Littelfuse, Inc. (NASDAQ: LFUS) — A 99-Year-Old Fuse Company Repriced as a Data-Center Power Story

Report date: 2026-07-25 · Price reference: $404.86 (close, 2026-07-24) · Market cap: ~$10.1bn Standing note: The analysis below contains no investment recommendation and no price target. Valuation is discussed only as embedded expectations and scenarios. The single, deliberate exception is the Claude's Take block immediately below, which is clearly marked as the author’s own opinion.

⚡ Claude’s Take

This block is the author’s own subjective opinion. It is general information, not investment advice, and no part of it is a recommendation to buy or sell any security. The detailed analysis that follows carries no position and no price target.

Verdict: HOLD — and AVOID adding here at ~$405. Accumulate on weakness in the $270–330 zone (roughly 14–17x a normalised ~$19–20 of 2030 EPS, or ~11–13x 2030 EV/EBITDA).

Tag: “The fuse business is good. The multiple is priced off the part of the business they just wrote off.”

Littelfuse contains a genuinely good company inside a mediocre one. The legacy circuit-protection franchise — fuses, TVS diodes, MOVs, gas discharge tubes — is a real Greenwald-style customer-captivity moat: a $0.30 part that is safety-agency qualified into a $30,000 system, where requalification costs vastly exceed any savings from switching. That business grew 18.3% in FY2025 and is the reason the stock deserves a premium to a generic component maker. But it is roughly 44% of revenue, and it is not what has re-rated the stock.

What has re-rated the stock is a data-center-and-grid narrative attached to the two segments with the weakest economics. The tell is arithmetic: in FY2025 the company earned GAAP operating income of $37.5M on $2,386M of revenue — a 1.6% margin — and lost $71.7M, because it wrote off $301.2M of goodwill on the Electronics-Semiconductor unit, the IXYS power-semiconductor franchise. Strip the acquired Dortmund fab out and that unit’s revenue is 19% below its FY2023 level. And the write-down has a specific author: the 10-K says it was “primarily driven by lower projected volumes in the power semiconductor business, largely associated with the Dortmund Fab” — a ~€94M wafer fab that closed on 31 December 2024 and became the principal driver of the largest impairment in company history within twelve months. Management’s own discounted cash flows for the power-semiconductor business fell far enough in Q4 2025 to force a third-of-a-billion-dollar write-down; the stock has roughly doubled since. Both things cannot be right. Meanwhile the incentive plan paid 143% on an “AIP EPS” of $10.69 in the year GAAP EPS was −$2.89, and the proxy contains zero mentions of ROIC — while the Investor Day promises mid-teens ROIC against the 8.4% the company actually earned last year on an ex-impairment basis, and 0.9% on GAAP.

The framing is a high-beta cyclical riding a thematic re-rating, not a quality compounder — and the factor data says so rather than my intuition: beta 1.54, R² 62.8% (so the majority of the move is systematic, not company-specific), nearest factor neighbours ESI, VSH, BorgWarner, Sensata, Lear and Arrow, and a three-year annualised return of 11.9% with a 50% drawdown and a 0.27 Sharpe. That is the honest base rate for this security. The one-year Sharpe of 1.97 is the anomaly, not the norm. Momentum has already stalled — the stock is 17.2% below its 22 June high and the three-month annualised return is negative. At 4.15x EV/sales and 19.7x EV/EBITDA, sitting at the 93.7th percentile of its own ten-year price-to-sales history, the price embeds close to full credit for a 2030 plan in which half the revenue growth must be bought — roughly $2.7bn of acquisitions at the Basler multiple — by the same M&A machine that has impaired 22.9% of everything it bought since 2018.

I am not short this and would not be: the cash flow is real ($433.8M operating, $366.1M free in FY2025), net leverage is ~1.0x, Q1 2026 was a genuinely strong quarter (GAAP operating margin 15.4%, +270bp), and the Basler deal is off to a better start than underwritten. This is a decent business at a demanding price in a crowded trade, not a broken one. Conviction: medium. What would flip me bullish: two or three consecutive quarters showing the power-semiconductor business growing organically at a double-digit rate with segment margins expanding — i.e. evidence the impaired unit is actually inflecting rather than being rationalised — plus a ROIC metric appearing in the incentive plan. What would flip me bearish: a data-center order digestion cycle (book-to-bill falling below 1.0) landing while the company is mid-way through a large debt-funded acquisition, or any further impairment, which would confirm the M&A engine is value-destroying rather than merely inconsistent.

📈 Stock Price Action — Five-Year Event Map

Littelfuse did nothing for four years and then tripled in fifteen months. From January 2021 through January 2025 the shares oscillated in a $180–310 band with no trend — five successive January closes of $230.83, $257.90, $247.47, $235.51 and $234.56. The stock bottomed at a five-year closing low of $147.74 on 8 April 2025, ran to an all-time closing high of $489.10 on 22 June 2026 — a 231% advance in fourteen months — and has since given back 17.2% to $404.86. The 52-week range is $228.05–$489.10. (Price levels: FACT, from the AZI daily series. Attributed drivers: INTERPRETATION.)

Cycle narrative. (1–3) For three full years after the 2021 components boom, Littelfuse was a range-bound industrial: the 2022 rate shock took 20.6% out of the stock in a month, and the 2023–24 destocking cycle drove Electronics segment revenue down 12.1% in FY2024 with segment margins falling from 22.3% to 14.3%. (4) The February 2025 trough combined two distinct shocks — the 27 February Form 12b-25 disclosing a material weakness in inventory controls at a non-US plant, and the April tariff selloff that hit the whole components complex. Director Anthony Grillo bought $4.09M of stock on 7 February 2025 at ~$233, before the low. (5) The 43.5% April–May rebound was a beta recovery, not a fundamental event. (6–7) The re-rating proper runs from January 2026: the Q4’25 print and 2026 guidance, the closing of Basler on 11 December 2025, a Q1’26 beat of $3.31 against a $2.83 consensus, and the 14 May Investor Day at which management published a 2030 ambition of $4.5bn revenue and >$25.00 EPS. (8) Since 22 June the stock has fallen 17.2% on no company-specific news, with the three-month annualised return at −1.6% — the momentum trade has stalled ahead of the 29 July Q2 print.

1. Executive Summary

Littelfuse designs and manufactures circuit-protection and power-control components — fuses, circuit breakers, transient-voltage-suppression diodes, varistors, power semiconductors, sensors, relays and, since December 2025, high-voltage excitation systems. FY2025 revenue was $2,386.3M across three segments: Electronics ($1,345.5M), Transportation ($676.4M) and Industrial ($364.4M).

The business is two companies wearing one ticker. The first is a high-quality protection franchise: passive protection and industrial protection, roughly 44% of revenue, where safety-agency qualification and design-in specification create real customer captivity. Electronics-Passive Products & Sensors grew 18.3% in FY2025 to $675.9M, and management describes the Protection franchise as “one of the most profitable” in the portfolio, growing double digits. The second is a power-semiconductor business built by acquiring IXYS in 2018 — roughly 28% of revenue — which competes against Infineon, onsemi and STMicroelectronics in a capacity-scale game it cannot win. In Q4 2025 the company wrote off $301.2M of goodwill on that unit.

