash $778M ≈ $6,613M. EBIT yield = $460M / $6,613M ≈ 6.96%. For a semiconductor company with near-zero capex ($14M on $2B revenue), a 52%+ gross margin, and strong FCF conversion (FCF margin 31.9%), this earnings yield is reasonable but not screaming cheap — the DCF intrinsic value of $177.83 per share represents a 21% upside to the current $146.50 price, suggesting meaningful but not extreme margin of safety. The business earns genuinely high returns on the minimal tangible capital it employs (fabless model = working capital is the main capital employed), recurring EBIT converts cleanly to FCF ($637M FCF vs $460M EBIT, the gap reflecting low-cash-tax and minimal capex), and the capital structure is pristine (net cash of $778M). Revenue CAGR of 1.7% over 3 years is modest but FCF margin expansion from 5.5% to 31.9% over 4 years shows dramatically improving capital efficiency. The P/FCF of 11.6x is genuinely attractive for a business of this quality. No special-situation catalyst is present, which limits the upside multiple; and customer concentration in Apple remains an unquantifiable but real risk. Insider selling via 10b5-1 is a mild negative signal but pre-planned. Overall: a good-to-excellent business at a fair-to-moderately-cheap price — squarely in Greenblatt's buy zone.

Key points

  • EBIT/EV earnings yield ~6.96% ($460M EBIT / ~$6,613M EV with $778M cash netted) — respectable for a high-quality semi
  • Fabless model means tangible capital base is almost entirely working capital; ROIC on that lean base is very high, consistent with Magic Formula top-decile quality
  • FCF margin expanded from 5.5% (FY2021) to 31.9% (FY2025) — dramatic improvement in capital efficiency, not a one-time spike
  • P/FCF of 11.6x is cheap for a business generating 32% FCF margins with zero debt
  • DCF intrinsic value $177.83 vs $146.50 price = 21% upside; bear case $159.36 provides some floor support
  • Net cash $778M provides EV deflation and balance sheet safety; current ratio 7.6x
  • Operating margin 23%, net margin 20.7%, ROE 19.7% — all consistent with durable competitive advantage in custom silicon design

Red flags

  • Customer concentration: Apple dependency is unquantified but dominant — a single OEM cycle miss collapses the EBIT base
  • Revenue CAGR only 1.7% over 3 years — the business is not growing the top line meaningfully, relying on margin expansion for earnings growth
  • EBIT/EV yield at ~7% is attractive but not a screaming bargain; stock trades near 52-week high, limiting margin of safety
  • Non-smartphone adjacencies (PC, voice-AI, battery health) are immaterial current revenue — growth optionality is real but unproven and not in the Magic Formula calculus
  • Fabless model excludes CHIPS Act subsidies and creates foundry concentration (GlobalFoundries wafer commitment through 2026) — supply disruption could impair near-term EBIT
  • PEG of 10.37 signals the market is pricing in growth that hasn't materialized yet in the top line; Greenblatt cares about current EBIT yield, but elevated PEG warns against paying for future growth that may not arrive

Michael Mauboussin — 🟢 pass · 74/100 · medium confidence

Cirrus Logic presents a compelling ROIC-above-WACC story with measurable moat sources, though the durability of the spread and the expectations embedded in the current price require careful decomposition. ROIC of 17.3% against an estimated WACC of ~10.3% yields a spread of roughly 700 bps — not enormous by franchise standards, but real and positive. The moat mechanism is identifiable and specific: deep switching costs embedded in custom silicon co-development with a dominant OEM (Apple), proprietary codec/amplifier IP that takes years to qualify, and scale economics in R&D amortization across a 5–6 year product lifecycle. These are not adjective-moats — they are structural barriers. The current price of $146.50 implies roughly 21% upside to the DCF base case of $177.83, suggesting the market is pricing in modest 2% FCF growth (the revenue CAGR used as the growth assumption) — which is arguably conservative given the FCF margin expansion from 5.5% (FY2021) to 31.9% (FY2025). This is a key insight from expectations investing: the embedded expectations appear beatable if the PC and mixed-signal adjacencies deliver even modest traction. However, the base rate for semiconductor companies this deeply concentrated in one customer cycle is mean-reversion pressure on ROIC, and the unproven adjacencies (PC, voice-AI, battery health) are not yet in the numbers. The distribution of outcomes skews modestly positive from here, but the fat downside tail — Apple concentration unwinding, or AI-PC failing to materialize — is real and unquantified. Management's capital allocation is disciplined (buybacks, zero debt, $778M cash, capex of only $14M reflecting fabless efficiency), and the 10b5-1 insider selling is routine rather than alarming at this scale. The PEG of 10.37 is elevated, but this metric is misleading here since the 1.7% revenue CAGR massively understates FCF trajectory — the correct growth input should be FCF CAGR, which is dramatically higher. On balance, the embedded expectations are achievable but not comfortably discounted, and the Apple opacity is a genuine epistemic handicap for any moat-durability assessment.

Key points

  • ROIC of 17.3% vs WACC of ~10.3% — 700 bps positive spread, confirmed value creation
  • Moat mechanism is specific and measurable: deep switching costs in custom silicon (5–6 year qualified lifecycle), proprietary codec/amplifier IP, R&D amortization scale economics across Apple's production volumes
  • FCF margin expansion from 5.5% (FY2021) to 31.9% (FY2025) is exceptional — not a mean-reverting accident but a structural margin improvement as custom silicon matures and capex remains minimal ($14M on $2B revenue)
  • Expectations investing read: DCF uses 2% FCF growth (equal to revenue CAGR) but FCF has compounded at ~60% CAGR over 4 years — market may be dramatically underestimating the FCF trajectory if margins stabilize at current levels
  • Current P/FCF of 11.6x is strikingly low for a fabless semicap with 31.9% FCF margin, ROIC of 17%, and zero debt — base case intrinsic value of $177.83 implies the stock is priced at an ~18% discount to DCF fair value
  • Capital allocation is disciplined: fabless model (capex $14M), zero debt, $778M cash, active buybacks — management reinvests at high ROIC and returns excess capital rather than pursuing empire-building M&A
  • Three-pillar adjacency strategy (PC, voice-AI, battery health) represents genuine optionality not yet in the numbers — PC mainstream consumer win expected FY2027 revenue; voice-enabled PC sampling December quarter

Red flags

  • Apple customer concentration is the dominant risk and is deliberately opaque — management refuses to quantify, making moat-durability assessment partially blind; losing or losing share at Apple would crater ROIC rapidly
  • Adjacencies (PC, voice-AI, battery health) are pre-revenue and unproven — outside view base rate for fabless semis successfully diversifying from a dominant OEM anchor is poor; these should be weighted as options, not base case
  • Revenue CAGR of only 1.7% over 3 years despite strong FCF improvement — top-line growth is effectively flat; FCF margin improvement cannot continue indefinitely from efficiency gains alone if revenue stagnates
  • DCF terminal value represents 69.2% of enterprise value — highly sensitive to terminal growth and WACC assumptions; a 50 bps WACC increase or terminal growth reduction would meaningfully compress intrinsic value
  • Foundry concentration (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints from AI demand) is an unquantified supply-chain tail risk that could disrupt the 5–6 year product longevity advantage
  • PEG of 10.37 screams expensive if one takes revenue CAGR at face value — requires understanding that FCF is the correct growth numerator; risk that analysts anchoring to revenue growth dismiss the stock before FCF story is understood
  • Insider selling (EVP liquidating ~$650K+ in June 2026) is pre-planned but concentrated — small sample signal but directionally worth monitoring against the conviction narrative

Forensic Short-Seller (Chanos/Einhorn-style) — 🟢 pass · 72/100 · medium confidence

Cirrus Logic passes the forensic short-seller's core accounting tests with unusually clean results. The earnings-vs-cash test — the Chanos/Einhorn lead indicator — goes strongly against a short thesis: operating cash flow of $650.6M materially EXCEEDS net income of $414.4M in FY2025, and FCF of $636.6M exceeds net income by 54%. This is the opposite of the classic accrual red flag. Historical pattern holds: in FY2024, FCF ($421.6M) exceeded net income ($331.5M); in FY2023, FCF ($384M) vs net income ($275M); FCF margins have expanded from 5.5% in FY2021 to 31.9% in FY2025. Cash genuinely converts from earnings — no earnings quality crisis. Balance sheet is fortress-class: zero debt, $778M cash, current ratio 7.65x, net cash position of ~$778M offsets any valuation concern. Working capital signals are clean: no evidence of receivables or inventory outrunning revenue (DSO data not granularly available in the fact base, but flagging absence). Capex is minimal ($14M on $2B revenue), consistent with fabless model — no aggressive capitalization of costs that flatter margins. Share buybacks appear genuine: shares outstanding at ~50.6M; the fact base does not show share count expansion, suggesting buybacks are at least partially reducing count rather than just treading dilution water, though SBC quantum is not fully disclosed. Insider selling (EVP Baumgartner, ~5,000 shares in June under 10b5-1) is modest relative to float and pre-planned — not a cluster red flag. No auditor changes, restatements, or going-concern language visible. The genuine forensic concerns are: (1) customer concentration opacity — management refuses to discuss largest customer (Apple), creating a material disclosure gap that Chanos would flag as governance/transparency risk; (2) the revenue CAGR of 1.7% over 3 years barely justifies the growth story embedded in the valuation, while new adjacency revenues (PC, voice-AI, battery) are immaterial and unquantified — this is a 'story stock' risk without a falsifiable catalyst yet; (3) GlobalFoundries wafer commitment flagged in 10-K as potentially onerous through 2026 — a write-down risk if demand softens; (4) the DCF relies on 2% FCF growth (matching tepid revenue CAGR) but 69% of value sits in terminal value, creating sensitivity to any demand disruption; (5) a PEG of 10.37 signals the market is pricing growth that the historical record does not yet support. The kill question for a short thesis: a meaningful Apple design loss or smartphone content reduction (e.g., Apple insourcing audio chips) would collapse the revenue base. No evidence this is imminent, but the opacity makes it unverifiable. Overall: forensic accounting gives CRUS a clean bill of health — this is not a fraudulent or accrual-inflated business. The bear case rests on concentration/disruption risk and narrative valuation, not accounting manipulation, which is outside the core forensic toolkit.

Key points

  • Operating cash flow ($650.6M) exceeds net income ($414.4M) by 57% in FY2025 — the opposite of the short-seller's core accrual red flag; pattern holds across all 5 years of history
  • FCF margin expanded from 5.5% (FY2021) to 31.9% (FY2025) — genuine cash conversion, not earnings engineering
  • Zero debt, $778M net cash, 7.65x current ratio — no debt wall or refinancing dependence; the business cannot be structurally distressed
  • Capex of $14M on $2B revenue (0.7%) is consistent with fabless model; no evidence of aggressive cost capitalization masking real expense
  • Insider selling (EVP, ~5,000 shares June 2026) is under pre-planned 10b5-1, modest relative to float — not a cluster warning signal
  • No auditor changes, restatements, or going-concern language in available filings
  • Revenue CAGR of 1.7% over 3 years is well below what a 10.37x PEG implies — valuation rests on adjacency growth that is currently immaterial and unquantified

Red flags

  • Customer concentration opacity: management refuses to discuss largest customer (widely presumed Apple) per stated policy — a material disclosure gap; an Apple insourcing or design-loss event is unverifiable and would be a revenue cliff
  • GlobalFoundries wafer commitment through 2026 flagged as potentially onerous in 10-K — a write-down or take-or-pay liability risk if smartphone demand softens
  • 69% of DCF value sits in terminal value with only 2% FCF growth assumed — extreme sensitivity to any demand disruption or multiple compression
  • PC/voice-AI/battery adjacencies are 'story pillars' with no quantified current revenue contribution; management acknowledged 'early-stage ramp' and 'only sampling' — classic unproven TAM risk
  • DSO and DIO trend data not available in fact base — flagging data gap; inability to fully run receivables/inventory quality test
  • SBC quantum not separately broken out in the fact base — cannot fully test whether buybacks genuinely reduce share count or merely offset dilution treadmill

AI & Disruption Referee (Christensen-style) — 🟢 pass · 72/100 · medium confidence

Cirrus Logic is a fabless semiconductor company whose core value is custom-designed mixed-signal silicon — audio codecs, boosted amplifiers, haptic drivers, camera controllers — embedded directly into physical devices. The 'job' it does is not knowledge work or intermediation: it converts analog signals to digital and vice versa, manages power delivery, and provides ultra-low-power voice wake functionality at the hardware layer. AI cannot replace this job by software substitution; physics requires the silicon to exist between the real world and the digital processor. The disruption question for CRUS is therefore not 'can AI do this cheaper via software' but rather (1) does AI commoditize the chip design process making CRUS's IP less defensible, (2) do hyperscalers vertically integrate their own audio/mixed-signal silicon, (3) does AI-driven demand (AI PCs, on-device inference) expand or contract Cirrus's addressable market, and (4) does AI accelerate or decelerate Apple's hardware cycle. On balance the disruption scorecard is net-positive-to-neutral. The AI PC and on-device voice processing wave is structurally additive — Cirrus explicitly called out an ultra-low-power AI/voice wake product sampling in Q4 FY2026, directly leveraging its smartphone voice IP for PC; this is a genuine tailwind as AI PCs require always-listening, low-power audio front-ends precisely in Cirrus's wheelhouse. On-device AI inference on smartphones also increases the premium on efficient mixed-signal front-ends (microphones, codecs, camera analog front-ends) because data must be captured cleanly before neural processing. The moat-commoditization question is more nuanced: EDA tools and AI-assisted chip design (Synopsys DSO.ai, Cadence Cerebrus, internal hyperscaler flows) do lower barriers to analog/mixed-signal design at the margin, but custom silicon for a tier-1 OEM requires deep co-design, years of silicon iterations, and process-node-specific tuning that is not yet automatable end-to-end. Cirrus's 5-6 year product lifecycle with Apple-class customers reflects sticky, co-engineered relationships that are not easily replicated by a model plus commodity IP. The vertical integration risk (Apple designing its own audio codec) is a real and ongoing concern — Apple already designs its own application processors — but audio mixed-signal remains a specialized domain where Apple has historically chosen to source externally; there is no public evidence of Apple in-housing codec/amp design as of FY2026. The hyperscaler capture risk is low: Google, Microsoft, and Amazon do not have a meaningful path to supplying physical audio silicon to Apple or Android OEMs. The AI-chip-design commoditization risk is real on a 5-10 year horizon but not yet a near-term displacement mechanism — Cirrus's defensibility rests on OEM co-design relationships and process-specific analog tuning, not on being the only party that can run SPICE simulations. Management's treatment of AI is mixed: they frame AI PCs and voice features as tailwinds (credible and specific — sampling product disclosed), but the 10-K risk factors do flag AI-driven foundry capacity constraints (a second-order supply risk, not a demand-side displacement, but real). No explicit engagement with whether AI chip design tools erode their R&D moat — a gap, though not an immediate red flag. The falsifiable call: over the next 2-4 years, evidence of displacement would be Apple announcing in-house audio codec production, Cirrus losing a design cycle to an AI-native mixed-signal startup, or take-rate compression on custom silicon contracts. Evidence of reinforcement would be AI PC voice products achieving material revenue, camera controller content per device expanding, and gross margin holding above 51% as AI-driven device complexity increases. Current data — 52.5% gross margin, FCF margin rising to 32%, no disclosed competitive displacement — is consistent with the reinforcement scenario. Score 72: genuine hardware moat that AI makes somewhat more valuable on the demand side (AI PC, on-device voice), partially offset by longer-term analog design commoditization risk and Apple insourcing optionality.

