Murata Manufacturing Co., Ltd. (6981.T) — Graham Screen

Layer: L9-MLCC-RF | Date: 2026-05-28 | Analyst: Claude (claude-sonnet-4-6) Exchange: Tokyo Stock Exchange | Sector: Technology / Electronic Components


PRICE CHECK (shr-020 pre-screen)

Metric Value
Current price JPY 8,538
52-week high JPY 8,899 (2026-05-26)
52-week low JPY 2,089 (2025-06-23)
Distance from 52w high -4.1%
YTD return (from 2026-01-05 open JPY 3,299) +137%
3-month surge (Mar 2026 JPY 3,409 → May 2026) +151%
Market cap JPY 15.5T (~EUR 95B)

ABBREVIATED THRESHOLD CHECK: YTD > 80% = YES (+137%). Within 10% of 52w high = YES (-4.1%).
Per protocol, abbreviated screen is triggered. However, given thesis question asks specifically about MLCC commoditization vs AI rerating — FULL SCREEN executed to provide complete analytical picture.


BACKGROUND

Murata Manufacturing is the global #1 producer of MLCCs (multilayer ceramic capacitors), holding ~40% world market share. Revenue mix: ~45% capacitors (MLCC), ~22% communications (RF filters, modules), ~18% inductors/ferrites, ~15% other. FY ends March 31.

Key subsidiaries: Murata Electronics (components), acquired Peregrine Semiconductor (2014, RF), TDK connectivity. Customer concentration in Apple (est. 15-20% of revenue), with Greater China as ~48% of sales (assembly hub for global OEMs, not end-demand proxy).


1. PRICE ACTION AND RERATING ANATOMY

The 4x move from JPY ~2,100 (May 2025) to JPY 8,538 (May 2026) with NO stock split is almost entirely multiple expansion, not earnings growth:

| | FY2025 Trough (May 2025) | Today (May 2026) | |--|--|--| | Price | JPY 2,089 | JPY 8,538 | | Trailing EPS | ~JPY 128 | JPY 127 | | Trailing P/E | ~16x | 67x | | Net income (FY26) | — | JPY 233.9B (+0% YoY) |

P/E expanded 312% while earnings were flat. The market re-priced from "cyclical consumer electronics compounder at trough" (16x) to "AI infrastructure secular compounder" (67x). This is the key thesis question: is the new multiple justified, or is it narrative overshoot?

Monthly price ladder (yfinance):


2. REVENUE AND EARNINGS TRAJECTORY

Annual (FY ends March 31)

Fiscal Year Revenue (JPY B) Operating Income (JPY B) Net Income (JPY B) Op Margin
FY2022 1,812.5 424.1 314.1 23.4%
FY2023 1,686.8 292.9 243.9 17.4%
FY2024 1,640.2 215.4 180.8 13.1%
FY2025 1,743.4 279.7 233.8 16.0%
FY2026 (actual Apr 30) 1,830.6 281.8 233.9 15.4%
FY2027 guidance 1,960.0 380.0 293.0 19.4%

Critical observation: FY2026 net income flat at JPY 233.9B vs FY2025 (233.8B). The stock is pricing in FY2027's recovery, not current earnings. Trough was FY2024 (180.8B) — the cycle is real but recovery to FY2022 peaks (314.1B) is still not forecast until FY2028+.

Quarterly Margin Recovery

Quarter Revenue (JPY B) Op Income (JPY B) Op Margin
Q2 FY25 (Sep 2024) 461.8 91.8 19.9%
Q3 FY25 (Dec 2024) 448.0 76.0 17.0%
Q4 FY25 (Mar 2025) 411.9 45.5 11.1% — cyclical trough
Q1 FY26 (Jun 2025) 416.2 61.6 14.8%
Q2 FY26 (Sep 2025) 486.6 103.5 21.3% — inflection

The Q2 FY26 (Sep 2025) operating margin of 21.3% demonstrates the leverage inherent in Murata's business — volume recovery + MLCC pricing power = sharp margin expansion.


3. MLCC PRICING TRAJECTORY (KEY THESIS ITEM A)

Pricing environment has fundamentally shifted from commoditization to allocation:

Bifurcated market: Consumer-grade commodity MLCCs remain flat (no panic buying, oversupplied). Premium AI server and automotive MLCCs are under genuine allocation. This bifurcation is exactly what favors Murata's portfolio mix — their consumer exposure is declining share and margin, while AI/auto is growing share and receives pricing power.

Commoditization assessment: The MLCC commoditization thesis was accurate for 2019-2024 for commodity capacitors. It does NOT apply to:

  1. High-capacitance, high-voltage AI server MLCCs (technically difficult, few qualified suppliers)
  2. Automotive AEC-Q200 grade (long qualification cycles, Murata holds dominant share)
  3. RF passives and BAW/SAW filters (patent-protected process technology)

Conclusion: Commodity MLCC pressure is priced into the trough already. The re-rate reflects justified recognition that Murata's value-add is in the non-commodity segments.