FY2025 GAAP results were poor and are widely misreported. GAAP operating income was $37.5M — a 1.57% margin — and the company lost $71.7M ($2.89 per diluted share). Third-party data services report FY2025 “operating income” of $357.6M because they exclude the $320.1M of restructuring and impairment charges. Both numbers appear in this memo, labelled. The gap between GAAP and adjusted operating income was $2–17M annually in FY2019–FY2023; it was $108.4M in FY2024 and $320.0M in FY2025. When the size of the adjustment grows twenty-fold in two years, the adjustment has stopped being a footnote and has become the story.

Cash generation, by contrast, did not deteriorate. Operating cash flow was $433.8M in FY2025 (up from $367.6M), capex $67.6M, free cash flow $366.1M. Net debt is $239.2M against $2,426.0M of equity; net leverage is approximately 1.0x. The impairment is non-cash and reflects the price paid for past acquisitions, not the current cash economics of the operating business.

Returns on capital are the weak point. On an ex-impairment basis, FY2025 NOPAT/average invested capital was 8.4%; on GAAP, 0.9%. The cyclical peak was 13.7% in FY2022. Management’s 2030 ambition targets “mid-teens ROIC” — a level the company has never sustained. The proxy contains no ROIC metric of any kind: FY2025 annual incentives were weighted 30% sales, 30% EPS, 20% operating cash flow, 20% individual, and paid out at 153%/143%/200% on those three financial measures in the year the company reported a GAAP loss.

Capital allocation is the central question. Since 2018 Littelfuse has spent $1,882.1M in cash on acquisitions and impaired $430.4M of the resulting goodwill and intangibles — 22.9%. The 2030 plan requires more of this: +6–8% of annual revenue CAGR is explicitly inorganic, implying roughly $2.7bn of further acquisition spend at the multiple paid for Basler. The record does not yet support underwriting that at par. Against this, Basler itself is off to a strong start (39pp of Industrial segment growth in Q1 2026, accretion raised to $0.15+), and the company deserves credit for defining its incentive sales metric to exclude the acquisition.

Recent operating momentum is genuinely good. Q1 FY2026 revenue was $657.0M, +18.5% year over year and +9% organic, with GAAP operating margin of 15.4% (up 270bp) and adjusted EPS of $3.31 against a $2.83 consensus. Book-to-bill was “well above 1.0” with bookings up more than 20%. Data centre and grid infrastructure were the fastest-growing end markets.

The valuation is the problem, not the business. At $404.86 the shares trade at 4.15x EV/sales, 19.7x EV/EBITDA, 26.7x EV/adjusted EBIT, 16.3x tangible book and a 3.63% free-cash-flow yield — the 93.7th percentile of the stock’s own price-to-sales history. Against management’s fully-successful 2030 case of >$25.00 EPS, the market is already paying 16.2x terminal-year earnings, implying a low-single-digit annualised return if the entire five-year plan lands. Context matters, though: the whole electrification complex is extended (Amphenol at the 94.0th percentile of its own history, nVent 94.9th, Hubbell 91.1st, Vishay 98.1st). Littelfuse is not uniquely expensive — it is expensive alongside its peer group while earning a lower return on capital than most of it.

2. Business Overview

Littelfuse was founded in Chicago in 1927 to make the first small, fast-acting fuse for the automotive industry. Ninety-nine years later the company describes itself as “a leader in developing smart solutions that enable safe and efficient electrical energy transfer.” The underlying product logic has not changed: the company sells components whose job is to fail safely so that something far more expensive does not.

2.1 Segments and product lines

Electronics ($1,345.5M, 56.4% of FY2025 revenue). Two distinct businesses share one segment label, and the distinction matters more than any other structural fact about this company.

• Electronics — Passive Products and Sensors ($675.9M, +18.3% y/y): fuses and fuse accessories, PTC resettable fuses, electromechanical switches and interconnect, polymer ESD suppressors, varistors, reed-switch magnetic sensing and gas discharge tubes.
• Electronics — Semiconductor ($669.6M, +8.8% y/y as reported): discrete TVS diodes, TVS diode arrays, protection and switching thyristors, MOSFETs, diodes and IGBTs. This is the IXYS franchise acquired in January 2018, plus the Dortmund 200mm wafer fab acquired from Elmos Semiconductor SE. Management further splits this unit into “Protection” semiconductors — which it calls “a model franchise… with a demonstrated track record of execution and operating discipline” — and “power semiconductors,” where it is “rationalizing lower value products and optimizing our footprint.”

Transportation ($676.4M, 28.3%). Blade, resettable, high-current and high-voltage fuses, battery cable protectors, switches, relays, circuit breakers and power distribution modules, plus sensors. Three sub-lines: Commercial Vehicle Products ($320.5M, flat), Passenger Car Products ($293.6M, +5.5%) and Automotive Sensors ($62.2M, −15.4%, shrinking on a deliberate exit of low-margin products).

Industrial ($364.4M, 15.3%, +9.9%). Industrial fuses, protection relays, contactors, transformers, residual current devices, GFCIs, residual current monitors, arc-fault detection and temperature sensors, serving renewables, energy storage, EV charging infrastructure, HVAC, industrial safety, non-residential construction, MRO, mining and factory automation. Basler Electric — high-voltage excitation and high-energy protection systems for grid, utility and power-generation customers, ~$130M of annualised sales — was added here on 11 December 2025.

2.2 How it makes money

Littelfuse is a component manufacturer selling through a mix of independent distributors, a direct salesforce and manufacturers’ representatives. Revenue is recognised on shipment, arrangements are “predominately short term,” and there is essentially no contracted recurring revenue in the software sense. What creates the recurring character of the business is design-in: once a Littelfuse part is specified into a customer’s board or wiring harness and qualified to a UL or IEC standard, it tends to remain there for the life of the platform. The company operates a “ship and debit” programme allowing distributors price adjustments on inventory already shipped — standard in the industry, and a reminder that reported distributor-channel revenue carries a rebate estimate.

Two structural features deserve emphasis. First, unit economics are dominated by the mismatch between part price and system value. A fuse costs cents; the motor drive, EV battery pack or data-centre rack it protects costs thousands to hundreds of thousands. Customers do not optimise this line item. Second, the business is cyclical and channel-levered. FY2024 revenue fell 7.3% on distributor destocking; Electronics fell 12.1%. The same channel amplifies upswings, which is a fair description of what is happening now.

2.3 Geographic and customer profile

Sales span the Americas, Asia-Pacific and Europe. The customer base is broad — the Investor Day describes the Computing/Communications/Diversified Industrials market as “Leading: 33% • Emerging: 33% • Broad: 34%” — and no single customer concentration is disclosed as material. This diversification is a genuine risk mitigant relative to peers dependent on a handful of OEM platforms. Notably, the March 2026 appointment of Holly Paeper (President, Commercial HVAC at Trane Technologies) to the board came with disclosure that Littelfuse had sold ~$19.0M of product to Trane and affiliates since 29 December 2024 — a related-party relationship small enough to be immaterial financially but worth noting for governance.