Key points

  • Core product is physical mixed-signal silicon — AI cannot substitute software for hardware that sits at the analog-digital boundary; the 'obsolescence via AI software' mechanism does not apply here
  • AI PC and on-device voice inference is a structural tailwind: Cirrus is sampling an ultra-low-power AI/voice wake product for PCs leveraging its smartphone codec IP, directly monetizing the AI device wave
  • On-device AI inference on smartphones increases demand for clean, efficient analog front-ends (microphones, codecs, camera AFEs) — AI complexity at the processor layer raises, not lowers, the value of high-fidelity signal capture
  • 5-6 year custom silicon lifecycle with Apple-class OEMs reflects deep co-design relationships; these are not replicable by a foundation model plus commodity IP on any near-term horizon
  • EDA/AI-assisted chip design tools (Synopsys DSO.ai, Cadence) are a margin risk on R&D efficiency over 5-10 years but do not yet commoditize analog mixed-signal co-design with tier-1 OEMs
  • Hyperscaler capture risk is structurally low — Google/Microsoft/Amazon have no supply-chain path to displace Cirrus in Apple or Android OEM silicon
  • Gross margin expansion to 52.5% and FCF margin at 32% show no current pricing pressure — leading indicators of commoditization are absent

Red flags

  • Apple vertical integration optionality is the single largest AI-adjacent disruption risk: Apple's demonstrated willingness to in-house silicon (APs, modems, now potentially more) combined with AI-assisted analog design tools could shorten the path to in-house audio codec over a 7-10 year horizon
  • AI-assisted EDA tools are lowering analog design barriers at the margin; if this accelerates, new entrants (or Apple itself) could challenge Cirrus's design-cycle advantage sooner than historical precedent suggests
  • 10-K flags AI-driven foundry capacity constraints (TSMC, GlobalFoundries) as a supply-side risk — AI semiconductor demand is competing for the same process nodes Cirrus uses, a second-order cost pressure
  • Management does not explicitly address the risk that AI chip design tools could erode their R&D moat — absence of honest engagement with this vector is a mild credibility gap
  • Customer concentration (widely presumed Apple) means a single AI-strategy decision by one OEM could impair >50% of revenue; this concentration amplifies any disruption scenario disproportionately

Chuck Akre — 🟡 watch · 62/100 · medium confidence

Cirrus Logic passes two of Akre's three legs convincingly but stumbles badly on the reinvestment runway — the most important leg. The business quality is genuinely impressive: 31.9% FCF margins, 17.3% ROIC, 19.7% ROE, all achieved with zero debt and minimal capex ($14M against $637M FCF). The capital-light, high-IP fabless model with deep switching costs at Apple is a real franchise. Management appears disciplined — clean balance sheet, active buybacks (shares outstanding declining), candid tone on adjacency timelines. But the reinvestment problem is acute: revenue CAGR of only 1.7% over three years, an overwhelming single-customer dependency (Apple), and nascent non-smartphone markets that are years from material contribution. The company is generating $637M in annual FCF from a $7.4B market cap business but cannot clearly demonstrate where it redeploys that cash at similarly high rates. The 'compounding machine' test requires not just generating cash but reinvesting it at high rates for years — CRUS is currently accumulating cash ($778M) and returning it via buybacks, which is rational but signals limited organic reinvestment opportunity at scale. The DCF intrinsic value of $177.83 vs. $146.50 current price offers ~21% upside in the base case, making valuation not egregious (P/FCF of 11.6x, P/E of 17.8x are reasonable for the quality). However, the 2% FCF growth assumption embedded in the DCF captures the reinvestment problem well — this is not a high-growth compounder, it's a high-quality cash cow. For Akre's framework, a 'pass' requires all three legs; the reinvestment runway leg is clearly impaired by customer concentration and unproven adjacencies.

Key points

  • FCF margin of 31.9% and ROIC of 17.3% achieved with zero debt and $14M capex — genuinely capital-light franchise economics
  • ROE of 19.7% driven by business economics, not leverage — current ratio 7.6x, debt-to-equity effectively zero, net cash $778M
  • FCF per share compounding is real: FCF grew from $98.6M (2021) to $636.6M (2025), a massive improvement, though partly cyclical
  • Management communication is measured and honest — they explicitly acknowledged unproven adjacencies and refused to overpromise on PC/battery timelines in Q2 call
  • P/FCF of 11.6x and P/E of 17.8x are reasonable for the quality level; DCF base case $177.83 implies ~21% upside — not a valuation disaster
  • Active buyback program with $414M remaining authorization; insider selling is pre-planned 10b5-1, modest in scale relative to market cap
  • PC market early wins (mainstream consumer design win FY2027 revenue expected) and AI/voice PC product sampling represent potential reinvestment avenue
  • Three-year revenue CAGR of only 1.7% demonstrates the fundamental growth constraint

Red flags

  • Reinvestment runway severely constrained by single-customer dependency — Apple widely understood to be 80%+ of revenue, limiting organic growth ceiling to Apple's own unit cycle
  • Revenue growth of 1.7% CAGR over three years is far below what Akre's compounding framework demands; this is a cash cow, not a compounder
  • Non-smartphone adjacencies (PC, battery/health, voice-AI) are nascent — mainstream PC revenue only expected FY2027, battery IP has 'no design wins to announce,' voice-PC only sampling; total non-smartphone revenue contribution is immaterial and timeline uncertain
  • $778M cash accumulation signals limited high-ROIC reinvestment options internally — cash piling up is the symptom of a mature franchise without adequate runway
  • Foundry concentration risk (GlobalFoundries wafer commitment through 2026, TSMC AI-driven capacity tightening) threatens supply-chain reliability for a company that can't fab its own chips
  • Insider selling by EVP — while pre-planned, multiple transactions in June 2026 (~$850K+ total) at near 52-week highs is a mild concern
  • PEG ratio of 10.37 reflects the mismatch between earnings quality and growth rate — expensive on a growth-adjusted basis

Warren Buffett — 🟡 watch · 62/100 · medium confidence

Cirrus Logic is an interesting semiconductor business that sits at the edge of my circle of competence — it is profitable, has a long operating history, generates genuine free cash flow, and possesses elements of a moat through deep customer integration and custom silicon design. However, several features give me pause. The business is heavily reliant on what appears to be a single dominant customer (Apple), which is both its greatest strength and its most significant vulnerability. The moat is real but narrow and customer-dependent rather than broadly durable in the way I prefer. On the positive side: ROE of ~19.7%, ROIC of ~17.3%, FCF margin expanding to ~31.9%, zero debt, current ratio of 7.6, and owner earnings that are genuinely clean (capex is only $14M against $650M operating cash flow — this is not a capital-intensive business, which I greatly admire). The P/E of 17.8x and price-to-FCF of 11.6x are modest for a business of this quality. The DCF suggests ~21% upside to intrinsic value of $177.83, which provides a thin but real margin of safety. Revenue CAGR of only 1.7% over three years tempers enthusiasm — this is a slow-growth business at current scale, not a compounder expanding its addressable market rapidly. The PC and AI/voice adjacencies are early-stage and unproven, which I discount heavily. Insider selling under 10b5-1 plans is noted but not alarming. What prevents a higher score: customer concentration opacity (I cannot assess the moat without understanding Apple dependency quantitatively), modest revenue growth suggesting limited pricing power expansion, and the semiconductor industry's inherent cyclicality which runs counter to my preference for businesses I can forecast a decade out with high confidence.

Key points

  • Owner earnings are genuinely excellent: capex of only $14M vs. $650M operating cash flow — an asset-light fabless model that converts earnings to cash at ~31.9% FCF margin
  • ROE of ~19.7% and ROIC of ~17.3% achieved with ZERO debt and $778M net cash — exactly the kind of high-quality economics I look for
  • Clean balance sheet: current ratio 7.6x, no debt-to-equity ratio (debt is effectively zero), $778M cash provides fortress-like financial security
  • P/E of 17.8x and price-to-FCF of 11.6x are not expensive for a business of this quality — the DCF base case of $177.83 offers ~21% margin of safety at current $146.50 price
  • Custom silicon with 5–6 year product lifecycles creates real switching costs and R&D leverage — this is a genuine moat element, though narrow
  • Honest, disciplined management tone noted in earnings call; active buyback program with $414M remaining authorization
  • No reliance on turnarounds or accounting gimmicks; five years of consistent profitability and improving FCF margins (5.5% in 2021 to 31.9% in 2025)

Red flags

  • Extreme customer concentration — largest customer (almost certainly Apple) is not disclosed but likely represents the vast majority of revenue; this is the antithesis of the durable, diversified moat I prefer — the castle has only one drawbridge
  • Revenue CAGR of only 1.7% over three years signals that the core business is not compounding; a wonderful business should be able to reinvest retained earnings at high rates into growing markets
  • Semiconductor industry cyclicality makes decade-out earnings forecasting difficult — this is near the edge of my circle of competence for predictability
  • PC, AI/voice, and battery adjacencies are early-stage with immaterial revenue — I do not pay for stories, only for proven, repeatable earnings streams
  • Insider selling (EVP sold ~5,000+ shares in June 2026) and foundry concentration risks (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints) are concerning
  • DCF is highly sensitive to terminal value (69.2% of equity value comes from terminal period) — if growth stagnates at 2%, that terminal assumption becomes the entire bet
  • GuruFocus flags stock as overvalued on its own methodology, which warrants attention even if I weigh the DCF more heavily

Valuation Referee (Damodaran-style) — 🟡 watch · 62/100 · medium confidence

The DCF model provided yields an intrinsic value of $177.83/share vs. the current price of $146.50, implying a 21% upside and a margin of safety that, on the surface, looks attractive. However, the DCF's core assumption — 2% FCF growth using the 3-year revenue CAGR of 1.72% — deserves serious scrutiny. The base FCF of $636M in FY2025 is unusually high (31.9% FCF margin vs. a 3-year average closer to 22–25%), partly reflecting capex of only $14M on $2B revenue (a fabless model with very low reinvestment). This creates an artificially inflated base from which to project. If I normalize FCF to, say, a 25% margin on current revenue, the base FCF would be ~$499M, and the intrinsic value per share drops materially — potentially to $130–140 range, near or below current price. The WACC of 10.27% is defensible for a semiconductor name with beta ~1.19 and no debt, though arguably slightly low given customer concentration risk and supply-chain exposure. The terminal growth rate of 2.5% is slightly above my preferred ceiling (equal to long-run nominal GDP) but not egregious. The bigger concern is the reverse-engineering exercise: at $146.50, the market is implying continued strong FCF generation with modest growth — achievable IF FY2025 FCF margins are sustainable. But ROIC of 17.3% comfortably exceeds the WACC of 10.27%, meaning growth genuinely creates value. Reinvestment intensity is minimal (capex $14M, sales-to-capital ratio extremely high for a fabless model), which is real for this asset-light structure. The story-to-numbers translation is coherent: dominant position in smartphone audio (Apple dependency), emerging PC/AI voice optionality (unquantified but real), clean balance sheet ($778M cash, zero debt adds ~$15/share). The narrative is honest and aligned with financials. However, 69% of DCF value sits in the terminal value, which is high sensitivity territory. Under a mild stress test — WACC +100bps to 11.27%, terminal growth 2.0%, normalized FCF base of $500M — intrinsic value falls to roughly $120–135, which is BELOW the current price. This eliminates the margin of safety entirely in the bear case. Revenue growth of only 1.72% CAGR over 3 years for a company trading at 3.7x sales and 11.6x FCF demands that the market is pricing in adjacency optionality (PC, voice-AI) that has not yet materialized in numbers. That is speculative value, not DCF-supported value. Net verdict: the base case DCF shows a modest margin of safety, but the assumptions are optimistic enough that sensitivity analysis erases it. The investment is not obviously overpriced (no heroic growth required), but it is not clearly cheap enough to call a 'pass' with discipline. A 'watch' reflects: real FCF quality, ROIC > WACC, defensible WACC, but insufficient margin of safety under conservative normalization, 69% terminal value dependence, and unquantified adjacency optionality baked into sentiment but not yet numbers.

Key points

  • Base DCF yields $177.83 intrinsic value vs. $146.50 price — 21% stated upside, but built on FY2025 FCF of $636M (31.9% margin), which is the highest in the 5-year history and may not be sustainable
  • ROIC of 17.3% clearly exceeds WACC of 10.27%, confirming growth creates value — fabless model with $14M capex on $2B revenue is genuinely asset-light and capital-efficient
  • Revenue CAGR of 1.72% over 3 years used as growth proxy is conservative and realistic; the 2% FCF growth assumption in the DCF is consistent with history
  • Clean balance sheet: $778M cash, zero debt, adds ~$15/share to equity value; current ratio 7.65x confirms financial resilience
  • Terminal value = 69% of total equity value at 2.5% terminal growth — high sensitivity; a 100bps WACC increase or normalization of FCF margin erases the margin of safety
  • At 11.6x price-to-FCF and 17.8x P/E, multiples are not demanding IF FY2025 FCF margins are sustainable; if they revert to 3-year avg ~22%, FCF drops to ~$440M and the stock looks fairly to slightly overvalued
  • PC/voice-AI optionality is real but produces zero current revenue — it is a call option embedded in the price, not a DCF input
  • Insider selling (EVP June 2026, pre-planned 10b5-1) is a weak but noted signal; not alarming but not confirming peak conviction
  • GuruFocus 'overvalued' tag and DCF base-case 'buy' signal reflect the uncertainty range — both are defensible depending on FCF normalization assumption

Red flags

  • FY2025 FCF of $636M (31.9% margin) is an outlier vs. 5-year history (2021–2024 avg FCF margin ~19%); using this as the DCF base overstates normalized earnings power by potentially 25–35%
  • 69% of intrinsic value sits in terminal value — model is highly sensitive to WACC and terminal growth; a modest stress (WACC 11.3%, tg 2.0%, normalized FCF $500M) puts intrinsic value at or below current price
  • Customer concentration: Apple is widely presumed to be the dominant customer but management refuses to quantify; the inability to stress-test this exposure is a fundamental valuation uncertainty
  • 3-year revenue CAGR of 1.72% is barely above inflation — the story of smartphone audio dominance is a stable/mature business, not a growth narrative; the price-to-sales of 3.7x requires either margin expansion or adjacency growth
  • Supply chain risks (GlobalFoundries wafer commitment through 2026, AI-driven foundry capacity constraints, geopolitical logistics) are flagged in 10-K but completely unquantified — adds tail risk not captured in the WACC
  • Non-smartphone adjacencies (PC, voice-AI, battery) are still pre-revenue or immaterial — pricing in optionality at a 17.8x P/E on a 1.7% revenue-growth business requires a stretch of imagination not supported by current numbers

Charlie Munger — 🟡 watch · 62/100 · medium confidence

Cirrus Logic is a genuinely high-quality semiconductor business that sits comfortably within a definable circle of competence: it makes specialized mixed-signal chips (audio codecs, amplifiers, camera controllers) that are deeply embedded in premium smartphones. The moat is real but narrower and more concentrated than I prefer. ROIC of 17.3%, ROE of 19.7%, operating margins of 23%, and a FCF margin of nearly 32% on $637M of free cash flow all speak to a business that earns well above its cost of capital. The balance sheet is pristine — $778M net cash, zero debt, current ratio of 7.6x. Management appears disciplined: low capex ($14M on nearly $2B revenue — a fabless model that is capital-light and elegant), active buybacks, and candid acknowledgment of concentration risk. Owner earnings are real; FCF conversion is excellent. The DCF at $177/share intrinsic value vs. $146.50 current price suggests a 21% upside — not a screaming bargain but a fair price for a quality business. However, three issues prevent a full 'pass.' First, the customer concentration problem is severe and deliberately opaque — Apple is almost certainly >80% of revenue, and management policy prohibits discussion. This is not a moat; it is a dependency. If Apple decides to in-source audio chips (as it has done with modems, displays, and other components), Cirrus Logic's economics deteriorate rapidly. Second, the revenue CAGR of 1.7% over three years is inconsistent with a compounder. Growth depends heavily on Apple's iPhone cycle and content gains per device rather than organic market expansion. The adjacency strategy (PC, AI-voice, battery management) is promising but entirely unproven at commercial scale — 'sampling' and 'early wins' are not earnings. Third, the PEG ratio of 10.4x is a glaring warning: the market is pricing in far more growth than the historical record supports. At 17.8x earnings and 11.6x FCF the stock is reasonably priced for the quality, but the terminal value constitutes 69% of the DCF — heavily dependent on assumptions that Apple concentration risk could violate. The DCF's 2% FCF growth assumption may actually be generous given 1.7% revenue CAGR. I can understand this business. Management is not obviously stupid or dishonest. The economics are good. But the concentration in one customer who is also a sophisticated technology company with incentives to vertically integrate keeps me from paying up enthusiastically. Inversion test: how does this go wrong? Apple in-sources or shifts supplier, as they have done repeatedly. That is not a remote scenario — it is the base-rate behavior of Apple. That alone caps my enthusiasm.