4. AI SERVER PCB CONTENT PER BOARD (KEY THESIS ITEM B)

Murata's own mid-term strategy (updated post-FY2026 results):

Murata also entering VPD (Vertical Power Delivery) power modules for AI servers — mass production begins FY2027, targeting JPY 50B revenue by FY2027-end. This is a new product category adjacent to MLCCs that leverages their ceramic substrate expertise.

MLCC CAGR for AI server segment: Murata raised forecast to 30% CAGR (per Investing.com, May 2026). This is their most recent disclosed estimate.

Countervailing factor: AI server CapEx cycles can compress rapidly (hyperscaler digestion periods). The 2024 inventory correction in standard semis could recur in MLCC if hyperscaler CapEx plateaus. Management's 3.3x to 2030 forecast assumes sustained datacenter investment growth — a reasonable but not guaranteed assumption.


5. SMARTPHONE EXPOSURE AND CHINA DEMAND (KEY THESIS ITEM C)

Revenue Geography

The China 48% overstates genuine China-consumption risk — a large portion is Apple's Foxconn/TSMC supply chain. Tariff risk is real but partially offset by supply chain diversification (Apple's India shift benefits Murata's India exposure).

Smartphone Mix Shift

In Q3 FY2026 (nine months ended Dec 2025), revenue split:

This reflects Apple's continued RF module insourcing (RFICs moved in-house) and competitive pressure from Qorvo/Skyworks on Android. Murata's smartphone revenue is under structural pressure from RF module insourcing — partially offset by MLCC volume recovery.

Net assessment: Smartphone is a declining-mix, margin-compressing segment for Murata. The investment thesis is precisely that AI server/automotive is replacing smartphone as the earnings driver, at higher ASPs and margins. The data supports this transition being underway.


6. SAMSUNG / YAGEO COMPETITION (KEY THESIS ITEM D)

Samsung Electro-Mechanics (SEMCO):

Yageo (2327.TW) / Walsin:

Competitive moat assessment: Murata's moat is real but not impenetrable:

  1. Ceramic material science IP: 70+ years of proprietary dielectric formulations. Genuine barrier.
  2. Process scale: thin-layer MLCC down to 0.01mm — Samsung is 2-3 years behind in layer thinness
  3. RF/SAW/BAW filters: Patent-intensive, semiconductor-adjacent. Hard to replicate quickly.
  4. Qualification lock-in: Automotive AEC-Q200, aerospace customers take 2-3 years to qualify alternates

Risk: China state-sponsored MLCC producers (Fenghua Advanced Technology, CCECC) — receiving government subsidies to close the gap. Still 5+ years behind on premium grades.


7. GRAHAM 7 FILTERS

Applied to FY2026 actuals (year ended March 31, 2026). Note: shr-017 guidance on distinguishing quality premium from cyclicality.

Filter Criterion Value Result
F1: Adequate size Revenue > JPY 200B JPY 1,830.6B PASS
F2: Financial condition Current ratio > 2.0x 5.4x PASS
F3: Earnings stability Positive earnings 10yr Yes (all years positive; FY2024 trough at JPY 180.8B still strongly positive) PASS
F4: Dividend record Uninterrupted 20yr Continuous since 1990s; JPY 65/sh FY26 vs JPY 60/sh FY25 PASS
F5: Earnings growth +33% over 10yr FY2016 EPS ~JPY 200 → FY2026 JPY 128 = -36% (MLCC downcycle) FAIL
F6: Moderate P/E < 15x trailing; < 15x avg Trailing 67x, FY27 forward 53x FAIL
F7: Moderate P/B < 1.5x; P/E×P/B < 22.5 P/B 6.0x; P/E×P/B = 402 FAIL

Graham Filter Score: 4/7

F5 note (shr-017 application): The F5 failure is a cyclical trap risk in reverse — EPS declined from the FY2022 cycle peak (JPY ~172) to the FY2026 trough (JPY 127). This is cyclicality suppressing trailing metrics, not permanent earnings deterioration. FY2022 earnings were cycle-peak (AI/5G/EV build-out and post-COVID double-ordering), FY2024 was demand destruction trough, FY2027 is recovery. F5 FAIL is cyclicality-driven, not value-destructive.

F6/F7 note: Both fail substantially and are NOT cyclicality artifacts — they reflect genuine quality premium pricing. At FY2022 peak earnings (JPY 314B net income → ~JPY 172 EPS), the stock traded at JPY 2,700-3,200 = P/E of ~16-19x. Even at peak earnings, Murata has historically commanded a premium. The current 67x trailing P/E is AI-narrative re-rate, not normative Murata valuation.