Verdict: a well-diversified, genuinely long-lived component franchise with sound economics in its protection core and a structurally weaker power-semiconductor business bolted on by acquisition. The mix — roughly 44% high-quality protection, 28% commodity-adjacent power semiconductors, 28% cyclical transportation — is the single most important thing to hold in mind when reading the rest of this memo, because the market is currently valuing the whole at a multiple that only the first third arguably justifies.

3. Industry Dynamics

3.1 Structure and market size

Management sizes its serviceable addressable market at approximately $22bn in 2025 growing to ~$30bn by 2030, a ~6.4% CAGR (company estimate, Investor Day 14 May 2026). The framing is credible in direction if not in precision: global electricity demand is rising, power architectures are moving to higher voltages and current densities, and the number of protection points per system is increasing. Littelfuse’s own framing — “from milliwatts to megawatts,” “grid to chip” — is a reasonable description of where protection content shows up.

The industry is not one industry. It is at least four, with materially different structures:

Circuit protection (fuses, breakers, TVS, MOVs) — structurally attractive. Fragmented, specification-driven, high-mix/low-unit-cost, with safety certification as a barrier. Competitors are Eaton (Bussmann), Mersen, Bourns, TDK and Vishay. Nobody is building greenfield fuse capacity to chase share; the returns do not justify it and the qualification cycle makes share transfer slow. This is a good business and Littelfuse is a leader in it.

Power semiconductors (IGBTs, MOSFETs, diodes) — structurally poor for a sub-scale participant. This is a capacity and process-node game dominated by Infineon, onsemi, STMicroelectronics, Mitsubishi Electric and Fuji Electric, with aggressive Chinese entrants adding capacity. Fixed costs are enormous, pricing is set by the marginal supplier, and the 2023–25 downcycle in industrial and EV power semis produced exactly the margin compression theory predicts. Littelfuse’s response — buying a single 200mm fab in Dortmund for ~€94m — buys supply-chain resilience, not scale advantage. The FY2025 impairment is the market clearing price of that strategic position.

Transportation electrical (fuses, PDMs, sensors) — mediocre. OEM-facing, price-down by contract, with content growth from electrification partly offsetting annual give-backs. Competitors include Sensata, Aptiv, Eaton and Mersen. Littelfuse’s Transportation segment earned a 5.0% operating margin as recently as FY2023.

Industrial protection and grid equipment — currently the best place to be. Protective relays, excitation systems and industrial fuses serving grid modernisation and data-centre power. Competitors are Schweitzer Engineering (SEL), ABB, Schneider Electric and GE Vernova. Demand is genuinely strong and order books are long. This is where Basler sits and where the incremental narrative value is being created.

3.2 The capital cycle — the analytically important point

Applying the Marathon capital-cycle lens, the electrification and data-centre power chain is squarely in the capital-attracting phase. The evidence is in the valuations of the entire supply chain, measured against each company’s own multi-year history rather than cross-sectionally:

Source: AZI valuation_index own-history percentile ranks, 2026-07-24. Percentiles compare each stock only to its own ~10-year history, never cross-sectionally.

Six of seven names in the complex sit above the 79th percentile of their own decade. That is not a Littelfuse-specific observation; it is a statement that the market has repriced an entire supply chain on the same thesis. High returns and high multiples attract capital, and capital attracted to power electronics and grid equipment will arrive as capacity. The classic Marathon inference is that this is the precondition for margin mean-reversion, not for the margin expansion embedded in a 24–26% adjusted EBITDA target. The counter-argument — and it is a serious one — is that grid and data-centre demand is supply-constrained by permitting, transformer lead times and skilled labour rather than by component capacity, so the normal supply response is slower than usual.

The one genuinely notable exception in the table is Sensata at the 59.8th percentile, which is a useful sanity check: the auto-levered parts of this complex have not re-rated. The re-rating is specifically about data centres and the grid.

3.3 Regulation and secular drivers

Safety regulation is a tailwind and a barrier: UL, IEC, and regional electrical codes mandate protection devices and make substitution slow. Electrification of vehicles adds protection content per unit even where unit volumes are flat — a genuine offset that showed up in Q1 2026 as passenger-vehicle organic growth of 4% “amid a soft global production environment.” Higher-voltage architectures (800V systems in both vehicles and data-centre racks) require new, higher-value protection devices, which is the strongest structural argument for content growth. Tariffs are a live variable: the April 2025 selloff was tariff-driven, and management describes managing commodity inflation (silver, copper) toward “price-cost neutral.”

Verdict: a mixed industry, and the mix is deteriorating in the places the story now depends on. Circuit protection is a structurally good business and Littelfuse’s position in it is defensible. Power semiconductors are structurally poor for a participant of this scale, which the company has now acknowledged in cash terms. Grid and data-centre exposure is genuinely attractive today but is where capital is flooding in, and the supply-side discipline that protects the fuse business does not exist there. The honest summary is that Littelfuse’s best business is the one growing at mid-single digits and its fastest-growing exposure is the one with the weakest barriers.

4. Competitive Position

4.1 Naming the moat

Under the Greenwald taxonomy there are three genuine competitive advantages: supply/cost, demand/customer captivity, and economies of scale reinforced by captivity. Littelfuse has one of the three, in one part of the business.

The moat is customer captivity in circuit protection, created by qualification cost asymmetry. The mechanism is specific and testable. A protection device is designed into a customer’s circuit, then qualified as a system to a safety standard (UL 248 for fuses, IEC 60269, automotive AEC-Q). Changing supplier requires re-testing and, frequently, re-certifying the end product. The cost of that process is measured in engineering months; the saving available from a cheaper fuse is measured in cents per unit. No rational procurement organisation makes that trade for a non-strategic component. The result is that share moves slowly, pricing is defensible, and incumbency compounds across product generations.

The financial evidence that this moat exists is exactly what the the output standards applied here standard demands — an outcome that would deteriorate without it. Electronics-Passive Products and Sensors grew from $571.4M to $675.9M in FY2025 (+18.3%) while the segment’s operating margin recovered from 14.3% to 16.4%; management attributes the margin recovery specifically to “volume leverage from the electronics products business, partially offset by lower gross margin from the semiconductor business.” Growth and price and margin recovery arriving together in a component business is the signature of pricing power. In a commodity, volume growth of that magnitude comes with price erosion.

The moat does not extend to power semiconductors, and the company has now conceded this in cash. Here the required advantage would be cost, and cost in power discretes comes from fab scale and yield. Infineon and onsemi operate at multiples of Littelfuse’s wafer volume. The FY2025 write-off of $301.2M of Electronics-Semiconductor goodwill is not an accounting curiosity — it is management’s own discounted-cash-flow model concluding that the unit is worth materially less than the price paid for IXYS. That is the cleanest possible disproof of a moat claim. Management’s current language is consistent: they are “rationalizing lower value products,” “optimizing our footprint,” and “exiting areas of low value,” with margin improvement promised “over the mid to long term.”