Key points

  • ROIC 17.3%, ROE 19.7%, FCF margin 31.9% — genuinely high-quality economics well above cost of capital
  • Pristine balance sheet: $778M net cash, zero debt, 7.6x current ratio — survives any downturn
  • Fabless model: $14M capex on $2B revenue is capital-light elegance; owner earnings closely track GAAP earnings
  • FCF of $637M at 11.6x price-to-FCF represents reasonable value for the quality on offer
  • DCF intrinsic value $177.83 vs $146.50 price implies ~21% upside — fair price, not a gift
  • Management tone disciplined and candid; active buybacks ($414M authorization remaining); no empire-building
  • PC/AI-voice adjacency is logical extension of existing IP; first mainstream consumer laptop win is meaningful milestone
  • Simple, understandable business model: design specialized silicon for premium devices, collect royalty-like economics per device shipped

Red flags

  • Customer concentration: Apple widely understood to be >80% of revenue, management policy prohibits disclosure — single point of failure for the entire franchise
  • Apple has demonstrated repeated pattern of in-sourcing chips (modems, displays, neural engines) — audio codec is not immune to this risk
  • Revenue CAGR of only 1.7% over three years contradicts the 'compounder' thesis; growth is episodic, tied to iPhone cycles and content gains
  • PEG of 10.4x is alarming — market pricing in growth the historical record does not support
  • 69% of DCF value sits in terminal value — highly sensitive to assumptions that customer concentration risk could invalidate
  • PC/AI-voice/battery adjacencies are entirely unproven at commercial scale; 'sampling' and 'early wins' generate no material revenue today
  • Insider selling (EVP, multiple transactions June 2026) under 10b5-1 plans — modest signal but worth noting after a strong run
  • Foundry dependency (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints from AI demand) adds supply-chain fragility
  • CHIPS Act subsidy exclusion as fabless player is structural competitive disadvantage vs. integrated peers

Terry Smith (Fundsmith) — 🟡 watch · 62/100 · medium confidence

Cirrus Logic passes several Fundsmith quality screens but fails others, leaving a mixed picture. On the positive side: the business is genuinely asset-light (capex of only $14M against $637M FCF, a capex/revenue ratio of ~0.7%), cash conversion is excellent (FCF of $637M exceeded net income of $414M, with FCF margin of 32% vs. net margin of 21%), the balance sheet is pristine (zero debt, $778M net cash, current ratio 7.65x), and ROCE/ROIC of ~17-20% is respectable. Operating margins at 23% and gross margins implied around 52% (per Q2 call) are healthy for a semiconductor company. However, several quality concerns temper enthusiasm. First, ROCE at ~17-20% is good but not exceptional by Fundsmith standards — Smith typically demands sustained pre-tax ROCE of 20%+ across the cycle; Cirrus has delivered this inconsistently, with FCF margins ranging from 5.5% (2021) to 32% (2025), suggesting cyclicality rather than the steady compounding Smith prizes. Second, and most critically, customer concentration in Apple represents an existential moat concern from a quality-investing perspective — the moat belongs partly to the customer, not Cirrus; Apple could in-source or switch suppliers, and management cannot even discuss the relationship publicly. This is the antithesis of the recurring, diversified, brand-driven revenue Smith seeks. Third, revenue CAGR of just 1.7% over three years reveals that this is a slow-growth, cyclically lumpy business tied to smartphone upgrade cycles — not the resilient, predictably compounding demand profile Smith rewards. Fourth, non-smartphone adjacencies (PC, battery/health, voice-AI) are nascent and unquantified, so the growth runway is speculative. On valuation: P/FCF of 11.6x is genuinely attractive for a quality business, and the DCF intrinsic value of $178 implies ~21% upside at $146.50. The bear-case DCF of $159 still represents an upside, providing some margin of safety. However, the DCF uses a 2% FCF growth rate — anchored to a 3-year revenue CAGR of 1.7% — which may be appropriate given concentration risk and episodic smartphone cycles. The terminal value represents 69% of enterprise value, which is high and sensitive to assumptions. Net-net: excellent cash economics and balance sheet quality, but insufficient moat independence, too much customer concentration, and insufficient demand predictability to earn a full Fundsmith pass.

Key points

  • Exceptional FCF conversion: FCF of $637M vs. net income of $414M (FCF/NI ratio >1.5x), FCF margin 32% — cash conversion is outstanding
  • Truly asset-light: capex of only $14M (0.7% of revenue) against $637M FCF; business does not eat capital to grow
  • Clean balance sheet: zero debt, $778M cash, current ratio 7.65x — no leverage risk whatsoever
  • ROIC of 17.3% and ROE of 19.7% are solid but fall slightly below Smith's preferred 20%+ threshold and have been cyclical
  • P/FCF of 11.6x is a genuinely fair-to-cheap valuation for a quality industrial/semi business; DCF suggests 21% upside to intrinsic value of $178
  • Gross margins ~52% (per Q2 2026 call) and operating margins 23% indicate meaningful pricing power within its design wins
  • Revenue CAGR of only 1.7% over 3 years is below what Smith seeks; this is a mature, cyclical smartphone-driven business, not a compounder
  • PC/voice-AI adjacencies provide a theoretical growth runway but contribute immaterial current revenue; commercialization timelines are speculative

Red flags

  • Apple customer concentration: widely understood largest customer represents the dominant share of ~$2B revenue; moat resides partly with customer, not Cirrus — antithetical to Fundsmith's requirement for independently defensible competitive positions
  • Revenue cyclicality: FCF margin swung from 5.5% (FY2021) to 32% (FY2025); this is episodic smartphone-cycle performance, not the steady compounding Smith prizes
  • Demand not recurring/repeat-purchase in the consumer-staples sense — tied to handset upgrade cycles and OEM design cycles with multi-year gaps
  • Insider selling: EVP liquidating shares (5,077 shares, ~$850K+) in June 2026 under 10b5-1 plan is not alarming per se but is a mild negative signal at current prices near 52-week highs
  • 69% of DCF value sits in terminal value, making the intrinsic value estimate highly sensitive to long-term growth assumptions — limited true margin of safety if Apple concentration or smartphone saturation worsens
  • Foundry dependency on GlobalFoundries (wafer commitment through 2026) and TSMC adds supply-chain risk that a quality investor must price in

Philip Fisher — 🟡 watch · 58/100 · medium confidence

Cirrus Logic presents a mixed picture through the Fisher lens. On the positive side, the company exhibits strong margin quality (52.5% gross margin, 23% operating margin, 32% FCF margin), a disciplined R&D-driven product pipeline spanning audio codecs, camera controllers, and emerging PC/AI-voice products, and management that communicates candidly with measured tone — acknowledging Android softness and adjacency immaturity without promotional spin. The 5–6 year longevity of custom silicon demonstrates genuine technical differentiation and R&D productivity. However, the most critical Fisher criterion — sustained, above-average organic sales growth with a visible multi-year runway — is conspicuously weak. Revenue CAGR of only 1.72% over three years is far below what Fisher demands: this is barely inflation-level growth for a semiconductor company. The core smartphone business (dominated by one undisclosed customer, almost certainly Apple) is mature and cyclical, not expanding. Non-smartphone adjacencies (PC, AI-voice, battery health) are exciting strategically but contribute immaterial current revenue and carry unquantified timelines. The diversification narrative is real but early-stage — Fisher would want to see these new markets actually convert to revenue scale before awarding a growth premium. Customer concentration in one OEM also creates episodic rather than compounding growth dynamics. R&D spend appears productive (new codec nodes, camera controllers, SoundWire adoption) but the pipeline has not yet demonstrated it can move the revenue needle beyond the Apple dependency. Management depth and communication quality are solid positives. Overall, CRUS sits in Fisher's 'interesting but not yet ready' zone — a franchise with genuine R&D capability and margin quality, but lacking the growth velocity and market expansion evidence that Fisher requires before concentrating.

Key points

  • Revenue CAGR only 1.72% over 3 years — far below Fisher's above-industry-growth threshold; smartphone volume growth is episodic, not compounding
  • Gross margin ~52.5%, operating margin ~23%, FCF margin ~32% — margin quality is excellent and well-defended through custom silicon and mix management
  • R&D pipeline is genuine: 22nm smart codec, boosted amplifiers, camera controllers, SoundWire PC products, AI-voice sampling — products convert to multi-year socket wins (5–6 year lifecycle per management)
  • PC mainstream consumer design win and AI-voice product sampling are credible early-stage Fisher-style market expansion, but revenue contribution immaterial through at least FY2027
  • Management communication is candid and measured — acknowledged Android softness, adjacency immaturity, and foundry risks without promotional spin; passes Fisher's integrity test
  • Zero debt, $896M cash, active buybacks — balance sheet supports long-term R&D investment without dilution pressure; management willing to increase spend on high-confidence opportunities
  • Custom silicon strategy creates deep customer switching costs and long product lifecycles — structural moat quality Fisher respects

Red flags

  • 1.72% revenue CAGR is near-zero real growth — the single biggest Fisher disqualifier; a company this R&D-heavy should be growing at 8–15%+ to justify the growth label
  • Extreme customer concentration in one OEM (Apple) makes revenue episodic and dependent on one buyer's product cycle — antithetical to Fisher's preference for diversified demand pull
  • Non-smartphone adjacencies (PC, voice-AI, battery health) are all pre-revenue or immaterial — the growth story is a thesis, not yet a demonstrated track record
  • Android segment weakness further narrows the growth base; no clear volume catalyst in core markets to accelerate beyond 2–4% organic growth near-term
  • 10-K flagged onerous GlobalFoundries wafer commitment through 2026 and foundry capacity constraints from AI demand — supply risk could crimp growth execution
  • Insider selling (EVP multiple tranches in June 2026) under 10b5-1, while pre-planned, signals insiders not accumulating at current prices after a strong run
  • GuruFocus rates stock as overvalued despite strong quality metrics — valuation leaves little margin for error if PC ramp or Apple cycle disappoints

Bruce Greenwald — 🟡 watch · 58/100 · medium confidence

Cirrus Logic presents a genuinely interesting EPV case but falls into a 'watch' zone due to customer concentration opacity, uncertain moat durability, and a market price that is close to but not decisively below my EPV estimate. Let me work through the methodology rigorously.

EPV Calculation: I normalize operating earnings rather than use the provided DCF anchor. Reported operating income for FY2026 (period ending March 28, 2026) is $460M on $1,997M revenue — a 23.1% operating margin. Scanning the 5-year history, net margins ranged from 9.3% (FY2022, likely R&D/inventory cycle) to 20.8% (FY2026). The FCF margin trajectory is steep (5.5% → 31.9%), partly reflecting capex normalization (capex is remarkably low at $14M on $2B revenue — fabless model). I will use a normalized operating margin closer to the mid-cycle average, approximately 20%, giving normalized NOPAT ≈ $1,997M × 20% × (1 - 0.17 effective tax rate) ≈ ~$332M. Maintenance capex for a fabless chip designer is minimal (actual capex $14M vs. D&A likely ~$60-70M based on asset base), so I add back excess D&A to distributable earnings: adjusted distributable earnings ≈ $332M + ~$50M excess D&A ≈ ~$382M. Capitalizing at WACC of 10.27% (per the provided model): EPV of operations = $382M / 0.1027 ≈ $3,720M. Adding net cash ($778M): EPV equity ≈ $4,498M. Dividing by ~50.6M shares: EPV per share ≈ $89. At the current price of $146.50, the stock trades at a ~65% premium to my EPV — i.e., the market is pricing in substantial growth. A more generous normalization using the current-year margin (23.1%) yields: NOPAT ≈ $1,997M × 23.1% × 0.83 ≈ $383M, plus excess D&A ~$50M = $433M distributable; EPV ≈ $433M / 0.1027 ≈ $4,216M equity, or ~$99/share. Still a ~48% premium to EPV. Even a bull-case normalization (25% operating margin) yields EPV ~$110/share — still below $146.50.

Asset Reproduction Value: Cirrus Logic is fabless, so tangible assets are modest. Total assets $2.46B, but much is goodwill, intangibles, and cash. Book equity is $2.1B (~$41/share), largely cash ($778M) plus net working capital and IP. The true reproduction cost for a competitor to replicate CRUS's position — deep custom silicon relationships with Apple, multi-year audio codec expertise, 22nm custom process knowledge, engineering talent, and qualification cycles — is substantially higher than book. I estimate reproduction value at $70-100/share (cash at $15/share + $55-85/share for IP, talent, customer relationships). This puts EPV ($89-110) comfortably above reproduction value ($70-100), which is the right pattern: it signals a genuine franchise. The EPV/asset gap is positive, supporting the moat hypothesis.

Moat Assessment: EPV > reproduction value confirms a franchise exists, but the nature and durability of the moat requires scrutiny. Customer captivity is real: custom silicon for Apple's audio codec involves years of co-development, qualification, and switching costs — Apple cannot swap suppliers without major disruption. Economies of scale within the audio niche are present (Cirrus is the dominant supplier). Proprietary technology (22nm codec, camera controllers, boosted amplifiers) provides some protection. However: (1) The moat is heavily concentrated on ONE customer (Apple), whose bargaining power is immense; (2) The franchise value is contingent on Apple not vertically integrating audio (Apple has a track record of internalizing chips — e.g., cellular modems); (3) PC and battery adjacencies are nascent with no proven moat extension yet. The moat is real but narrow and dependent.

Margin of Safety: At $146.50, I'm paying a 33-65% premium to EPV — the opposite of a margin of safety. The market price is essentially the bear case of the provided DCF ($159) or requires ~2-5% growth indefinitely within a moat to be justified. That may be true, but it is speculative growth pricing, not EPV investing. The DCF's terminal value constitutes 69.2% of enterprise value — precisely the kind of far-future assumption I distrust most. The 21% DCF upside ($177 intrinsic) depends on 2% FCF growth perpetually, which is essentially the minimum case — yet the stock is only 21% below even that conservative DCF estimate.

Other Red Flags: The FY2022 net margin collapsed to 9.3% — suggesting the current margins may be somewhat cyclically elevated or benefiting from product mix. Revenue CAGR of 1.7% over 3 years confirms this is not a high-growth business; the FCF improvement is largely margin and working capital driven. Insider selling (EVP liquidating shares via 10b5-1) is mild noise but not alarming. The GlobalFoundries wafer commitment through 2026 is a supply-chain overhang not fully quantified.