Murata has NEVER been a Graham defensive stock. It is a quality compounder that occasionally trades at trough valuations (16x in May 2025) when cycles compress. The current 67x is multiple expansion driven by AI narrative capture.


8. GRAHAM INTRINSIC VALUE SENSITIVITY TABLE

Per shr-003 and shr-012. AAA bond yield proxy: 4.4% (US AAA equivalent).

8a. Trailing EPS (FY2026 actual: JPY 127.47)

Growth Rate Graham IV (JPY) vs Current 8,538 MOS
g = 0% 1,083 -87.3% DEEPLY NEGATIVE
g = 3% 1,848 -78.4% DEEPLY NEGATIVE
g = 5% 2,358 -72.4% DEEPLY NEGATIVE
g = 7.5% 2,996 -64.9% DEEPLY NEGATIVE
g = 10% 3,633 -57.5% DEEPLY NEGATIVE

8b. FY2027 Management Guidance EPS (JPY 161 = JPY 293B / 1,820M shares)

Growth Rate Graham IV (JPY) vs Current 8,538 MOS
g = 0% 1,368 -84.0% DEEPLY NEGATIVE
g = 3% 2,334 -72.7% DEEPLY NEGATIVE
g = 5% 2,978 -65.1% DEEPLY NEGATIVE
g = 7.5% 3,783 -55.7% DEEPLY NEGATIVE
g = 10% 4,588 -46.3% DEEPLY NEGATIVE
g = 15% 6,198 -27.4% NEGATIVE
g = 20% 7,808 -8.6% NEAR FAIR VALUE

8c. FY2028 Consensus EPS (JPY 237 — likely FY2028 forward estimate via yfinance)

Growth Rate Graham IV (JPY) vs Current 8,538 MOS
g = 0% 2,017 -76.4% DEEPLY NEGATIVE
g = 5% 4,389 -48.6% DEEPLY NEGATIVE
g = 10% 6,762 -20.8% NEGATIVE
g = 15% 9,135 +7.0% SLIM POSITIVE
g = 20% 11,494 +34.6% POSITIVE (but g=20% required)

shr-003 implied growth gap analysis:

The gap between trailing (29%) and 2-yr forward (14%) implied growth is 15 percentage points. Per shr-003, this is precisely the "profitability inflection" pattern — the market is pricing acceleration from the FY2024 trough to FY2027+ recovery, then expects a lower long-run growth rate. The question is whether the FY2028+ sustainable growth rate (implied ~14%) is achievable given AI server MLCC CAGR of 30% and automotive MLCC growth — plausibly yes, but the stock already prices it.

Graham IV conclusion: Graham's formula yields negative MOS at all realistic EPS and growth assumptions except using FY2028 EPS with 15%+ sustained growth. This is solidly a growth stock under shr-004 framing. The Graham IV tool is being used here as a sensitivity framework, not a defensive filter.


9. VALUATION SUMMARY

Metric Value Context
Trailing P/E 67x vs historical 16-25x normalized range
Forward P/E (FY27 guidance) 53x vs semi-component peer 25-35x
Forward P/E (FY28 consensus) 36x Approaching reasonable if growth sustains
P/B 6.0x Fully premium; JPY 1,494/share book value
EV/EBITDA (FY25 last full year) 31x vs peers ~15-20x
P/FCF 60x Based on FY26 FCF JPY 259B
Net cash JPY 565B Fortress balance sheet
Dividend yield (forward JPY 70) 0.82% Minimal; buyback 1% of mktcap
Consensus analyst PT JPY 5,203 -39% vs current — but set pre-rally, stale

Goldman Sachs added 6981 to APAC Conviction List post-results (May 2026). The consensus PT of 5,203 reflects a snapshot before the Apr-May 2026 rally; the actual analyst community is mid-upgrade cycle.


10. BALANCE SHEET AND CAPITAL RETURNS (FY2026 actual, March 31, 2026)

Item Value
Total assets JPY 3,028B
Stockholders equity JPY 2,581B
Total debt JPY 60B
Cash & equivalents JPY 625B
Net cash position JPY +565B
Current ratio 5.4x
Long-term debt JPY 1.7B (negligible)
R&D spend (FY25) JPY 149B (8.6% of revenue)
Capex (FY26) JPY 193B
FCF (FY26) JPY 259B
Share buyback (announced May 2026) JPY 150B (75M shares, by Jan 2027)
Annual dividend FY27 JPY 70/share (+JPY 5 YoY)

Capital allocation is conservative and prudent. The JPY 150B buyback is 1.0% of current market cap — symbolic rather than transformative. R&D at 8.6% of revenue is essential for maintaining process leadership. Capex-heavy model (10.5% of revenue) reflects manufacturing complexity.