4.2 Market-share stability test

Greenwald’s operational test for a moat is share stability. The evidence is mixed and splits along exactly the same line:

• Protection/passives: share appears stable to rising. Management reports design wins in data centre “up more than double year over year” in 2025 and “double-digit design win growth” targeted in transportation for 2026, with Q1 2026 transportation growth of high single digits attributed to “content and share gains” in a soft production environment. Winning share while the end market shrinks is strong evidence.
• Power semiconductors: share is being lost or abandoned. Revenue of $669.6M in FY2025 includes $49.0M of acquired Dortmund fab revenue; the organic base of ~$620.6M compares with $767.4M in FY2023 — a 19.1% organic decline over two years. Some of that is deliberate exit of low-value product, which is the right decision, but a business that must shrink to become profitable is not a business with an advantage.
• Automotive sensors: unambiguously shrinking, $88.5M → $73.6M → $62.2M over three years, on “the strategic exit of certain lower margin products.”

4.3 Direct competitive comparison

Against Amphenol and TE Connectivity, Littelfuse is far smaller ($10.1bn market cap versus $193.7bn and $57.6bn) and materially less profitable on capital. Against Vishay, Littelfuse has a better margin structure and a cleaner balance sheet. Against Sensata, it has better growth and less automotive concentration. Against Eaton and Schneider in industrial protection and grid, it is a component specialist selling into their systems as much as competing with them — a position that is defensible but caps the value captured. Against Infineon and onsemi in power semiconductors, it is sub-scale and has now said so with a write-down.

The factor model offers an independent read on the peer set. FactorsToday’s nearest factor neighbours for LFUS are Element Solutions (0.939), Vishay (0.938), BorgWarner (0.928), Sensata (0.926), Advanced Energy (0.921), Lear (0.915) and Arrow Electronics (0.902) — an auto-and-industrial-cyclical cluster with semi-capital-equipment names behind it. The market is not trading Littelfuse alongside Amphenol or the quality-compounder complex; it is trading it alongside cyclicals. That is a useful corrective to the narrative.

Verdict: a real but narrow moat. Littelfuse has genuine, financially-evidenced customer captivity in circuit protection — roughly 44% of revenue — and essentially no durable advantage in power semiconductors (28%) or transportation electrical (28%). This is a company with one very good business, one mediocre one, and one it overpaid for. The valuation currently applied to the whole is the valuation the first business alone might deserve. That is the central tension of the investment case.

5. Growth History and Forward Opportunities

5.1 The historical record

Two things stand out. First, this is a cyclical, not a compounder: revenue is up 58.7% over six years but with two down years and a peak-to-trough of 12.9% from FY2022 to FY2024. FY2025 revenue of $2,386.3M is still 5.1% below the FY2022 peak of $2,513.9M. Second, the divergence between GAAP and adjusted margins is new and enormous. Through FY2023 the two lines tracked within 100bp. In FY2024 the gap was 500bp and in FY2025 it was 1,340bp. An investor reading only the adjusted line would conclude FY2025 was a good year with 15.0% margins; the GAAP line says the company barely broke even at the operating level.

5.2 Organic versus acquired

The growth record leans heavily on acquisition. Cash paid for acquisitions from FY2018 to FY2025 totals $1,882.1M — against FY2025 revenue of $2,386.3M, the company has spent nearly a year of revenue buying growth in eight years. Named contributors: IXYS (2018), Hartland Controls (2021, ~$70M sales), Carling Technologies (2021, ~$170M sales), Western Automation (2023), Dortmund Fab (closed 31 Dec 2024, $49.0M of FY2025 revenue), Basler Electric (Dec 2025, ~$130M sales).

FY2025’s 8.9% headline growth decomposes as: Electronics +$158.7M (+13.4%), of which $49.0M was the Dortmund fab and $9.7M favourable FX — leaving ~$100M organic, essentially all from passive products; Transportation +$4.0M (+0.6%), of which $7.5M was FX — i.e. organically down $3.5M; Industrial +$32.8M (+9.9%), organic. Adjusting for acquisition and currency, FY2025 organic growth was approximately 5%, concentrated almost entirely in passive electronics and industrial.

5.3 Current momentum

Q1 FY2026 was the strongest quarter in years: revenue $657.0M, +18.5% reported, +9% organic, with Basler contributing 6pp and FX 3pp. By segment, Electronics +18% reported and +15% organic (passives +22% organic, semiconductors +8% organic), Transportation +5% reported and +1% organic, Industrial +45% reported and +5% organic with Basler contributing 39pp. Book-to-bill was “well above 1.0” and bookings up more than 20% year over year, with sequential improvement through the quarter. Q2 guidance of $690–710M implies 14% growth (8% organic).

Two observations temper the enthusiasm. First, the semiconductor business grew 8% organically in Q1 2026 — a real improvement, but from a base 19% below FY2023, and management continues to describe active product rationalisation that will subtract from growth. Second, the Industrial segment’s spectacular 45% growth is 39pp acquisition; organic Industrial growth was 5%, held back by “soft residential HVAC volumes.”

5.4 Forward opportunities

Management’s 2030 ambition (Investor Day, 14 May 2026) is explicit and quantified — commendably so, and it gives us something falsifiable:

• Revenue $4.5bn by 2030 from $2.4bn — a ~13.5% CAGR
• Adjusted EBITDA $1.1bn at a 24–26% margin (from 20.6% in FY2025)
• Free cash flow $600M (from $366M)
• EPS >$25.00, “high-teens EPS CAGR”
• Mid-teens ROIC
• Composition: +7–9% organic revenue CAGR and +6–8% inorganic, with product rationalisation a −1% to −2% annual drag

The three named high-growth opportunities are grid & utility infrastructure, data centre, and aerospace & defence, each targeted to grow at double-digit CAGRs against a core growing mid-single digits. The company estimates its aerospace and defence SAM grows at ~15% CAGR to 2030 and that data-centre-and-building-infrastructure rises from 9% to 21% of the Computing/Communications/Diversified Industrials market revenue.

The organic half of this plan is defensible. 7–9% organic against a SAM growing ~6.4% requires modest share gain and content growth, which the design-win data supports. The inorganic half is the problem. Adding 6–8% of revenue growth per year inorganically over five years means acquiring roughly $1.0bn of revenue. At the ~2.7x sales paid for Basler, that is approximately $2.7bn of acquisition spend against current free cash flow of $366M/year — the entirety of five years’ free cash flow plus meaningful incremental leverage, executed by a company that has impaired 22.9% of what it bought in the last cycle.

Verdict: growth quality is improving at the margin but remains structurally mixed, and the forward plan is half-dependent on the weakest part of the track record. The organic engine — passive protection, industrial protection, content growth in transportation — is real, evidenced by share gains in declining end markets and by doubling design wins. But the FY2025 organic rate was ~5%, the FY2026 rate is running at 8–9% on a cyclical upswing with book-to-bill above 1.0, and the 2030 plan needs 7–9% sustained plus $2.7bn of successful M&A. Growth that must be bought at 2.7x sales by an acquirer with this impairment record is low-quality growth until proven otherwise.