Key points

  • EPV estimate ranges $89-110/share depending on margin normalization assumption — current price of $146.50 represents a 33-65% premium to EPV, meaning the market is paying for growth, not just earnings power
  • EPV materially exceeds asset reproduction value ($70-100/share), confirming a genuine franchise with real barriers to entry — customer captivity (Apple co-development) and niche scale economies are concrete and verifiable
  • Fabless model produces exceptionally low maintenance capex ($14M on $2B revenue), which inflates FCF and supports higher normalized distributable earnings than GAAP capex-adjusted figures suggest
  • Revenue CAGR of only 1.7% over 3 years and history of margin cyclicality (net margin collapsed to 9.3% in FY2022) suggest current 31.9% FCF margin may be near-peak rather than sustainable baseline for EPV normalization
  • Clean balance sheet ($778M net cash, zero debt, 7.65x current ratio) provides genuine downside protection and optionality; cash represents ~$15/share of intrinsic value directly
  • DCF terminal value = 69.2% of enterprise value — exactly the long-horizon speculative assumption I weight least; the DCF upside is modest (21%) and depends on perpetual modest growth being achievable within a protected franchise

Red flags

  • Price substantially above EPV at all reasonable normalization assumptions — buyer is paying for growth in a business where growth must be proven moat-protected to be value-creating
  • Extreme customer concentration on a single undisclosed customer (widely understood to be Apple) who has both the incentive and capability to internalize audio silicon as they have done with modems and application processors — this is an existential franchise risk
  • PC, voice-AI, and battery adjacencies are strategically interesting but contribute immaterial current revenue and have no demonstrated moat; paying for growth in unproven adjacencies is precisely the value-trap pattern I flag
  • FY2022 operating cycle showed margin compression to ~9% net — normalization must account for downside; using peak-cycle margins in EPV would overstate sustainable earnings power
  • GlobalFoundries wafer commitment through 2026 and AI-driven foundry capacity constraints flagged in 10-K represent unquantified supply-side risk to the earnings power thesis
  • Terminal value constitutes 69.2% of the DCF enterprise value — the provided valuation is heavily back-loaded and should not be used as an anchor for margin-of-safety calculations

Ray Dalio — 🟡 watch · 52/100 · medium confidence

Cirrus Logic presents a mixed picture through the Dalio macro lens. On the balance sheet, it is genuinely fortress-like: zero debt, $778M net cash, current ratio of 7.6x, and 31.9% FCF margin generating $637M in free cash flow against essentially no financing obligations. This passes the debt-cycle and self-funding test with flying colors — the company can survive a credit contraction, a deleveraging episode, or a sustained higher-for-longer rate environment without distress or capital markets access. That is a real strength in a risk-parity framework. However, the macro regime analysis reveals concentrated single-regime dependence that Dalio would penalize heavily. CRUS is overwhelmingly a falling-inflation, rising-growth, consumer-discretionary semiconductor play tethered to Apple's smartphone upgrade cycle. In stagflation (rising inflation + falling growth), consumer electronics demand compresses sharply, Apple's premium pricing power erodes at the margin, and fabless semiconductor suppliers face input cost pressure (wafer costs, energy, logistics) with limited pass-through leverage since pricing is largely negotiated with a single dominant OEM. In a deflationary bust, smartphone unit volumes historically decline meaningfully (discretionary spend, credit-dependent consumer in emerging markets, inventory correction cycles). The 3-year revenue CAGR of only 1.7% across the recent history already reflects this episodic, cycle-dependent nature. The PEG of 10.4x is essentially pricing permanent smooth growth into a cyclical business. Customer concentration risk — overwhelmingly Apple, per widely understood but deliberately opaque management policy — creates massive single-counterparty macro exposure; if Apple cycles down or loses premium market share in a slowdown, CRUS has no meaningful revenue diversification shield. PC and non-smartphone adjacencies remain immaterial revenue contributors today. The DCF intrinsic value of $177.83 (21% upside at $146.50) uses only 2% FCF growth — arguably too conservative given recent FCF surge — but the WACC of 10.27% assumes mid-cycle conditions; in a stagflationary regime with 10-year yields at 5-6%, the appropriate WACC rises materially and the terminal value (69.2% of enterprise value) compresses significantly. The foundry concentration risk (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints from AI) represents a supply-side inflation pass-through vulnerability. Geographic revenue concentration in a single-currency, US-dominated supply chain with geopolitical tail risk (Taiwan) is a Dalio-style hidden tail risk. Net assessment: the balance sheet is exemplary and earns meaningful credit, but the regime dependence, customer concentration, and valuation sensitivity to terminal assumptions in a rate-shift scenario limit the score to watch territory.

Key points

  • Zero debt and $778M net cash position: extraordinary balance sheet resilience; self-funding through any credit contraction without capital markets access
  • FCF margin of 31.9% ($637M) provides thick operational cushion against cyclical downturns
  • Revenue CAGR of only 1.7% over 3 years reveals this is not a structurally high-growth name; cyclical, episodic revenue tied to smartphone upgrade cycles
  • DCF terminal value represents 69.2% of enterprise value — extreme sensitivity to WACC and terminal growth assumptions; a 150bps WACC increase (stagflation regime) would materially impair the bear-case $159 floor
  • Bear-case DCF of $159 implies minimal downside from $146.50, but assumes no regime shift in cost of capital or structural demand compression
  • PC and voice-AI adjacencies are real optionality but contribute immaterial revenue today — cannot be weighted in a regime-stress test

Red flags

  • Single-regime dependence: earnings overwhelmingly tied to rising-growth, low-inflation, consumer-electronics boom conditions; stagflation or deflationary bust scenario would compress both volume and margin simultaneously
  • Extreme customer concentration (Apple presumed >80% revenue) creates a single-counterparty macro exposure that cannot be diversified away — one Apple inventory correction equals a CRUS earnings crisis
  • Foundry supply risk: GlobalFoundries wafer commitment through 2026 and AI-driven TSMC capacity constraints introduce input cost and supply-side inflation exposure with limited pass-through leverage against a dominant OEM customer
  • Terminal value constitutes 69.2% of DCF enterprise value — valuation is highly sensitive to the persistence of current low-rate, moderate-growth regime assumptions; a sustained higher-for-longer rate environment materially degrades intrinsic value
  • PEG ratio of 10.4x implies market is pricing smooth, persistent growth into a business with 1.7% 3-year revenue CAGR — significant re-rating risk in a growth slowdown
  • Insider selling by EVP ($650K+ in June 2026) under pre-planned 10b5-1 is not alarming alone but combined with stock at 52-week high creates cautionary signal
  • CHIPS Act subsidy exclusion as fabless player: disadvantaged vs. integrated peers in a geopolitically driven industrial policy environment favoring domestic capacity

Stanley Druckenmiller — 🟡 watch · 52/100 · medium confidence

CRUS presents a mixed picture through my lens. The directional thesis has genuine merit — the Q2 FY2026 record $561M revenue, expanding FCF margins (31.9% vs 22.2% three years ago), and emerging PC/AI-voice catalyst are real. The tape confirmed the thesis through April 2026 when CRUS hit 52-week highs. But right now the stock has pulled back from $180 to $146 — it is trading 18.8% BELOW its 52-week high while the narrative is still bullish. That tape disagreement is a critical red flag for me. The second derivative of revenue growth is also soft: 3-year revenue CAGR is only 1.72% — this is not an accelerating earnings story, it is a margin expansion story. Margin expansion is good but it does not give me the forward earnings INFLECTION I need to size up aggressively. The Apple concentration risk means I cannot independently model the earnings trajectory with confidence — the key variable is opaque. The PC adjacency, AI-voice products, and battery IP are in sampling/early-ramp stage with IMMATERIAL current revenue contribution; these catalysts are 12-24+ months from being earnings-moving. The liquidity/macro backdrop is ambiguous — no clear Fed easing tailwind that would compress the discount rate and re-rate the multiple. The FCF is exceptional (price-to-FCF of 11.6x is genuinely attractive) but I am not in the business of buying cheap — I am in the business of buying what is going UP. For me to size up, I need: (1) price reclaiming and sustaining above the $160-170 range showing the tape is back on my side, (2) evidence of forward estimate REVISIONS upward driven by PC design win revenue materializing, and (3) a clearer macro/Fed catalyst. Until then, this is a watch — high-quality business, interesting setup, but not yet the right moment to concentrate.

Key points

  • FCF margin expansion from 5.5% (2021) to 31.9% (2025) shows genuine quality improvement and operational leverage
  • Price-to-FCF of 11.6x and P/E of 17.8x are compellingly cheap for a business of this quality — but cheap alone is not my trigger
  • PC mainstream consumer design win (FY2027 revenue) and AI/voice sampling represent real but early-stage catalysts — not yet earnings-moving
  • Zero debt, $778M cash, pristine balance sheet gives flexibility and makes thesis management clean
  • Q2 FY2026 record $561M revenue shows the smartphone audio franchise remains durable with 5-6 year product cycle longevity
  • Revenue CAGR of only 1.72% over 3 years is not the accelerating growth profile I typically demand for maximum conviction
  • DCF intrinsic value of $177.83 suggests ~21% upside in base case — meaningful margin of safety exists if thesis plays out

Red flags

  • Stock is 18.8% below its 52-week high of $180.42 — the tape has rolled over from highs, disagreeing with the bullish fundamental narrative; I do not fight a broken tape
  • Apple customer concentration is deliberately opaque — I cannot independently model forward earnings trajectory with precision, making a concentrated directional bet unmanageable
  • Revenue CAGR 1.72% over 3 years signals earnings DIRECTION is not accelerating; margin expansion is masking revenue growth stagnation
  • Non-smartphone adjacencies (PC, AI-voice, battery/health) are in early-ramp or sampling phase — immaterial to near-term EPS, making it hard to identify the 'why now' earnings catalyst
  • EVP insider selling in June 2026 under 10b5-1 — while pre-planned, liquidating near highs does not convey peak insider conviction
  • No clear Fed/liquidity tailwind identified — macro regime ambiguous; without central bank support, multiple expansion thesis is harder to own
  • Supply chain risks (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints from AI demand) are unquantified — potential earnings estimate destabilizer

Benjamin Graham — 🟡 watch · 52/100 · medium confidence

Cirrus Logic passes several of Graham's quantitative tests but falls short of the strict margin-of-safety requirement that defines a true defensive buy. On the positive side, the balance sheet is genuinely strong: current ratio of 7.65 (well above the 2x minimum), zero long-term debt, $778M in net cash, and stockholders' equity of $2.1B. Earnings have been positive in every year of the 5-year history provided, with net income rising from $176M (FY2022) to $414M (FY2025), demonstrating the stability Graham prizes. ROE of 19.7% and ROIC of 17.3% show productive capital deployment. The trailing P/E of 17.84x is modestly above Graham's defensive ceiling of 15x, but the earnings yield of 6.96% compares favorably to high-grade bond yields (approximately 4.5-5%), providing a reasonable spread. Free cash flow generation is exceptional — $637M on $2B revenue, a 31.9% FCF margin — and the price-to-FCF of 11.6x is actually quite attractive by Graham's cash-earnings yardstick. The DCF intrinsic value of $177.83/share suggests the stock trades at roughly a 18% discount to base-case intrinsic value, which is meaningful but falls well short of Graham's preferred one-third margin of safety (~$119 or below would qualify). The critical failure on Graham's asset-based tests: P/B of 3.51x is well above his P/E x P/B = 22.5 threshold (17.84 x 3.51 = 62.6, versus the 22.5 ceiling), and the stock does not remotely approach net-net territory. Net current assets are approximately $1.225B (current assets $1.409B minus all liabilities — total assets $2.456B minus equity $2.103B implies ~$353M total liabilities, so NCAV ≈ $1.409B - $353M = ~$1.056B), or roughly $20.9/share against a $146.50 price — stock trades at 7x NCAV, far from net-net. The 3-year revenue CAGR of 1.7% is anaemic, barely meeting Graham's one-third growth test over a decade at this rate. No dividend is paid, which is a notable negative for Graham's defensive criteria. Customer concentration in Apple (though undisclosed) represents a qualitative risk that Graham would view as an unstable earnings foundation. The company passes as an established, profitable, debt-free enterprise with genuine earnings power, but the price does not offer adequate margin of safety on either earnings or asset bases by strict Graham standards.

Key points

  • Zero long-term debt and net cash of $778M — pristine balance sheet by any Graham standard
  • Current ratio of 7.65x far exceeds Graham's 2x minimum; working capital is robust
  • Trailing P/E of 17.84x slightly exceeds Graham's 15x defensive ceiling but earnings yield of 6.96% exceeds high-grade bond yields by a useful spread
  • Price-to-FCF of 11.6x is genuinely cheap on a cash-earnings basis
  • 5-year earnings history is positive throughout with rising net income (FY2021-FY2025) — meets stability test
  • DCF intrinsic value $177.83 implies ~18% upside — meaningful but below Graham's one-third margin of safety threshold
  • No dividends paid — fails Graham's dividend reliability criterion for defensive investors

Red flags

  • P/E x P/B = 17.84 x 3.51 = 62.6, nearly 3x Graham's 22.5 combined multiple ceiling — stock is expensive on assets
  • NCAV per share approximately $20-21 vs. $146.50 price — trades at ~7x net current asset value, not remotely a net-net
  • 3-year revenue CAGR of only 1.7% — insufficient growth to justify premium; long-term earnings growth trajectory uncertain
  • No dividend history — Graham's defensive investor criterion explicitly requires uninterrupted dividend record
  • Heavy customer concentration (widely understood to be Apple) creates earnings instability risk inconsistent with Graham's preference for diversified, stable revenue
  • Adjacency businesses (PC, battery, voice-AI) are nascent with no quantified revenue — growth narrative depends on unproven future results, precisely the speculative element Graham warns against
  • Insider selling by EVP in June 2026 under 10b5-1 plan — minor negative signal on management confidence at current price

Seth Klarman — 🟡 watch · 52/100 · medium confidence

Cirrus Logic presents a genuinely interesting balance sheet and cash-flow profile, but falls short of a Klarman-quality margin of safety at current prices. The DCF yields an intrinsic value of ~$178/share against a $146.50 current price, a 21% upside — modest by Baupost standards and dependent on a 2% FCF growth assumption that underweights the real risk: Apple customer concentration. The company is fundamentally high-quality (32% FCF margins, zero debt, $778M cash, current ratio 7.65x, ROIC 17%), but quality alone does not create a margin of safety. The bear DCF scenario at $159 is only 9% above current price, meaning even the conservative scenario offers little cushion. From a liquidation/asset perspective, stockholders equity of $2.1B against a $7.4B market cap yields price-to-book of 3.51x — no net-net or asset-coverage comfort here. The business is fabless, so tangible assets are minimal; the real value is in IP and customer relationships, both difficult to appraise conservatively and heavily dependent on a single opaque customer. Normalized FCF of $636M at 11.6x price-to-FCF is genuinely cheap for a semiconductor company, and if one credits the full cash position ($778M, ~$15/share), the ex-cash P/FCF is roughly 10x — reasonable but not deeply discounted. The three-pillar diversification strategy (PC, voice-AI, battery health) is promising but contributes immaterial current revenue; valuing those options requires optimistic growth assumptions that I would not embed in a conservative appraisal. There is no special-situation catalyst, no forced selling, no distress — this is a well-followed, well-loved semiconductor name trading near 52-week highs with bullish retail sentiment. That is precisely the environment where Baupost holds cash rather than chases. The insider selling (EVP liquidating ~$850K in June under 10b5-1) is minor but consistent with a stock that has run hard. Supply chain risks (GlobalFoundries wafer commitments, TSMC AI capacity competition, geopolitical logistics) are real and unquantified. CHIPS Act exclusion as a fabless player is a structural disadvantage. The stock warrants watching for a better entry — a 30-40% decline to the $90-105 range would begin to create genuine margin of safety against normalized FCF and would represent a bear-scenario overshoot worth acting on.