F2/F3/F4 PASS — the balance sheet is pristine. Net cash of JPY 565B with minimal long-term debt. Earnings positive every year (trough FY2024: JPY 180.8B). Dividend uninterrupted and growing.


11. SHR-020 RED FLAG CHECK (Pre-Buy Diligence)

Applied as of 2026-05-28. This section answers whether the thesis has changed since the supply chain "quality compounder at trough multiple" observation.

Category Signal Flag
Price action +137% YTD, within 4% of 52w high, 3-month surge +151% RED
Valuation Trailing P/E 67x, P/B 6x, P/FCF 60x — all historically peak for Murata RED
Earnings (recent) FY2026 net income flat (+0% YoY), guidance missed consensus AMBER
Guidance FY2027 +25% NI, +7% revenue — genuine recovery but partially missed AMBER
MLCC pricing 15-35% hikes Apr 2026, oligopoly discipline intact, no undercutting GREEN
AI server content 10-20x smartphone content, 30% CAGR forecast, VPD power modules new GREEN
Competition Samsung/Yageo following hikes; Chinese producers 5yr+ behind premium GREEN
Balance sheet Net cash JPY 565B, current ratio 5.4x, minimal debt GREEN
Insider activity JPY 150B buyback (token 1%), no open-market C-suite purchases AMBER
China risk 48% Greater China (assembly hub risk, not demand proxy) AMBER
Shr-003 gap Trailing implied g=29% vs FY28 implied g=14% — 15pp gap AMBER

Net assessment: 2 RED, 4 AMBER, 4 GREEN. The REDs are both valuation/momentum-driven. The GREENs are structural thesis indicators. This is NOT a Graham defensive entry — the thesis has fundamentally rerated.

The "quality compounder at trough multiple" thesis was VALID at JPY 2,089 (May 2025) at 16x P/E. It is EXPIRED at JPY 8,538 at 67x P/E. The multiple has already done its work.


12. PORTFOLIO FIT ANALYSIS

At current valuation (67x trailing, 53x FY27, 6x P/B), Murata at JPY 8,538 does not fit the EUR 3,000 satellite Graham value mandate. The satellite is designed for companies with positive margin of safety (shr-016, shr-018). Murata at current price has NEGATIVE margin of safety at every realistic Graham IV computation.


VERDICT

VERDICT: RED — DO NOT ADD TO SATELLITE PORTFOLIO AT CURRENT LEVELS

Why RED and not YELLOW:

  1. Thesis has already executed: The "quality compounder at trough multiple" observation was correct — in May 2025 at JPY 2,089 and P/E ~16x. The subsequent 4x re-rate has fully captured the thesis and then some. There is no discount remaining to exploit.

  2. Valuation is extreme by any framework: Graham IV requires 20%+ sustained growth at FY2027 EPS just to justify current price. Implied trailing growth rate of 29% is 2x Murata's historical long-run growth. Neither Graham defensive (4/7 filters, F6/F7 fail massively) nor Graham Growth Addendum (forward P/E 53x means implied growth >> achievable growth — shr-004 says buy only when implied ≤ achievable).

  3. Narrative-driven, not fundamental-driven: 100% of the 4x price move represents P/E expansion (trailing earnings were flat). The narrative (AI server MLCC) is real and structural — but the price already fully prices a heroic growth recovery, consensus is in upgrade mode, and the stock trades near all-time highs. No margin of safety.

  4. MLCC commoditization thesis is dead for premium tier — but already in the price: the market has recognized AI server MLCC as non-commodity, hence the rerating. Being "right about the thesis" now generates zero alpha because everyone is right.

  5. Satellite mandate mismatch: EUR 3,000 satellite is Graham value (shr-018: positive EV/EUR deployed). At 67x P/E with negative Graham IV at every realistic scenario, this is a growth bet requiring 20-25% CAGR justification, not a margin-of-safety value entry.

What would change the verdict:

A re-entry opportunity exists if/when:

Monitor at:


VERDICT LINE: RED — Murata is a confirmed quality MLCC/RF compounder with a valid AI-server secular growth thesis, but the "trough multiple" entry window closed 12 months ago. Current 67x trailing P/E, 53x FY27 P/E, and -39% vs consensus PT means the satellite gets zero margin of safety. WATCH at JPY 4,500 for re-entry opportunity on any AI CapEx pause or earnings-reset drawdown.


Data sources: yfinance (price, fundamentals, dividends), Murata IR (FY2026/FY2027 results), BigGo Finance, TrendForce, Digitimes, Investing.com, TradingView/Quartr. All JPY figures. EUR/JPY ~163 as of 2026-05-28.