6. Financial Quality

6.1 The quality-of-earnings problem

The single most important analytical fact in this memo is that Littelfuse’s reported “operating income” depends entirely on which of two numbers you use, and the two have diverged violently.

GAAP operating income from SEC XBRL us-gaap:OperatingIncomeLoss; restructuring and impairment from us-gaap:RestructuringSettlementAndImpairmentProvisions.

Widely-used data services report the FY2025 figure as $357.6M of operating income at a 15.0% margin. That is the ex-impairment presentation. The audited GAAP figure is $37.5M at a 1.57% margin. This memo reports both, but an investor who never looks past the aggregator will have an entirely wrong picture of FY2025.

Below the operating line the arithmetic compounds. FY2025 pre-tax income was $3.6M. Income tax expense was $75.3M — because goodwill impairment is largely non-deductible — producing a net loss of $71.7M, or −$2.89 per diluted share. The Q4 2025 quarter alone showed a $242.1M net loss.

6.2 What was written off, and why it matters

The FY2025 charge was a $301.2M non-cash goodwill impairment of the Electronics-Semiconductor reporting unit, recognised in Q4 2025 following the annual test performed as of 28 September 2025. It followed FY2024’s $93.5M of impairments: $36.1M of goodwill in Industrial Controls and Sensors, $8.6M in Automotive Sensors, and $47.8M of intangibles (acquired customer relationships, developed technology and a tradename) in Industrial Controls and Sensors.

The company’s own explanation of the FY2025 charge is the most important sentence in the filing: the reduction in fair value came “from lower expectations for future revenue, profitability and cash flows… In the context of a recent leadership transition and strategic reassessment in the semiconductor business, the reduction was primarily driven by lower projected volumes in the power semiconductor business, largely associated with the Dortmund Fab.” The Dortmund fab was acquired from Elmos Semiconductor SE for approximately €94M and closed on 31 December 2024 — meaning the largest write-down in company history was driven principally by an asset bought twelve months earlier. The FY2024 Industrial Controls and Sensors charge was similarly attributed to “lower-than-expected demand in the electric vehicle end market as well as reduced government funding to support charging infrastructures for electric vehicles, primarily in Europe.”

After the charge, $238.5M of goodwill remains in the Electronics-Semiconductor unit. The company tests six reporting units; Automotive Sensors now carries no goodwill at all, and for the remaining five — Electronics-Passive Products and Sensors, Passenger Car Products, Commercial Vehicle Products, Industrial Controls and Sensors, and Industrial Circuit Protection — estimated fair values exceeded carrying values at the September 2025 test. No headroom percentages are disclosed.

Accumulated goodwill impairment on the balance sheet rose from $85.5M at FY2024-end to $397.0M at FY2025-end. Goodwill still carried is $1,211.4M and other intangibles $594.9M, against total equity of $2,426.0M — intangibles are 74.5% of book equity. Tangible book value is $619.7M, or $24.86 per share.

A goodwill impairment is a statement about the future, not the past. It means management’s own discounted cash flow for that reporting unit, tested at least annually, fell below its carrying value. Management concluded in Q4 2025 that the power-semiconductor business is worth $301.2M less than previously carried — and the stock has risen materially since. That is the disagreement at the heart of this investment case.

6.3 Cash flow — the genuine strength

Cash generation did not follow GAAP earnings down, which is the correct reason to remain constructive on the business:

Free cash flow conversion of 13–16% of revenue is healthy for a component manufacturer, and FY2025 free cash flow of $366.1M was achieved in the year of the $71.7M net loss — clean confirmation that the impairment is non-cash. Capex has fallen from $104.3M to $67.6M, which is worth watching: 2.8% of revenue is a light reinvestment rate for a company that owns wafer fabs and is promising to double in size. Working capital released $26.9M in FY2025, with inventories down $40.2M — the destocking cycle unwinding.

Q1 FY2026 operating cash flow was $80M and free cash flow $66M, up 55% year over year.

6.4 Returns on capital — the weak point

NOPAT at a 22% tax rate over average (total equity + total debt).

This is the table that matters most for the investment case. Even on the flattering ex-impairment basis, Littelfuse earned 8.4% on invested capital in FY2025 — below any plausible weighted average cost of capital for a business with a 1.54 beta. On GAAP it earned 0.9%. The cyclical peak was 13.7% in FY2022, and the trailing five-year average on the ex-impairment basis is roughly 9.9%. Management’s 2030 target of “mid-teens ROIC” would require exceeding the best year of the last cycle and sustaining it.

The reason returns are mediocre despite decent margins is the denominator: $1,806.3M of goodwill and intangibles sitting on the balance sheet from acquisitions. The operating business earns respectable returns on its tangible assets; the company earns poor returns on the capital shareholders have committed, because a large fraction of that capital was spent buying other companies.

One important counter-test: cumulative GAAP net income from FY2018 to FY2025 is +$1,378.7M. Unlike a genuinely capital-destructive cyclical, Littelfuse has produced substantial cumulative accounting profit across the full cycle. The impairment is a write-down of purchased goodwill, not evidence of a business that never earned. This distinction matters and argues against the most bearish reading.

6.5 Balance sheet

Conservative and not a source of risk. At FY2025-end: cash $563.4M, total debt $885.8M, net debt $239.2M, equity $2,426.0M. Net debt/EBITDA is roughly 0.5x on adjusted EBITDA; management describes net leverage as “approximately 1.0x.” The current ratio is 2.69x. In March 2026 the company refinanced into an $800M senior unsecured revolver. In Q1 2026 it repaid $66.3M of term loan and $100M of revolver. Pension liabilities are modest at $38.7M. There is ample capacity for the acquisition programme management describes — which is the point, and the risk.

Verdict: economics do improve with scale in the protection business, but the consolidated entity does not earn its cost of capital. Gross margin has recovered from 33.4% in Q4 2024 to 38.7% in Q1 2026 and incremental flow-through was 38% in Q1 2026 against a 30–35% long-term target — genuine operating leverage. Cash conversion is good and the balance sheet is sound. But the returns table is the verdict: a business earning 8.4% on capital at the top of an upcycle, with 74.5% of book equity in intangibles and a $301.2M write-off just taken, has not demonstrated that it converts scale into shareholder returns. The quality of earnings is also now a live question, given a GAAP-to-adjusted gap that has grown from $16.5M to $320.1M in two years.

7. Capital Allocation

7.1 The M&A record

Cash paid for acquisitions FY2018–FY2025: $1,882.1M. Cumulative impairment of acquired goodwill and intangibles: $430.4M — 22.9% of the spend.