Key points

  • Zero debt, $778M cash ($15.38/share), current ratio 7.65x — fortress balance sheet provides real downside cushion
  • FCF of $636M at price-to-FCF of 11.6x (ex-cash ~10x) is reasonably priced but not deeply discounted by Baupost standards
  • DCF intrinsic value $178/share implies only 21% upside at current price; bear scenario $159 is only 9% above current — insufficient margin of safety
  • Net income $414M, operating margin 23%, FCF margin 32% — genuine high-quality business that deserves a premium, but premium is already partially priced in
  • 3-year revenue CAGR of 1.7% is pedestrian; growth required to justify current valuation must come from unproven PC/voice-AI adjacencies
  • Price-to-book 3.51x on a largely intangible-asset business; no tangible asset floor to provide liquidation comfort

Red flags

  • Apple customer concentration is the central unquantifiable risk — management policy prohibits disclosure; any Apple supply-chain shift or design-out is potentially catastrophic and cannot be stress-tested from public data
  • Value thesis partly depends on PC/voice-AI/battery adjacencies that are pre-revenue; embedding these in valuation requires optimism, not conservatism
  • No special-situation catalyst, no forced selling, no distress — stock is near 52-week highs with bullish retail and analyst sentiment, exactly the wrong setup for value entry
  • GuruFocus rates overvalued despite 80/100 GF Score; retail sentiment is shallow and bullish ('easy $200 target'), suggesting Mr. Market is not offering a discount
  • Insider selling (EVP, $850K+ in June 2026) under 10b5-1 is pre-planned but directionally unfavorable at current price levels
  • Foundry risk: GlobalFoundries wafer commitment through 2026 is potentially onerous; AI-driven TSMC capacity constraints could squeeze fabless players
  • Revenue CAGR of 1.7% over 3 years while DCF uses 2% growth — the model is barely above stagnation; any demand shock from smartphone cycle turns FCF negative relative to current valuation

Peter Lynch — 🟡 watch · 52/100 · medium confidence

Cirrus Logic is a stalwart/slow-grower semiconductor company — not a fast grower — and must be judged on those terms. Revenue CAGR of just 1.7% over 3 years and a PEG of 10.37 are the two numbers that immediately concern me. The Lynch framework is unambiguous: you never want to pay a P/E far above the growth rate, and a PEG of 10+ is a flashing red light regardless of how good the business is. The stock's P/E of ~17.8x is being compared against anemic 1.7% revenue growth — that's the core problem here. Now, there are genuinely good things: the balance sheet is pristine ($778M net cash, zero debt, current ratio 7.6x), FCF generation is excellent ($636M, 31.9% FCF margin), and the company's P/FCF of 11.6x is actually quite attractive for a high-quality chip maker. The business is understandable — Cirrus makes audio chips that go inside Apple iPhones, which is as close to a 'product you can see in everyday life' as semiconductors get. But the Apple concentration is my biggest story-clarity concern: management won't even discuss their largest customer by name, which means the 'story' has a black box at its center. That's not how Lynch investing works — I want to explain the full story in two sentences. The three-pillar diversification strategy (PC, voice-AI, battery health) is promising but unproven; these are adjacencies with 'immaterial current revenue' by management's own admission. I like the PC mainstream consumer win and the AI/voice PC product sampling, but these are catalysts 2-3 years out, and I don't pay today's price for tomorrow's maybe. The FCF math is appealing and DCF intrinsic value of $177.83 (21% upside) provides some comfort at $146.50. But for Lynch, the earnings growth rate is the anchor — and 1.7% revenue CAGR with a 10x PEG means I'm being asked to overpay for a stalwart. The right stalwart buy is 30-50% gain potential at a fair P/E, and the DCF suggests that's achievable, but I need faster earnings growth to justify holding long-term. Insider selling (EVP liquidating ~$850K in June) adds a minor negative signal, though the 10b5-1 pre-plan context reduces alarm. This is a watch — attractive FCF and balance sheet quality, but the growth engine isn't firing fast enough to justify the PEG, and customer concentration opacity violates my 'understand the story' rule.

Key points

  • PEG of 10.37 is far above Lynch's 1.0 threshold — the single most damning metric for a growth investor; P/E of 17.8x vs. revenue CAGR of 1.7% does not compute
  • Business is understandable at a high level: premium audio chips for smartphones, primarily Apple iPhones; products visible in everyday devices
  • Balance sheet is exceptional: $778M net cash, zero debt, current ratio 7.6x, FCF margin 31.9% — Lynch loves cash-rich, debt-free companies
  • P/FCF of 11.6x is genuinely cheap for this quality level and provides a floor argument; DCF intrinsic value $177.83 implies 21% upside
  • Category classification: stalwart/slow-grower, not a fast grower — so I'm looking for 30-50% gain potential at a fair price, not a multi-bagger thesis
  • PC diversification is early but credible: first mainstream consumer laptop win with revenue expected FY2027 and AI/voice PC product sampling — a real adjacency with Lynch-style unit-expansion potential if it materializes
  • FCF conversion is excellent ($636M on $2B revenue) confirming the growth that did occur is real and not accounting-driven

Red flags

  • PEG of 10.37 is catastrophically high for Lynch-style investing — paying 10x the growth rate is textbook overpayment for a stalwart
  • Customer concentration black box: management refuses to discuss largest customer (Apple), making it impossible to fully 'understand the story' — a fundamental Lynch requirement
  • Revenue CAGR of only 1.7% over 3 years firmly places this in slow-grower territory, not a category where Lynch commits serious capital at 17.8x earnings
  • Non-smartphone adjacencies (PC, battery health, voice-AI) contribute immaterial current revenue — adjacency diversification is promising but resembles unproven 'next-big-thing' hype at this stage
  • Insider selling: EVP sold ~$850K+ in June 2026 via 10b5-1 plan — not alarming on its own but directionally negative signal at a 52-week high
  • Stock at 52-week high with GuruFocus flagging overvalued — heavy analyst coverage (CFRA, Argus, LSEG, Fidelity all bullish) means neglected-stock edge is absent
  • Foundry capacity risk (GlobalFoundries wafer commitment through 2026, TSMC AI-driven constraints) could crimp margins or force costly supply pivots

Howard Marks — 🟡 watch · 52/100 · medium confidence

From a Marks risk-first lens, Cirrus Logic presents a genuinely mixed picture. The balance sheet is pristine — zero debt, $778M cash, current ratio 7.65x, strong FCF generation ($636M, 31.9% FCF margin) — which eliminates the capital structure fragility I fear most. That is the core safety net. However, the risk-adjusted investment case is complicated by what appears to be a sentiment pendulum that has already swung meaningfully toward optimism. The stock sits at its 52-week high ($146.5 = 52w high per price data, though 52w high is listed as $180.42 elsewhere — the data shows conflicting signals suggesting recent price decline from peak), with retail chatter uniformly bullish, GuruFocus flagging overvaluation, and analyst consensus firmly in the Buy camp. This is not the environment where I find the best risk/return: the crowd is already on the right side. The DCF shows only 21.4% upside to intrinsic value ($177.83 base case), with that intrinsic value depending on a 2% FCF growth assumption that essentially flatlines a business with only 1.7% revenue CAGR over 3 years. The terminal value represents 69.2% of enterprise value — an uncomfortable dependence on a Gordon-growth perpetuity assumption in a semiconductor business with concentrated customer risk. The bear DCF case is $159.36, only 8.8% above current price, meaning the margin of safety is razor thin against a pessimistic but plausible scenario. On second-level thinking: what does the crowd believe? That Apple dependency is manageable, PC diversification is credible, margins are sustainable, and the fabless model is durable. The contrarian question is whether any of that is wrong — and frankly, with Apple concentration undisclosed but widely understood to dominate, a single OEM cycle disruption could materially impair revenue. The 10-K flagged onerous GlobalFoundries wafer commitments through 2026 and supply chain risk, neither quantified. The 4% YoY revenue growth in a 'record' quarter is not the growth profile that justifies elevated expectations. The PEG of 10.37 is alarming — it prices in growth that has not materialized. Insider selling under 10b5-1 (EVP liquidating ~$650K+ in June) is a mild negative signal. On positives: P/FCF of 11.6x is genuinely inexpensive for quality; the operating margin of 23% and ROE of 19.7% are solid; no debt means no forced selling or covenant risk in a downturn. The asymmetry, however, is the core problem: limited downside protection in price (bear case only 8.8% below current) with upside also constrained (21% to base DCF). This is a good company at a fair-to-full price, not a beaten-down asset with embedded pessimism. I want to be paid to take risk; here, I am not being paid enough for the concentration and cyclicality risks embedded in this business.

Key points

  • Balance sheet fortress: zero debt, $778M cash, 7.65x current ratio — eliminates capital structure fragility risk entirely
  • P/FCF of 11.6x is genuinely inexpensive for a business generating 31.9% FCF margins; earnings yield 6.96%
  • DCF base case intrinsic value $177.83 implies only 21.4% upside — not a deep discount, more like fair-to-full value
  • Bear DCF case $159.36 is only 8.8% above current price — thin margin of safety against plausible pessimism
  • Terminal value represents 69.2% of enterprise value, creating heavy dependence on perpetuity assumptions in a cyclical semiconductor business
  • Revenue CAGR of only 1.7% over 3 years yet PEG of 10.37 — growth expectations appear too optimistic relative to demonstrated trajectory
  • Retail sentiment uniformly bullish, analyst consensus Buy/Strong Buy — the pendulum has swung toward optimism, not fear

Red flags

  • Sentiment at peak popularity: GuruFocus overvalued tag, stock at/near 52-week high, unanimous bullish retail chatter — the crowd is already on board
  • Customer concentration opaque: Apple dependency undisclosed but dominant; a single OEM cycle shift could materially impair revenue with no visible hedge
  • Thin margin of safety vs. downside: bear DCF only 8.8% above current price — limited protection if cycle turns or Apple content share declines
  • PEG of 10.37 implies the market is paying for growth well above the 1.7% demonstrated 3-year CAGR — expectations are ahead of reality
  • GlobalFoundries wafer commitment through 2026 flagged as potentially onerous in 10-K — unquantified but real supply-cost risk
  • Insider selling: EVP liquidating ~$650K+ in June 2026 under 10b5-1; pre-planned but concentration of recent activity is a mild negative
  • Non-smartphone adjacencies (PC, voice-AI, battery health) contribute immaterial revenue yet are central to bull thesis — unproven optionality priced into a semiconductor name

Walter Schloss — 🔴 avoid · 28/100 · high confidence

Cirrus Logic fails virtually every Schloss deep-value criterion. The stock trades at 3.51x book value and 3.7x sales — far above the near-or-below tangible book standard that anchors my entire method. Price is at its 52-week HIGH ($146.50 equals the 52-week high per the fact base), not a multi-year low, which is the opposite of what I look for. I buy what is beaten down and unloved; this stock has been rising and is at peak pricing. Tangible book per share works out to roughly $41.50 (stockholders' equity $2.103B / 50.58M shares), and the stock at $146.50 trades at 3.5x that figure — nowhere near the discount to hard assets that provides my required margin of safety. Yes, the balance sheet is clean: zero long-term debt, current ratio of 7.65, $778M cash, and the company is asset-light with minimal capex ($14M). But 'asset-light' is precisely the problem for my approach — there are no heavy tangible assets (plant, real estate, inventory at scale) to appraise as downside protection independent of the earnings story. The business value lives almost entirely in intangible competitive position (custom silicon design wins, Apple relationship) and earnings power — exactly the kind of thesis I avoid because it requires faith in projections rather than verification against the balance sheet. P/E of 17.8x is not egregious for a quality compounder, but that is not my bar. FCF yield of ~8.6% (price-to-FCF of 11.6x) is decent but the asset anchor simply isn't there. The DCF intrinsic value of $177.83 suggests 21% upside, but DCF is an earnings/growth model, not an asset appraisal — I don't anchor to it. Insider activity is negative: EVP Jeffrey Baumgartner sold shares twice in June 2026 under a 10b5-1 plan, totaling several thousand shares. No insider buying reported. Revenue CAGR of only 1.7% over three years confirms this is a slow-growth business, which is fine, but slow growth combined with a rich price-to-book leaves no statistical bargain here. The business itself appears solidly run with excellent FCF margins (31.9%) and a pristine balance sheet — I have nothing against the company's quality. But Schloss Associates was never in the business of paying 3.5x book for quality; we paid below book for ordinary businesses and let mean reversion do the work.

Key points

  • Price-to-book of 3.51x versus my requirement of near or below tangible book value; tangible book roughly $41.50/share vs $146.50 stock price
  • Stock is at its 52-week HIGH, not a low — the complete opposite of a Schloss entry point
  • Zero long-term debt and $778M cash are genuinely excellent, but the asset base is mostly intangibles/goodwill in the form of design IP, not appraisable hard assets
  • Revenue CAGR of only 1.72% over 3 years confirms modest growth, yet valuation is pricing in a quality premium I refuse to pay
  • FCF of $636M and FCF margin of 31.9% reflect real earnings quality but are capitalized at a rich multiple, not a statistical bargain
  • No dividend paid — shareholder return via buybacks only, so the patient value holder receives no cash while waiting
  • Business is fabless semiconductor — almost entirely earnings/IP dependent, with de minimis tangible fixed assets ($14M capex signals no heavy plant)

Red flags

  • Price at 52-week high ($146.50 = high_52w per data) — the antithesis of a beaten-down Schloss candidate
  • Price-to-book 3.51x; paying 3.5x for tangible assets with no margin of safety in the balance sheet
  • Insider selling: EVP sold shares twice in June 2026 (~5,077 shares total) with no offsetting buying reported
  • Thesis entirely dependent on intangible competitive moat (Apple custom silicon relationship, design-win pipeline) — cannot be verified from balance sheet assets alone
  • Asset-light fabless model provides no hard-asset floor; if competitive position erodes, book value provides minimal protection at these prices
  • No dividend; capital return entirely discretionary via buybacks which can be suspended
  • GuruFocus independently flags stock as overvalued despite strong operating quality, consistent with my own assessment

Fact base appendix

Price

  • last_close: 146.5
  • as_of: 2026-06-27
  • high_52w: 146.5
  • low_52w: 146.5
  • pct_below_52w_high: 0.0