That headline ratio is damning but deserves fair handling. Carling, bought for $315.5M in 2021, sits inside a Transportation segment whose operating margin has gone from 5.0% to 12.5% in two years with commercial-vehicle price and cost actions cited as the driver — that looks like a good deal. Basler is early but ahead of underwriting: it contributed 39pp of Industrial segment growth in its first full quarter, management raised expected 2026 adjusted EPS accretion from $0.10–0.15 to $0.15+, and it carries high-teens adjusted EBITDA margins. The failures are concentrated in IXYS — the largest deal in company history, and the one that defined the strategic direction into power semiconductors — and in the smaller Industrial Controls and Sensors and Automotive Sensors units.

The pattern is not random. Littelfuse has allocated capital well when buying protection and control businesses adjacent to its core (Carling, Basler, Western Automation) and badly when buying its way into semiconductor manufacturing (IXYS and the Dortmund fab). That is a legible, teachable distinction, and the current management team’s stated strategy — “focus on accelerating high-growth opportunities,” grid/utility, data centre, A&D — points at the first category. Whether the discipline holds when $2.7bn needs deploying is the open question.

The Dortmund fab deserves particular emphasis, because it is the fastest capital-allocation failure in the record. The company paid approximately €94M to Elmos Semiconductor SE for a 200mm wafer fab, closing on 31 December 2024. Twelve months later, in explaining the $301.2M goodwill impairment, the 10-K states the fair-value reduction was “primarily driven by lower projected volumes in the power semiconductor business, largely associated with the Dortmund Fab.” An acquisition whose stated rationale was to “enhance the resilience and reliability of our power semiconductor supply chain” instead became the principal driver of the largest write-down in the company’s history within a year of closing. Whatever volume assumptions underwrote that purchase were wrong almost immediately.

7.2 Incentives — the most troubling finding

The 2026 proxy statement contains zero occurrences of “ROIC,” “return on invested capital,” or “EBITDA.” Executive compensation is built on scale and cash, not returns.

FY2025 annual incentive plan design and outcomes for the CEO and CFO:

The critical line is the second one. Management earned a 143% payout on “AIP EPS” of $10.69 in the fiscal year in which GAAP diluted EPS was −$2.89. The $13.58 difference is the impairment and associated charges added back. The cash-flow metric paid at the 200% maximum. In the year Littelfuse wrote off $301.2M of shareholder capital, the compensation scorecard registered an excellent year.

Two points of genuine credit are owed. First, AIP Corporate Sales was explicitly defined to exclude the Basler acquisition — an unusual and real discipline that prevents management from buying its way to a sales bonus. Second, the committee responded to shareholder feedback by replacing stock options with performance share units from FY2025, earned on total shareholder return relative to an index.

But the PSU change cuts both ways. A relative-TSR award pays out on a re-rating. Combined with the complete absence of any return-on-capital metric, the scorecard now rewards revenue scale, adjusted EPS, operating cash flow and share price — precisely the four outputs that a debt-and-cash-funded acquisition programme reliably delivers regardless of whether the acquisitions earn their cost of capital. Management is being asked at Investor Day to deliver mid-teens ROIC and is being paid on none of it.

7.3 Skin in the game

Insider ownership is thin. All current directors and executive officers as a group held 260,164 shares — 1.0% of the company as of 27 February 2026. CEO Greg Henderson held 5,917 shares; CFO Abhishek Khandelwal held zero. Both are recent appointees, which explains much of this, but it means the executives setting a five-year, multi-billion-dollar capital deployment plan have almost no personal capital riding on the outcome.

The transaction record reinforces the point. Across the complete five-year Form 3/4/5 corpus (421 of 421 filings parsed):

Four open-market purchases in five years, all by one director — Anthony Grillo — on a single day, 7 February 2025, at $232.92–$235.63. That trade was well-timed, made shortly before the April low, by a director whose board biography emphasises financial-markets and corporate-acquisition experience. It is a genuine signal, and the only one of its kind.

Against it: $98.0M of sales, a 24:1 ratio. Selling by year runs $23.5M (2021), $5.8M (2022), $11.7M (2023), $11.7M (2024), $13.0M (2025) and $32.4M in 2026 year to date — the heaviest year of the five, with five months still to run, executed into the re-rating. Named sellers include former CEO David Heinzmann ($21.2M), Deepak Nayar ($12.9M), departing Chief Legal Officer Ryan Stafford ($12.8M), Peter Kim ($7.3M), former CFO Meenal Sethna ($6.8M) and Chairman Gordon Hunter ($6.1M). Much of this is routine option-exercise-and-sell behaviour by executives with concentrated equity, and several of the largest sellers were departing. But no operating executive has bought a share on the open market in five years.

7.4 Returns of capital and management turnover

Buybacks have been modest and are now suspended: $40.9M in FY2024, $27.6M in FY2025, $0 in Q1 FY2026. Declining to repurchase stock at $400+ after buying at lower prices is defensible capital discipline — or simply cash preservation for M&A. Either reading is more favourable than the alternative. The dividend has risen steadily: $2.02 (2021), $2.26, $2.50, $2.70, $2.90 (2025), with $0.75 per quarter in Q1 2026 — $18.8M in the quarter. The current yield is approximately 0.74%.

Management turnover has been extraordinary. CEO David Heinzmann retired effective February 2025 after a long tenure; Greg Henderson, previously SVP of the Automotive & Energy, Communications and Aerospace Group at Analog Devices, became President and CEO on 10 February 2025. CFO Meenal Sethna transitioned out in June 2025; Abhishek Khandelwal was appointed CFO on 18 June 2025. EVP M&A and Chief Legal Officer Ryan Stafford separated effective 30 April 2026; Anne-Marie D’Angelo became Chief Legal Officer on 1 May 2026. CEO, CFO and Chief Legal Officer were all replaced within roughly fifteen months. A new team is entitled to clear the decks — and the timing of the FY2025 impairment, taken in the new CEO’s first full year, is consistent with exactly that. But it also means the 2030 plan is being underwritten by executives with no track record at this company and essentially no equity in it.

Verdict: capital allocation has been inconsistent, the incentive structure is misaligned with returns, and the forward plan asks for more of what has worked least well. The company has destroyed $430.4M of the $1,882.1M it deployed on acquisitions since 2018 while paying management above-target bonuses on metrics that add the destruction back. Balanced against that: the deals adjacent to the protection core have worked, Basler is ahead of plan, the sales metric excludes acquisitions, buybacks were sensibly halted at high prices, the balance sheet is conservative, and a genuinely new management team may allocate differently. The verdict is negative on the record and open on the future — but the burden of proof sits squarely with management, and the current share price grants them the benefit of the doubt in full.

8. Changes and Headwinds — Last Two Years

Accounting and control failure (Feb 2025). On 27 February 2025 Littelfuse filed a Form 12b-25 notifying it could not file its FY2024 10-K on time, disclosing “a failure in adherence to the Company’s inventory cycle count policies and procedures at one of its non-U.S. manufacturing legal entities” affecting inventory valuation, originating “in periods prior to 2023.” The company disclosed a material weakness in internal control over financial reporting as of 28 December 2024. FY2024 results included $13.5M of cumulative out-of-period adjustments, reducing FY2024 gross margin by 60bp and Transportation segment operating margin by 170bp. The 10-K was filed 13 March 2025. There is also SEC comment-letter correspondence on file from October–November 2023.

Resolution: the FY2025 10-K states the material weaknesses were remediated as of 27 December 2025, following replacement of key finance and operational personnel at non-US manufacturing locations, redesign of regional finance structures, new monitoring controls and independently-facilitated full physical inventory counts in three consecutive quarters. Deloitte & Touche LLP issued an unqualified attestation on ICFR. Basler’s controls were excluded from the assessment (0.2% of sales, 9% of assets) — a standard first-year exclusion, but it means 9% of the asset base is unaudited for control effectiveness.

Leadership transition (2025–2026). CEO, CFO and Chief Legal Officer all replaced within fifteen months (see the capital-allocation section). Holly Paeper (Trane Technologies) joined the board in March 2026, expanding it to nine; the appointment 8-K disclosed ~$19.0M of sales to Trane since December 2024.

The FY2024 and FY2025 impairments. $93.5M in FY2024 and $302.1M in FY2025 (the financial-quality section). Accumulated goodwill impairment rose from $85.5M to $397.0M.

Portfolio actions. Deliberate exit of low-margin automotive sensor products (revenue $88.5M → $62.2M over three years); exit of the marine business within commercial vehicle; ongoing “rationalization of lower value products” and footprint optimisation in power semiconductors, which management says will be a −1% to −2% annual revenue drag through 2030.

Basler Electric acquisition (Dec 2025). $350.3M net of cash for ~$130M of annualised sales in grid, utility and power-generation control and protection. Funded from cash on hand. The strategic pivot toward grid and data-centre infrastructure in physical form.

Financing (Mar 2026). Amended and restated $800M senior unsecured revolving credit facility replacing the June 2022 agreement — capacity for the acquisition programme.

Investor Day (May 2026). First detailed long-range plan under the new CEO: $4.5bn revenue, $1.1bn adjusted EBITDA, $600M FCF, >$25.00 EPS, mid-teens ROIC by 2030.

Demand environment. The dominant change is the emergence of data centre and grid infrastructure as the fastest-growing end markets, with management reporting 2025 data-centre design wins “up more than double,” strong double-digit data-centre revenue growth in Q1 2026, book-to-bill well above 1.0 and bookings up more than 20%. Offsetting headwinds: soft residential HVAC, a soft global vehicle production environment, and commodity inflation in silver and copper that management is managing toward “price-cost neutral.”

Verdict: the two-year record contains one serious governance failure now credibly remediated, one large admission of past capital misallocation, a complete leadership turnover, and a genuine improvement in demand and operating execution. On balance these changes strengthen the near-term operating thesis and weaken confidence in the institution’s historical judgment. The material weakness was handled properly — disclosed promptly, remediated within a year, independently attested — and should not be over-weighted now that it is closed. The impairments are more consequential, because they are the company’s own verdict on its largest strategic decision. The new team inherits a better demand environment than any of its predecessors and a strategy that is, on paper, correctly aimed.

9. Risk Analysis

The three risks that actually matter are (1), (2) and (3), and they are correlated. A cyclical component supplier trading at the top of its own valuation history, whose forward plan requires large-scale acquisition spending, is exposed to a single adverse sequence: orders normalise, the multiple compresses, and the company is mid-deployment into a deal financed when the shares were expensive. Risk (12) is genuinely low — the balance sheet is not the vulnerability here, and that materially caps downside relative to a leveraged peer.

10. Valuation Discussion

No price target and no recommendation appears in this section. The discussion is confined to what the current price implies and what would have to be true.

10.1 Where the stock trades

At the 2026-07-24 close of $404.86 on 24.926M shares: market capitalisation ~$10,092M, net debt $239.2M, enterprise value ~$10,331M.

10.2 Own-history context

The AZI own-history percentile ranks (2026-07-24) place the composite at the 89.9th percentile of the stock’s ~10-year range: P/B 4.09 at the 86.2nd percentile and P/S 4.08 at the 93.7th percentile.

Methodological caveat, stated explicitly: the P/E percentile is null because trailing GAAP EPS is negative (−$1.68 on a TTM basis), so the composite rests on only two components. Consistent with the standing rule that the P/E percentile must be ignored where GAAP EPS is distorted, the valuation read here rests on price-to-sales and price-to-book, both of which are unambiguous: this is close to the most expensive Littelfuse has been on revenue in a decade.

10.3 Embedded expectations — what the price requires

Management’s 2030 ambition is unusually explicit, which allows a direct test of what is priced in.

The plan: $4.5bn revenue, $1.1bn adjusted EBITDA (24–26% margin), $600M free cash flow, >$25.00 EPS, mid-teens ROIC, built from +7–9% organic and +6–8% inorganic annual revenue growth, less 1–2% of product rationalisation.

Test 1 — the terminal multiple. At $404.86 against a fully-delivered >$25.00 of 2030 EPS, the market is paying 16.2x terminal-year earnings today, four and a half years early. If the plan lands in full and the shares hold a 20x exit multiple, the 2030 value is ~$500, implying roughly 4–5% annualised price appreciation plus a 0.74% dividend. That is a low-single-digit expected return conditional on perfect execution of an ambitious five-year plan. There is very little margin for error in the price.

Test 2 — the capital required. Closing the revenue gap requires adding ~$2.1bn of revenue, of which roughly $1.0bn is explicitly inorganic. At the ~2.7x sales multiple paid for Basler, that implies approximately $2.7bn of acquisition spend over five years against current free cash flow of ~$366M per year (~$1.8bn cumulative, before any growth). The plan therefore requires deploying effectively all free cash flow plus incremental leverage — feasible given a 1.0x starting net leverage and an $800M revolver, but leaving no capacity for buybacks and no cushion if the deals disappoint.

Test 3 — the return on that capital. Mid-teens ROIC by 2030 requires exceeding the FY2022 cyclical peak of 13.7% and sustaining it, on an invested capital base that the plan itself expands by ~$2.7bn. Adding $2.7bn of goodwill-heavy invested capital while raising ROIC from 8.4% to ~15% requires the acquired businesses to earn substantially above the corporate average from day one. The last cycle’s acquisitions earned so far below it that 22.9% of them were written off.

Test 4 — the margin bridge. Adjusted EBITDA margin must rise from 20.6% (FY2025) to 24–26%. Q1 2026 delivered 22.9%, up 180bp, with incremental flow-through of 38% against a 30–35% long-term guide. The near-term trajectory is genuinely supportive. Sustaining it requires the power-semiconductor rationalisation to work and the acquired businesses to carry higher margins than the base — Basler at “high-teens” adjusted EBITDA margin is below the corporate target, which means the mix effect from that deal is dilutive to the margin goal even as it is accretive to EPS.

10.4 Scenario analysis

The base case — which assumes the company does most of what it says, just not all of it — produces a 2030 value that brackets today’s price. That is the cleanest statement of the valuation problem: at $404.86 the market has already paid for the base case in full and is underwriting a meaningful share of the bull case.