Fundamentals

  • last_price: 146.5
  • market_cap: 7391323369
  • fifty_two_week_high: 180.42
  • fifty_two_week_low: 92.02
  • beta: 1.1949037
  • currency: USD
  • exchange: NASDAQ NMS - GLOBAL MARKET
  • sector: Semiconductors
  • industry: Semiconductors
  • price_source: finnhub
  • bars: 1
  • entity: Cirrus Logic, Inc.
  • fiscal_year: 2025
  • revenue: 1997379000
  • revenue_period: 2026-03-28
  • net_income: 414408000
  • net_income_period: 2026-03-28
  • operating_income: 460380000
  • operating_income_period: 2026-03-28
  • operating_cash_flow: 650598000
  • operating_cash_flow_period: 2026-03-28
  • capex: 13988000
  • capex_period: 2026-03-28
  • total_assets: 2456520000
  • total_assets_period: 2025-12-27
  • current_assets: 1409349000
  • current_assets_period: 2025-12-27
  • current_liabilities: 184191000
  • current_liabilities_period: 2025-12-27
  • stockholders_equity: 2103229000
  • stockholders_equity_period: 2025-12-27
  • cash_and_equivalents: 778083000
  • cash_and_equivalents_period: 2025-12-27
  • shares_outstanding: 50582893
  • operating_margin: 0.2305
  • net_margin: 0.2075
  • roe: 0.197
  • debt_to_equity: None
  • current_ratio: 7.6516
  • roic: 0.1729
  • free_cash_flow: 636610000
  • fcf_margin: 0.3187
  • pe_ratio: 17.84
  • price_to_fcf: 11.61
  • price_to_sales: 3.7
  • revenue_cagr: 0.0172
  • revenue_cagr_years: 3
  • fundamentals_source: edgar_companyfacts
  • price_to_book: 3.51
  • earnings_yield: 0.0696
  • peg: 10.37

Filings reviewed

  • 10-K (2026-05-21) https://www.sec.gov/Archives/edgar/data/772406/000077240626000018/crus-20260328.htm
  • 8-K (2026-05-06) https://www.sec.gov/Archives/edgar/data/772406/000077240626000012/crus-20260504.htm
  • 8-K (2026-03-31) https://www.sec.gov/Archives/edgar/data/772406/000077240626000009/crus-20260325.htm
  • 10-Q (2026-02-03) https://www.sec.gov/Archives/edgar/data/772406/000077240626000006/crus-20251227.htm
  • 10-Q (2025-11-04) https://www.sec.gov/Archives/edgar/data/772406/000077240625000041/crus-20250927.htm
  • 10-K (2025-05-23) https://www.sec.gov/Archives/edgar/data/772406/000077240625000014/crus-20250329.htm

Other sources

  • [news] Cirrus Logic vs. Skyworks: Which Chip Stock is the Smarter Buy? - Yahoo Finance
  • [news] Insider Sell: Jeffrey Baumgartner Sells 1,170 Shares of Cirrus L - GuruFocus
  • [news] Cirrus Logic EVP Jeffrey Baumgartner sells $196,829 in stock. - Investing.com
  • [news] Form 4 Cirrus Logic Inc For: 24 June By Investing.com - Investing.com Canada
  • [news] Cirrus Logic (NASDAQ: CRUS) EVP sells 1,170 shares under 10b5-1 plan - Stock Titan
  • [news] Price to earnings ratio of Cirrus Logic, Inc. – XETR:CRU - TradingView
  • [news] CRUS Forecast — Price Target — Prediction for 2027 - TradingView
  • [news] Cirrus Logic Inc (CRUS) Stock Up 3.2% but GF Value Says Overvalued -- GF Score: 80/100 - GuruFocus
  • [news] Cirrus Logic, Inc. (CRUS) Hit a 52 Week High, Can the Run Continue? - Yahoo Finance
  • [news] CRUS Q4 Earnings & Sales Top, Up Y/Y as Diversification Gains Momentum - Yahoo Finance
  • [news] Cirrus Logic (NASDAQ: CRUS) EVP exercises options, sells 3,907 shares under 10b5-1 plan - Stock Titan
  • [news] Cirrus Logic, Inc. (CRUS) Hits Fresh High: Is There Still Room to Run? - Yahoo Finance
  • [news] Cirrus Logic to Release Q4 Earnings: Here's What to Expect - Yahoo Finance
  • [news] Cirrus Logic, Inc. Actuals & Estimates (NASDAQ:CRUS) - TradingView
  • [news] Why Is Cirrus Logic (CRUS) Up 2.3% Since Last Earnings Report? - Yahoo Finance
  • [discussion] [Bullish] $CRUS great price to buy and to add more today with the 8.8 bullish score out of 10.
  • [discussion] [Bullish] $CRUS great buy efore AAPL conference.
  • [discussion] $CRUS Brilliant piece that captures CRUS's situation perfectly. So if you want to update your un
  • [discussion] [Bullish] $CRUS I have covered Cirrus Logics for over 10 years. This company has a A rating with ne
  • [discussion] [Bullish] $CRUS Argus and Trading Center, Market Edge, LSEG and a few other list of analysts cover C
  • [discussion] [Bullish] $CRUS Stop trading. Cirrus Logics is an easy over $200 dollar stock by the end of the yea
  • [discussion] [Bullish] $CRUS Feels like this and $SWKS are the only semi-adjacent stocks that every broke fool on
  • [discussion] [Bullish] $CRUS Cirrus Logics is rated strong buy by CFRA, Market Edge, LSEG, Vikers and the list go
  • [discussion] [Bullish] $CRUS Trade with intention, your vision and your courage. Cirrus Logic is a great long ter
  • [discussion] $CRUS - Cirrus Logic Inc - 10K - Updated Risk Factors

CRUS flags tighter foundry and logistics ri

  • [discussion] $SPY

Trump went KRUS — think he meant $CRUS haaaa 😂

  • [discussion] Wall St is expecting 1.82 EPS for $CRUS Q1 [Reporting 08/05 AMC] http://www.estimize.com/intro/crus?
  • [discussion] $CRUS More potential upside.✅📈 Thanks to $AAPL Apple improving sales.
  • [discussion] $CRUS Price: $166.47 (-4.45%) Trend: Bullish Market Bias (7D): Bullish Bias 📈 Expected Range: ±2.29%
  • [discussion] $CRUS Current Stock Price: $174.09 Contracts to trade: $175.0 CRUS May 15 2026 Call Entry: $3.00 Exi
  • [earnings_call] Cirrus Logic, Inc CRUS Q2 2026 Earnings Call

Generated 2026-07-17T19:44:05 · est. cost $1.43

What each investor thinks

01

Joel Greenblatt Value

pass · 78

Cirrus Logic scores well on both legs of the Magic Formula. On ROIC: EBIT of $460M against a tangible capital base that is extraordinarily lean — net working capital (current assets $1,409M minus current liabilities $184M = $1,225M) plus net fixed assets (capex of only $14M implies minimal net PP&E as a fabless company; using total assets minus current assets minus intangibles, net fixed assets are likely sub-$200M) yields an EBIT/tangible capital ratio that is very high, consistent with the reported ROIC of 17.3% on a book basis. On earnings yield: EV = market cap $7,391M + zero debt - excess cash $778M ≈ $6,613M. EBIT yield = $460M / $6,613M ≈ 6.96%. For a semiconductor company with near-zero capex ($14M on $2B revenue), a 52%+ gross margin, and strong FCF conversion (FCF margin 31.9%), this earnings yield is reasonable but not screaming cheap — the DCF intrinsic value of $177.83 per share represents a 21% upside to the current $146.50 price, suggesting meaningful but not extreme margin of safety. The business earns genuinely high returns on the minimal tangible capital it employs (fabless model = working capital is the main capital employed), recurring EBIT converts cleanly to FCF ($637M FCF vs $460M EBIT, the gap reflecting low-cash-tax and minimal capex), and the capital structure is pristine (net cash of $778M). Revenue CAGR of 1.7% over 3 years is modest but FCF margin expansion from 5.5% to 31.9% over 4 years shows dramatically improving capital efficiency. The P/FCF of 11.6x is genuinely attractive for a business of this quality. No special-situation catalyst is present, which limits the upside multiple; and customer concentration in Apple remains an unquantifiable but real risk. Insider selling via 10b5-1 is a mild negative signal but pre-planned. Overall: a good-to-excellent business at a fair-to-moderately-cheap price — squarely in Greenblatt's buy zone.

02

Michael Mauboussin Quality

pass · 74

Cirrus Logic presents a compelling ROIC-above-WACC story with measurable moat sources, though the durability of the spread and the expectations embedded in the current price require careful decomposition. ROIC of 17.3% against an estimated WACC of ~10.3% yields a spread of roughly 700 bps — not enormous by franchise standards, but real and positive. The moat mechanism is identifiable and specific: deep switching costs embedded in custom silicon co-development with a dominant OEM (Apple), proprietary codec/amplifier IP that takes years to qualify, and scale economics in R&D amortization across a 5–6 year product lifecycle. These are not adjective-moats — they are structural barriers. The current price of $146.50 implies roughly 21% upside to the DCF base case of $177.83, suggesting the market is pricing in modest 2% FCF growth (the revenue CAGR used as the growth assumption) — which is arguably conservative given the FCF margin expansion from 5.5% (FY2021) to 31.9% (FY2025). This is a key insight from expectations investing: the embedded expectations appear beatable if the PC and mixed-signal adjacencies deliver even modest traction. However, the base rate for semiconductor companies this deeply concentrated in one customer cycle is mean-reversion pressure on ROIC, and the unproven adjacencies (PC, voice-AI, battery health) are not yet in the numbers. The distribution of outcomes skews modestly positive from here, but the fat downside tail — Apple concentration unwinding, or AI-PC failing to materialize — is real and unquantified. Management's capital allocation is disciplined (buybacks, zero debt, $778M cash, capex of only $14M reflecting fabless efficiency), and the 10b5-1 insider selling is routine rather than alarming at this scale. The PEG of 10.37 is elevated, but this metric is misleading here since the 1.7% revenue CAGR massively understates FCF trajectory — the correct growth input should be FCF CAGR, which is dramatically higher. On balance, the embedded expectations are achievable but not comfortably discounted, and the Apple opacity is a genuine epistemic handicap for any moat-durability assessment.

03

Forensic Short-Seller (Chanos/Einhorn-style) Referee

pass · 72

Cirrus Logic passes the forensic short-seller's core accounting tests with unusually clean results. The earnings-vs-cash test — the Chanos/Einhorn lead indicator — goes strongly against a short thesis: operating cash flow of $650.6M materially EXCEEDS net income of $414.4M in FY2025, and FCF of $636.6M exceeds net income by 54%. This is the opposite of the classic accrual red flag. Historical pattern holds: in FY2024, FCF ($421.6M) exceeded net income ($331.5M); in FY2023, FCF ($384M) vs net income ($275M); FCF margins have expanded from 5.5% in FY2021 to 31.9% in FY2025. Cash genuinely converts from earnings — no earnings quality crisis. Balance sheet is fortress-class: zero debt, $778M cash, current ratio 7.65x, net cash position of ~$778M offsets any valuation concern. Working capital signals are clean: no evidence of receivables or inventory outrunning revenue (DSO data not granularly available in the fact base, but flagging absence). Capex is minimal ($14M on $2B revenue), consistent with fabless model — no aggressive capitalization of costs that flatter margins. Share buybacks appear genuine: shares outstanding at ~50.6M; the fact base does not show share count expansion, suggesting buybacks are at least partially reducing count rather than just treading dilution water, though SBC quantum is not fully disclosed. Insider selling (EVP Baumgartner, ~5,000 shares in June under 10b5-1) is modest relative to float and pre-planned — not a cluster red flag. No auditor changes, restatements, or going-concern language visible. The genuine forensic concerns are: (1) customer concentration opacity — management refuses to discuss largest customer (Apple), creating a material disclosure gap that Chanos would flag as governance/transparency risk; (2) the revenue CAGR of 1.7% over 3 years barely justifies the growth story embedded in the valuation, while new adjacency revenues (PC, voice-AI, battery) are immaterial and unquantified — this is a 'story stock' risk without a falsifiable catalyst yet; (3) GlobalFoundries wafer commitment flagged in 10-K as potentially onerous through 2026 — a write-down risk if demand softens; (4) the DCF relies on 2% FCF growth (matching tepid revenue CAGR) but 69% of value sits in terminal value, creating sensitivity to any demand disruption; (5) a PEG of 10.37 signals the market is pricing growth that the historical record does not yet support. The kill question for a short thesis: a meaningful Apple design loss or smartphone content reduction (e.g., Apple insourcing audio chips) would collapse the revenue base. No evidence this is imminent, but the opacity makes it unverifiable. Overall: forensic accounting gives CRUS a clean bill of health — this is not a fraudulent or accrual-inflated business. The bear case rests on concentration/disruption risk and narrative valuation, not accounting manipulation, which is outside the core forensic toolkit.

04

AI & Disruption Referee (Christensen-style) Referee

pass · 72

Cirrus Logic is a fabless semiconductor company whose core value is custom-designed mixed-signal silicon — audio codecs, boosted amplifiers, haptic drivers, camera controllers — embedded directly into physical devices. The 'job' it does is not knowledge work or intermediation: it converts analog signals to digital and vice versa, manages power delivery, and provides ultra-low-power voice wake functionality at the hardware layer. AI cannot replace this job by software substitution; physics requires the silicon to exist between the real world and the digital processor. The disruption question for CRUS is therefore not 'can AI do this cheaper via software' but rather (1) does AI commoditize the chip design process making CRUS's IP less defensible, (2) do hyperscalers vertically integrate their own audio/mixed-signal silicon, (3) does AI-driven demand (AI PCs, on-device inference) expand or contract Cirrus's addressable market, and (4) does AI accelerate or decelerate Apple's hardware cycle. On balance the disruption scorecard is net-positive-to-neutral. The AI PC and on-device voice processing wave is structurally additive — Cirrus explicitly called out an ultra-low-power AI/voice wake product sampling in Q4 FY2026, directly leveraging its smartphone voice IP for PC; this is a genuine tailwind as AI PCs require always-listening, low-power audio front-ends precisely in Cirrus's wheelhouse. On-device AI inference on smartphones also increases the premium on efficient mixed-signal front-ends (microphones, codecs, camera analog front-ends) because data must be captured cleanly before neural processing. The moat-commoditization question is more nuanced: EDA tools and AI-assisted chip design (Synopsys DSO.ai, Cadence Cerebrus, internal hyperscaler flows) do lower barriers to analog/mixed-signal design at the margin, but custom silicon for a tier-1 OEM requires deep co-design, years of silicon iterations, and process-node-specific tuning that is not yet automatable end-to-end. Cirrus's 5-6 year product lifecycle with Apple-class customers reflects sticky, co-engineered relationships that are not easily replicated by a model plus commodity IP. The vertical integration risk (Apple designing its own audio codec) is a real and ongoing concern — Apple already designs its own application processors — but audio mixed-signal remains a specialized domain where Apple has historically chosen to source externally; there is no public evidence of Apple in-housing codec/amp design as of FY2026. The hyperscaler capture risk is low: Google, Microsoft, and Amazon do not have a meaningful path to supplying physical audio silicon to Apple or Android OEMs. The AI-chip-design commoditization risk is real on a 5-10 year horizon but not yet a near-term displacement mechanism — Cirrus's defensibility rests on OEM co-design relationships and process-specific analog tuning, not on being the only party that can run SPICE simulations. Management's treatment of AI is mixed: they frame AI PCs and voice features as tailwinds (credible and specific — sampling product disclosed), but the 10-K risk factors do flag AI-driven foundry capacity constraints (a second-order supply risk, not a demand-side displacement, but real). No explicit engagement with whether AI chip design tools erode their R&D moat — a gap, though not an immediate red flag. The falsifiable call: over the next 2-4 years, evidence of displacement would be Apple announcing in-house audio codec production, Cirrus losing a design cycle to an AI-native mixed-signal startup, or take-rate compression on custom silicon contracts. Evidence of reinforcement would be AI PC voice products achieving material revenue, camera controller content per device expanding, and gross margin holding above 51% as AI-driven device complexity increases. Current data — 52.5% gross margin, FCF margin rising to 32%, no disclosed competitive displacement — is consistent with the reinforcement scenario. Score 72: genuine hardware moat that AI makes somewhat more valuable on the demand side (AI PC, on-device voice), partially offset by longer-term analog design commoditization risk and Apple insourcing optionality.

05

Chuck Akre Quality

watch · 62

Cirrus Logic passes two of Akre's three legs convincingly but stumbles badly on the reinvestment runway — the most important leg. The business quality is genuinely impressive: 31.9% FCF margins, 17.3% ROIC, 19.7% ROE, all achieved with zero debt and minimal capex ($14M against $637M FCF). The capital-light, high-IP fabless model with deep switching costs at Apple is a real franchise. Management appears disciplined — clean balance sheet, active buybacks (shares outstanding declining), candid tone on adjacency timelines. But the reinvestment problem is acute: revenue CAGR of only 1.7% over three years, an overwhelming single-customer dependency (Apple), and nascent non-smartphone markets that are years from material contribution. The company is generating $637M in annual FCF from a $7.4B market cap business but cannot clearly demonstrate where it redeploys that cash at similarly high rates. The 'compounding machine' test requires not just generating cash but reinvesting it at high rates for years — CRUS is currently accumulating cash ($778M) and returning it via buybacks, which is rational but signals limited organic reinvestment opportunity at scale. The DCF intrinsic value of $177.83 vs. $146.50 current price offers ~21% upside in the base case, making valuation not egregious (P/FCF of 11.6x, P/E of 17.8x are reasonable for the quality). However, the 2% FCF growth assumption embedded in the DCF captures the reinvestment problem well — this is not a high-growth compounder, it's a high-quality cash cow. For Akre's framework, a 'pass' requires all three legs; the reinvestment runway leg is clearly impaired by customer concentration and unproven adjacencies.

06

Warren Buffett Quality

watch · 62

Cirrus Logic is an interesting semiconductor business that sits at the edge of my circle of competence — it is profitable, has a long operating history, generates genuine free cash flow, and possesses elements of a moat through deep customer integration and custom silicon design. However, several features give me pause. The business is heavily reliant on what appears to be a single dominant customer (Apple), which is both its greatest strength and its most significant vulnerability. The moat is real but narrow and customer-dependent rather than broadly durable in the way I prefer. On the positive side: ROE of ~19.7%, ROIC of ~17.3%, FCF margin expanding to ~31.9%, zero debt, current ratio of 7.6, and owner earnings that are genuinely clean (capex is only $14M against $650M operating cash flow — this is not a capital-intensive business, which I greatly admire). The P/E of 17.8x and price-to-FCF of 11.6x are modest for a business of this quality. The DCF suggests ~21% upside to intrinsic value of $177.83, which provides a thin but real margin of safety. Revenue CAGR of only 1.7% over three years tempers enthusiasm — this is a slow-growth business at current scale, not a compounder expanding its addressable market rapidly. The PC and AI/voice adjacencies are early-stage and unproven, which I discount heavily. Insider selling under 10b5-1 plans is noted but not alarming. What prevents a higher score: customer concentration opacity (I cannot assess the moat without understanding Apple dependency quantitatively), modest revenue growth suggesting limited pricing power expansion, and the semiconductor industry's inherent cyclicality which runs counter to my preference for businesses I can forecast a decade out with high confidence.

07

Valuation Referee (Damodaran-style) Referee

watch · 62

The DCF model provided yields an intrinsic value of $177.83/share vs. the current price of $146.50, implying a 21% upside and a margin of safety that, on the surface, looks attractive. However, the DCF's core assumption — 2% FCF growth using the 3-year revenue CAGR of 1.72% — deserves serious scrutiny. The base FCF of $636M in FY2025 is unusually high (31.9% FCF margin vs. a 3-year average closer to 22–25%), partly reflecting capex of only $14M on $2B revenue (a fabless model with very low reinvestment). This creates an artificially inflated base from which to project. If I normalize FCF to, say, a 25% margin on current revenue, the base FCF would be ~$499M, and the intrinsic value per share drops materially — potentially to $130–140 range, near or below current price. The WACC of 10.27% is defensible for a semiconductor name with beta ~1.19 and no debt, though arguably slightly low given customer concentration risk and supply-chain exposure. The terminal growth rate of 2.5% is slightly above my preferred ceiling (equal to long-run nominal GDP) but not egregious. The bigger concern is the reverse-engineering exercise: at $146.50, the market is implying continued strong FCF generation with modest growth — achievable IF FY2025 FCF margins are sustainable. But ROIC of 17.3% comfortably exceeds the WACC of 10.27%, meaning growth genuinely creates value. Reinvestment intensity is minimal (capex $14M, sales-to-capital ratio extremely high for a fabless model), which is real for this asset-light structure. The story-to-numbers translation is coherent: dominant position in smartphone audio (Apple dependency), emerging PC/AI voice optionality (unquantified but real), clean balance sheet ($778M cash, zero debt adds ~$15/share). The narrative is honest and aligned with financials. However, 69% of DCF value sits in the terminal value, which is high sensitivity territory. Under a mild stress test — WACC +100bps to 11.27%, terminal growth 2.0%, normalized FCF base of $500M — intrinsic value falls to roughly $120–135, which is BELOW the current price. This eliminates the margin of safety entirely in the bear case. Revenue growth of only 1.72% CAGR over 3 years for a company trading at 3.7x sales and 11.6x FCF demands that the market is pricing in adjacency optionality (PC, voice-AI) that has not yet materialized in numbers. That is speculative value, not DCF-supported value. Net verdict: the base case DCF shows a modest margin of safety, but the assumptions are optimistic enough that sensitivity analysis erases it. The investment is not obviously overpriced (no heroic growth required), but it is not clearly cheap enough to call a 'pass' with discipline. A 'watch' reflects: real FCF quality, ROIC > WACC, defensible WACC, but insufficient margin of safety under conservative normalization, 69% terminal value dependence, and unquantified adjacency optionality baked into sentiment but not yet numbers.

08

Charlie Munger Quality

watch · 62

Cirrus Logic is a genuinely high-quality semiconductor business that sits comfortably within a definable circle of competence: it makes specialized mixed-signal chips (audio codecs, amplifiers, camera controllers) that are deeply embedded in premium smartphones. The moat is real but narrower and more concentrated than I prefer. ROIC of 17.3%, ROE of 19.7%, operating margins of 23%, and a FCF margin of nearly 32% on $637M of free cash flow all speak to a business that earns well above its cost of capital. The balance sheet is pristine — $778M net cash, zero debt, current ratio of 7.6x. Management appears disciplined: low capex ($14M on nearly $2B revenue — a fabless model that is capital-light and elegant), active buybacks, and candid acknowledgment of concentration risk. Owner earnings are real; FCF conversion is excellent. The DCF at $177/share intrinsic value vs. $146.50 current price suggests a 21% upside — not a screaming bargain but a fair price for a quality business. However, three issues prevent a full 'pass.' First, the customer concentration problem is severe and deliberately opaque — Apple is almost certainly >80% of revenue, and management policy prohibits discussion. This is not a moat; it is a dependency. If Apple decides to in-source audio chips (as it has done with modems, displays, and other components), Cirrus Logic's economics deteriorate rapidly. Second, the revenue CAGR of 1.7% over three years is inconsistent with a compounder. Growth depends heavily on Apple's iPhone cycle and content gains per device rather than organic market expansion. The adjacency strategy (PC, AI-voice, battery management) is promising but entirely unproven at commercial scale — 'sampling' and 'early wins' are not earnings. Third, the PEG ratio of 10.4x is a glaring warning: the market is pricing in far more growth than the historical record supports. At 17.8x earnings and 11.6x FCF the stock is reasonably priced for the quality, but the terminal value constitutes 69% of the DCF — heavily dependent on assumptions that Apple concentration risk could violate. The DCF's 2% FCF growth assumption may actually be generous given 1.7% revenue CAGR. I can understand this business. Management is not obviously stupid or dishonest. The economics are good. But the concentration in one customer who is also a sophisticated technology company with incentives to vertically integrate keeps me from paying up enthusiastically. Inversion test: how does this go wrong? Apple in-sources or shifts supplier, as they have done repeatedly. That is not a remote scenario — it is the base-rate behavior of Apple. That alone caps my enthusiasm.

09

Terry Smith (Fundsmith) Quality

watch · 62

Cirrus Logic passes several Fundsmith quality screens but fails others, leaving a mixed picture. On the positive side: the business is genuinely asset-light (capex of only $14M against $637M FCF, a capex/revenue ratio of ~0.7%), cash conversion is excellent (FCF of $637M exceeded net income of $414M, with FCF margin of 32% vs. net margin of 21%), the balance sheet is pristine (zero debt, $778M net cash, current ratio 7.65x), and ROCE/ROIC of ~17-20% is respectable. Operating margins at 23% and gross margins implied around 52% (per Q2 call) are healthy for a semiconductor company. However, several quality concerns temper enthusiasm. First, ROCE at ~17-20% is good but not exceptional by Fundsmith standards — Smith typically demands sustained pre-tax ROCE of 20%+ across the cycle; Cirrus has delivered this inconsistently, with FCF margins ranging from 5.5% (2021) to 32% (2025), suggesting cyclicality rather than the steady compounding Smith prizes. Second, and most critically, customer concentration in Apple represents an existential moat concern from a quality-investing perspective — the moat belongs partly to the customer, not Cirrus; Apple could in-source or switch suppliers, and management cannot even discuss the relationship publicly. This is the antithesis of the recurring, diversified, brand-driven revenue Smith seeks. Third, revenue CAGR of just 1.7% over three years reveals that this is a slow-growth, cyclically lumpy business tied to smartphone upgrade cycles — not the resilient, predictably compounding demand profile Smith rewards. Fourth, non-smartphone adjacencies (PC, battery/health, voice-AI) are nascent and unquantified, so the growth runway is speculative. On valuation: P/FCF of 11.6x is genuinely attractive for a quality business, and the DCF intrinsic value of $178 implies ~21% upside at $146.50. The bear-case DCF of $159 still represents an upside, providing some margin of safety. However, the DCF uses a 2% FCF growth rate — anchored to a 3-year revenue CAGR of 1.7% — which may be appropriate given concentration risk and episodic smartphone cycles. The terminal value represents 69% of enterprise value, which is high and sensitive to assumptions. Net-net: excellent cash economics and balance sheet quality, but insufficient moat independence, too much customer concentration, and insufficient demand predictability to earn a full Fundsmith pass.

10

Philip Fisher Growth

watch · 58

Cirrus Logic presents a mixed picture through the Fisher lens. On the positive side, the company exhibits strong margin quality (52.5% gross margin, 23% operating margin, 32% FCF margin), a disciplined R&D-driven product pipeline spanning audio codecs, camera controllers, and emerging PC/AI-voice products, and management that communicates candidly with measured tone — acknowledging Android softness and adjacency immaturity without promotional spin. The 5–6 year longevity of custom silicon demonstrates genuine technical differentiation and R&D productivity. However, the most critical Fisher criterion — sustained, above-average organic sales growth with a visible multi-year runway — is conspicuously weak. Revenue CAGR of only 1.72% over three years is far below what Fisher demands: this is barely inflation-level growth for a semiconductor company. The core smartphone business (dominated by one undisclosed customer, almost certainly Apple) is mature and cyclical, not expanding. Non-smartphone adjacencies (PC, AI-voice, battery health) are exciting strategically but contribute immaterial current revenue and carry unquantified timelines. The diversification narrative is real but early-stage — Fisher would want to see these new markets actually convert to revenue scale before awarding a growth premium. Customer concentration in one OEM also creates episodic rather than compounding growth dynamics. R&D spend appears productive (new codec nodes, camera controllers, SoundWire adoption) but the pipeline has not yet demonstrated it can move the revenue needle beyond the Apple dependency. Management depth and communication quality are solid positives. Overall, CRUS sits in Fisher's 'interesting but not yet ready' zone — a franchise with genuine R&D capability and margin quality, but lacking the growth velocity and market expansion evidence that Fisher requires before concentrating.

11

Bruce Greenwald Value

watch · 58

Cirrus Logic presents a genuinely interesting EPV case but falls into a 'watch' zone due to customer concentration opacity, uncertain moat durability, and a market price that is close to but not decisively below my EPV estimate. Let me work through the methodology rigorously.

EPV Calculation: I normalize operating earnings rather than use the provided DCF anchor. Reported operating income for FY2026 (period ending March 28, 2026) is $460M on $1,997M revenue — a 23.1% operating margin. Scanning the 5-year history, net margins ranged from 9.3% (FY2022, likely R&D/inventory cycle) to 20.8% (FY2026). The FCF margin trajectory is steep (5.5% → 31.9%), partly reflecting capex normalization (capex is remarkably low at $14M on $2B revenue — fabless model). I will use a normalized operating margin closer to the mid-cycle average, approximately 20%, giving normalized NOPAT ≈ $1,997M × 20% × (1 - 0.17 effective tax rate) ≈ ~$332M. Maintenance capex for a fabless chip designer is minimal (actual capex $14M vs. D&A likely ~$60-70M based on asset base), so I add back excess D&A to distributable earnings: adjusted distributable earnings ≈ $332M + ~$50M excess D&A ≈ ~$382M. Capitalizing at WACC of 10.27% (per the provided model): EPV of operations = $382M / 0.1027 ≈ $3,720M. Adding net cash ($778M): EPV equity ≈ $4,498M. Dividing by ~50.6M shares: EPV per share ≈ $89. At the current price of $146.50, the stock trades at a ~65% premium to my EPV — i.e., the market is pricing in substantial growth. A more generous normalization using the current-year margin (23.1%) yields: NOPAT ≈ $1,997M × 23.1% × 0.83 ≈ $383M, plus excess D&A ~$50M = $433M distributable; EPV ≈ $433M / 0.1027 ≈ $4,216M equity, or ~$99/share. Still a ~48% premium to EPV. Even a bull-case normalization (25% operating margin) yields EPV ~$110/share — still below $146.50.

Asset Reproduction Value: Cirrus Logic is fabless, so tangible assets are modest. Total assets $2.46B, but much is goodwill, intangibles, and cash. Book equity is $2.1B (~$41/share), largely cash ($778M) plus net working capital and IP. The true reproduction cost for a competitor to replicate CRUS's position — deep custom silicon relationships with Apple, multi-year audio codec expertise, 22nm custom process knowledge, engineering talent, and qualification cycles — is substantially higher than book. I estimate reproduction value at $70-100/share (cash at $15/share + $55-85/share for IP, talent, customer relationships). This puts EPV ($89-110) comfortably above reproduction value ($70-100), which is the right pattern: it signals a genuine franchise. The EPV/asset gap is positive, supporting the moat hypothesis.

Moat Assessment: EPV > reproduction value confirms a franchise exists, but the nature and durability of the moat requires scrutiny. Customer captivity is real: custom silicon for Apple's audio codec involves years of co-development, qualification, and switching costs — Apple cannot swap suppliers without major disruption. Economies of scale within the audio niche are present (Cirrus is the dominant supplier). Proprietary technology (22nm codec, camera controllers, boosted amplifiers) provides some protection. However: (1) The moat is heavily concentrated on ONE customer (Apple), whose bargaining power is immense; (2) The franchise value is contingent on Apple not vertically integrating audio (Apple has a track record of internalizing chips — e.g., cellular modems); (3) PC and battery adjacencies are nascent with no proven moat extension yet. The moat is real but narrow and dependent.

Margin of Safety: At $146.50, I'm paying a 33-65% premium to EPV — the opposite of a margin of safety. The market price is essentially the bear case of the provided DCF ($159) or requires ~2-5% growth indefinitely within a moat to be justified. That may be true, but it is speculative growth pricing, not EPV investing. The DCF's terminal value constitutes 69.2% of enterprise value — precisely the kind of far-future assumption I distrust most. The 21% DCF upside ($177 intrinsic) depends on 2% FCF growth perpetually, which is essentially the minimum case — yet the stock is only 21% below even that conservative DCF estimate.

Other Red Flags: The FY2022 net margin collapsed to 9.3% — suggesting the current margins may be somewhat cyclically elevated or benefiting from product mix. Revenue CAGR of 1.7% over 3 years confirms this is not a high-growth business; the FCF improvement is largely margin and working capital driven. Insider selling (EVP liquidating shares via 10b5-1) is mild noise but not alarming. The GlobalFoundries wafer commitment through 2026 is a supply-chain overhang not fully quantified.

12

Ray Dalio Risk

watch · 52

Cirrus Logic presents a mixed picture through the Dalio macro lens. On the balance sheet, it is genuinely fortress-like: zero debt, $778M net cash, current ratio of 7.6x, and 31.9% FCF margin generating $637M in free cash flow against essentially no financing obligations. This passes the debt-cycle and self-funding test with flying colors — the company can survive a credit contraction, a deleveraging episode, or a sustained higher-for-longer rate environment without distress or capital markets access. That is a real strength in a risk-parity framework. However, the macro regime analysis reveals concentrated single-regime dependence that Dalio would penalize heavily. CRUS is overwhelmingly a falling-inflation, rising-growth, consumer-discretionary semiconductor play tethered to Apple's smartphone upgrade cycle. In stagflation (rising inflation + falling growth), consumer electronics demand compresses sharply, Apple's premium pricing power erodes at the margin, and fabless semiconductor suppliers face input cost pressure (wafer costs, energy, logistics) with limited pass-through leverage since pricing is largely negotiated with a single dominant OEM. In a deflationary bust, smartphone unit volumes historically decline meaningfully (discretionary spend, credit-dependent consumer in emerging markets, inventory correction cycles). The 3-year revenue CAGR of only 1.7% across the recent history already reflects this episodic, cycle-dependent nature. The PEG of 10.4x is essentially pricing permanent smooth growth into a cyclical business. Customer concentration risk — overwhelmingly Apple, per widely understood but deliberately opaque management policy — creates massive single-counterparty macro exposure; if Apple cycles down or loses premium market share in a slowdown, CRUS has no meaningful revenue diversification shield. PC and non-smartphone adjacencies remain immaterial revenue contributors today. The DCF intrinsic value of $177.83 (21% upside at $146.50) uses only 2% FCF growth — arguably too conservative given recent FCF surge — but the WACC of 10.27% assumes mid-cycle conditions; in a stagflationary regime with 10-year yields at 5-6%, the appropriate WACC rises materially and the terminal value (69.2% of enterprise value) compresses significantly. The foundry concentration risk (GlobalFoundries wafer commitment through 2026, TSMC capacity constraints from AI) represents a supply-side inflation pass-through vulnerability. Geographic revenue concentration in a single-currency, US-dominated supply chain with geopolitical tail risk (Taiwan) is a Dalio-style hidden tail risk. Net assessment: the balance sheet is exemplary and earns meaningful credit, but the regime dependence, customer concentration, and valuation sensitivity to terminal assumptions in a rate-shift scenario limit the score to watch territory.

13

Stanley Druckenmiller Risk

watch · 52

CRUS presents a mixed picture through my lens. The directional thesis has genuine merit — the Q2 FY2026 record $561M revenue, expanding FCF margins (31.9% vs 22.2% three years ago), and emerging PC/AI-voice catalyst are real. The tape confirmed the thesis through April 2026 when CRUS hit 52-week highs. But right now the stock has pulled back from $180 to $146 — it is trading 18.8% BELOW its 52-week high while the narrative is still bullish. That tape disagreement is a critical red flag for me. The second derivative of revenue growth is also soft: 3-year revenue CAGR is only 1.72% — this is not an accelerating earnings story, it is a margin expansion story. Margin expansion is good but it does not give me the forward earnings INFLECTION I need to size up aggressively. The Apple concentration risk means I cannot independently model the earnings trajectory with confidence — the key variable is opaque. The PC adjacency, AI-voice products, and battery IP are in sampling/early-ramp stage with IMMATERIAL current revenue contribution; these catalysts are 12-24+ months from being earnings-moving. The liquidity/macro backdrop is ambiguous — no clear Fed easing tailwind that would compress the discount rate and re-rate the multiple. The FCF is exceptional (price-to-FCF of 11.6x is genuinely attractive) but I am not in the business of buying cheap — I am in the business of buying what is going UP. For me to size up, I need: (1) price reclaiming and sustaining above the $160-170 range showing the tape is back on my side, (2) evidence of forward estimate REVISIONS upward driven by PC design win revenue materializing, and (3) a clearer macro/Fed catalyst. Until then, this is a watch — high-quality business, interesting setup, but not yet the right moment to concentrate.

14

Benjamin Graham Value

watch · 52

Cirrus Logic passes several of Graham's quantitative tests but falls short of the strict margin-of-safety requirement that defines a true defensive buy. On the positive side, the balance sheet is genuinely strong: current ratio of 7.65 (well above the 2x minimum), zero long-term debt, $778M in net cash, and stockholders' equity of $2.1B. Earnings have been positive in every year of the 5-year history provided, with net income rising from $176M (FY2022) to $414M (FY2025), demonstrating the stability Graham prizes. ROE of 19.7% and ROIC of 17.3% show productive capital deployment. The trailing P/E of 17.84x is modestly above Graham's defensive ceiling of 15x, but the earnings yield of 6.96% compares favorably to high-grade bond yields (approximately 4.5-5%), providing a reasonable spread. Free cash flow generation is exceptional — $637M on $2B revenue, a 31.9% FCF margin — and the price-to-FCF of 11.6x is actually quite attractive by Graham's cash-earnings yardstick. The DCF intrinsic value of $177.83/share suggests the stock trades at roughly a 18% discount to base-case intrinsic value, which is meaningful but falls well short of Graham's preferred one-third margin of safety (~$119 or below would qualify). The critical failure on Graham's asset-based tests: P/B of 3.51x is well above his P/E x P/B = 22.5 threshold (17.84 x 3.51 = 62.6, versus the 22.5 ceiling), and the stock does not remotely approach net-net territory. Net current assets are approximately $1.225B (current assets $1.409B minus all liabilities — total assets $2.456B minus equity $2.103B implies ~$353M total liabilities, so NCAV ≈ $1.409B - $353M = ~$1.056B), or roughly $20.9/share against a $146.50 price — stock trades at 7x NCAV, far from net-net. The 3-year revenue CAGR of 1.7% is anaemic, barely meeting Graham's one-third growth test over a decade at this rate. No dividend is paid, which is a notable negative for Graham's defensive criteria. Customer concentration in Apple (though undisclosed) represents a qualitative risk that Graham would view as an unstable earnings foundation. The company passes as an established, profitable, debt-free enterprise with genuine earnings power, but the price does not offer adequate margin of safety on either earnings or asset bases by strict Graham standards.

15

Seth Klarman Value

watch · 52

Cirrus Logic presents a genuinely interesting balance sheet and cash-flow profile, but falls short of a Klarman-quality margin of safety at current prices. The DCF yields an intrinsic value of ~$178/share against a $146.50 current price, a 21% upside — modest by Baupost standards and dependent on a 2% FCF growth assumption that underweights the real risk: Apple customer concentration. The company is fundamentally high-quality (32% FCF margins, zero debt, $778M cash, current ratio 7.65x, ROIC 17%), but quality alone does not create a margin of safety. The bear DCF scenario at $159 is only 9% above current price, meaning even the conservative scenario offers little cushion. From a liquidation/asset perspective, stockholders equity of $2.1B against a $7.4B market cap yields price-to-book of 3.51x — no net-net or asset-coverage comfort here. The business is fabless, so tangible assets are minimal; the real value is in IP and customer relationships, both difficult to appraise conservatively and heavily dependent on a single opaque customer. Normalized FCF of $636M at 11.6x price-to-FCF is genuinely cheap for a semiconductor company, and if one credits the full cash position ($778M, ~$15/share), the ex-cash P/FCF is roughly 10x — reasonable but not deeply discounted. The three-pillar diversification strategy (PC, voice-AI, battery health) is promising but contributes immaterial current revenue; valuing those options requires optimistic growth assumptions that I would not embed in a conservative appraisal. There is no special-situation catalyst, no forced selling, no distress — this is a well-followed, well-loved semiconductor name trading near 52-week highs with bullish retail sentiment. That is precisely the environment where Baupost holds cash rather than chases. The insider selling (EVP liquidating ~$850K in June under 10b5-1) is minor but consistent with a stock that has run hard. Supply chain risks (GlobalFoundries wafer commitments, TSMC AI capacity competition, geopolitical logistics) are real and unquantified. CHIPS Act exclusion as a fabless player is a structural disadvantage. The stock warrants watching for a better entry — a 30-40% decline to the $90-105 range would begin to create genuine margin of safety against normalized FCF and would represent a bear-scenario overshoot worth acting on.

16

Peter Lynch Growth

watch · 52

Cirrus Logic is a stalwart/slow-grower semiconductor company — not a fast grower — and must be judged on those terms. Revenue CAGR of just 1.7% over 3 years and a PEG of 10.37 are the two numbers that immediately concern me. The Lynch framework is unambiguous: you never want to pay a P/E far above the growth rate, and a PEG of 10+ is a flashing red light regardless of how good the business is. The stock's P/E of ~17.8x is being compared against anemic 1.7% revenue growth — that's the core problem here. Now, there are genuinely good things: the balance sheet is pristine ($778M net cash, zero debt, current ratio 7.6x), FCF generation is excellent ($636M, 31.9% FCF margin), and the company's P/FCF of 11.6x is actually quite attractive for a high-quality chip maker. The business is understandable — Cirrus makes audio chips that go inside Apple iPhones, which is as close to a 'product you can see in everyday life' as semiconductors get. But the Apple concentration is my biggest story-clarity concern: management won't even discuss their largest customer by name, which means the 'story' has a black box at its center. That's not how Lynch investing works — I want to explain the full story in two sentences. The three-pillar diversification strategy (PC, voice-AI, battery health) is promising but unproven; these are adjacencies with 'immaterial current revenue' by management's own admission. I like the PC mainstream consumer win and the AI/voice PC product sampling, but these are catalysts 2-3 years out, and I don't pay today's price for tomorrow's maybe. The FCF math is appealing and DCF intrinsic value of $177.83 (21% upside) provides some comfort at $146.50. But for Lynch, the earnings growth rate is the anchor — and 1.7% revenue CAGR with a 10x PEG means I'm being asked to overpay for a stalwart. The right stalwart buy is 30-50% gain potential at a fair P/E, and the DCF suggests that's achievable, but I need faster earnings growth to justify holding long-term. Insider selling (EVP liquidating ~$850K in June) adds a minor negative signal, though the 10b5-1 pre-plan context reduces alarm. This is a watch — attractive FCF and balance sheet quality, but the growth engine isn't firing fast enough to justify the PEG, and customer concentration opacity violates my 'understand the story' rule.

17

Howard Marks Risk

watch · 52

From a Marks risk-first lens, Cirrus Logic presents a genuinely mixed picture. The balance sheet is pristine — zero debt, $778M cash, current ratio 7.65x, strong FCF generation ($636M, 31.9% FCF margin) — which eliminates the capital structure fragility I fear most. That is the core safety net. However, the risk-adjusted investment case is complicated by what appears to be a sentiment pendulum that has already swung meaningfully toward optimism. The stock sits at its 52-week high ($146.5 = 52w high per price data, though 52w high is listed as $180.42 elsewhere — the data shows conflicting signals suggesting recent price decline from peak), with retail chatter uniformly bullish, GuruFocus flagging overvaluation, and analyst consensus firmly in the Buy camp. This is not the environment where I find the best risk/return: the crowd is already on the right side. The DCF shows only 21.4% upside to intrinsic value ($177.83 base case), with that intrinsic value depending on a 2% FCF growth assumption that essentially flatlines a business with only 1.7% revenue CAGR over 3 years. The terminal value represents 69.2% of enterprise value — an uncomfortable dependence on a Gordon-growth perpetuity assumption in a semiconductor business with concentrated customer risk. The bear DCF case is $159.36, only 8.8% above current price, meaning the margin of safety is razor thin against a pessimistic but plausible scenario. On second-level thinking: what does the crowd believe? That Apple dependency is manageable, PC diversification is credible, margins are sustainable, and the fabless model is durable. The contrarian question is whether any of that is wrong — and frankly, with Apple concentration undisclosed but widely understood to dominate, a single OEM cycle disruption could materially impair revenue. The 10-K flagged onerous GlobalFoundries wafer commitments through 2026 and supply chain risk, neither quantified. The 4% YoY revenue growth in a 'record' quarter is not the growth profile that justifies elevated expectations. The PEG of 10.37 is alarming — it prices in growth that has not materialized. Insider selling under 10b5-1 (EVP liquidating ~$650K+ in June) is a mild negative signal. On positives: P/FCF of 11.6x is genuinely inexpensive for quality; the operating margin of 23% and ROE of 19.7% are solid; no debt means no forced selling or covenant risk in a downturn. The asymmetry, however, is the core problem: limited downside protection in price (bear case only 8.8% below current) with upside also constrained (21% to base DCF). This is a good company at a fair-to-full price, not a beaten-down asset with embedded pessimism. I want to be paid to take risk; here, I am not being paid enough for the concentration and cyclicality risks embedded in this business.

18

Walter Schloss Value

avoid · 28

Cirrus Logic fails virtually every Schloss deep-value criterion. The stock trades at 3.51x book value and 3.7x sales — far above the near-or-below tangible book standard that anchors my entire method. Price is at its 52-week HIGH ($146.50 equals the 52-week high per the fact base), not a multi-year low, which is the opposite of what I look for. I buy what is beaten down and unloved; this stock has been rising and is at peak pricing. Tangible book per share works out to roughly $41.50 (stockholders' equity $2.103B / 50.58M shares), and the stock at $146.50 trades at 3.5x that figure — nowhere near the discount to hard assets that provides my required margin of safety. Yes, the balance sheet is clean: zero long-term debt, current ratio of 7.65, $778M cash, and the company is asset-light with minimal capex ($14M). But 'asset-light' is precisely the problem for my approach — there are no heavy tangible assets (plant, real estate, inventory at scale) to appraise as downside protection independent of the earnings story. The business value lives almost entirely in intangible competitive position (custom silicon design wins, Apple relationship) and earnings power — exactly the kind of thesis I avoid because it requires faith in projections rather than verification against the balance sheet. P/E of 17.8x is not egregious for a quality compounder, but that is not my bar. FCF yield of ~8.6% (price-to-FCF of 11.6x) is decent but the asset anchor simply isn't there. The DCF intrinsic value of $177.83 suggests 21% upside, but DCF is an earnings/growth model, not an asset appraisal — I don't anchor to it. Insider activity is negative: EVP Jeffrey Baumgartner sold shares twice in June 2026 under a 10b5-1 plan, totaling several thousand shares. No insider buying reported. Revenue CAGR of only 1.7% over three years confirms this is a slow-growth business, which is fine, but slow growth combined with a rich price-to-book leaves no statistical bargain here. The business itself appears solidly run with excellent FCF margins (31.9%) and a pristine balance sheet — I have nothing against the company's quality. But Schloss Associates was never in the business of paying 3.5x book for quality; we paid below book for ordinary businesses and let mean reversion do the work.

Want this on a name you own?Send a ticker; we convene the full council on it.

Request Analysis →