10.5 Comparative context

The peer table in the capital-cycle discussion in the Industry section is essential context and cuts in Littelfuse’s favour on relative terms. Amphenol sits at the 94.0th percentile of its own history, nVent at 94.9th, Hubbell at 91.1st, Vishay at 98.1st. Littelfuse at 89.9th is not the most extended name in its complex. An investor who believes the electrification re-rating is durable would not single out Littelfuse as the expensive one.

The counter is the returns comparison. Littelfuse is paying a peer-group multiple while earning 8.4% on invested capital ex-impairment — below Amphenol, below Hubbell, below nVent. Paying a similar multiple for a lower return on capital is the specific form of the mispricing argued here.

Verdict: the price embeds close to full credit for a five-year plan whose inorganic half is unproven. The market is correctly underwriting a genuine improvement in end-market demand, a real moat in circuit protection, strong recent operating execution and a sound balance sheet. It appears to be underwriting incorrectly the assumption that an acquirer which has impaired 22.9% of its deal spend, pays management nothing for returns on capital, and has never sustained mid-teens ROIC will now deploy $2.7bn at above-average returns. The distribution of outcomes around $404.86 is asymmetric to the downside.

11. Variant Perception

11.1 What consensus believes

Consensus is constructive and momentum-driven. Sell-side coverage is thin — five firms, “Moderate Buy,” with a consensus price target around $364 as of late March 2026 (below the current price, and below where the stock has traded since April). Zacks has repeatedly ranked LFUS a #1 (Strong Buy) through May and June 2026 on rising earnings estimates and has featured it on multiple “momentum stocks to buy” lists, explicitly linking it to “AI trade strength.” Heartland Value Plus described it in its Q2 2026 review as one of several “long-standing holdings that have recently added Data Centers as a new end market.”

The consensus view, stated fairly: Littelfuse is a quality niche component leader whose end markets have inflected from destocking to a multi-year electrification and data-centre upcycle. A new management team has articulated a credible 2030 plan, the balance sheet supports acquisitions, margins are expanding with 38% incremental flow-through, and bookings are up more than 20% with book-to-bill well above 1.0. Nearly every element of that is factually correct.

11.2 The strongest bull case

• The protection moat is real and under-appreciated. Qualification-driven customer captivity is a genuine Greenwald advantage, and passive products growing 22% organically in Q1 2026 with segment margin up 300bp is precisely what pricing power looks like.
• Content growth is structural, not cyclical. 800V architectures in both vehicles and data-centre racks require more and higher-value protection per system. Winning high-single-digit transportation growth in a soft production environment is evidence of content and share gain, not end-market beta.
• Data centre is early. Design wins doubled in 2025 and the pipeline was “up meaningfully” again in Q1 2026, with the company participating from “grid to chip” across all three segments. Basler’s first design win with a US grid utility ships in 2027 — visibility, not hope.
• Self-help is genuine. Transportation segment margins have gone 5.0% → 8.7% → 12.5% in two years on price and productivity, not volume. That is real operational improvement under new leadership.
• The write-off clears the decks. A new CEO taking a $301.2M impairment in his first full year resets the base. Ex-impairment ROIC of 8.4% understates the forward return because the denominator now carries $397.0M less goodwill.
• The balance sheet is a genuine option. Net leverage of 1.0x and an $800M revolver allow the company to buy through a downturn when others cannot.

11.3 The strongest bear case

• The re-rating is priced off the impaired business. The market is applying a data-centre-power multiple to a company whose power-semiconductor unit is organically 19% smaller than in FY2023 and was written down by $301.2M six months ago.
• The returns do not justify the multiple. 8.4% ex-impairment ROIC, 0.9% GAAP, against 19.7x EV/EBITDA and 16.3x tangible book. The company has never sustained the mid-teens ROIC it now targets.
• Half the plan must be bought. ~$2.7bn of acquisitions at ~2.7x sales, by a management team paid nothing for return on capital, with a 22.9% historical impairment rate on deal spend.
• The adjustment has become the story. A GAAP-to-adjusted operating income gap that ran $2–17M for five years and then hit $108M and $320M is a quality-of-earnings signal, not a one-off.
• The tape is a cyclical, not a compounder. Three-year annualised return 11.9%, Sharpe 0.27, max drawdown 50%; beta 1.54; factor neighbours BorgWarner, Lear, Sensata, Vishay, Arrow. The one-year Sharpe of 1.97 is the outlier.
• Insiders are not buying. Four open-market purchases in five years, all by one director on one day, against $98.0M of sales — $32.4M of it in 2026 alone, the heaviest year on record, into the re-rating.
• Momentum has already broken. The stock is 17.2% below its 22 June high with a negative three-month return, on no company-specific news.

11.4 The factor-positioning read

This is where the tape informs the thesis rather than decorating it. FactorsToday shows beta 1.54, alpha −0.159, and model R² of 62.8% — meaning roughly 63% of the stock’s variance is explained by systematic factors, with idiosyncratic volatility of 26.3%. This is not an event-driven or company-specific story; the majority of the move is factor exposure. The nearest factor neighbours — Element Solutions 0.939, Vishay 0.938, BorgWarner 0.928, Sensata 0.926, Advanced Energy 0.921, Lear 0.915, Arrow 0.902, then robotics and semiconductor ETFs — describe an auto/industrial-cyclical and semi-capital-equipment cluster.

The inference for consensus positioning: the market is not holding Littelfuse as a quality compounder alongside Amphenol; it is holding it as a high-beta expression of the electrification and data-centre trade. That has two implications. First, the stock will de-rate with the theme rather than with company fundamentals — the 17.2% drawdown since June came without any company news. Second, the risk-adjusted history is the honest base rate: 11.9% annualised over three years with a 50% drawdown and a 0.27 Sharpe. A one-year Sharpe of 1.97 in a thematic upcycle is what the top of a cycle feels like, not a new regime. The dividend-yield loading of +0.685 is a further tell that part of the historical shareholder base owned this as a steady industrial compounder — a base that is not obviously the buyer at 4.15x sales.

11.5 The five assumptions that matter, and what falsifies each

Verdict: the variant perception is not that the business is bad — it is that the market has repriced the whole company on the strength of its weakest third. Consensus is right about the demand inflection, right about the moat in protection, and right that execution has improved sharply. Where consensus appears offside is in paying a compounder’s multiple for an 8.4%-ROIC cyclical whose five-year plan requires $2.7bn of acquisitions from a team paid nothing for returns on capital — and in treating a 79% one-year return as evidence of quality rather than of a crowded thematic trade that has already stalled.

12. Fact vs. Interpretation

13. Open Questions

1. What is data-centre revenue, in dollars? Management calls it the fastest-growing end market and reports design wins doubling, but has never disclosed the revenue base. Without it, neither the growth contribution nor the concentration risk can be sized. (Asked directly on the Q1 call; deferred to Investor Day; still not quantified in dollars.)
2. How much goodwill headroom remains, in percentage terms? (Partially resolved.) $1,211.4M of goodwill remains, of which $238.5M sits in Electronics-Semiconductor; Automotive Sensors now carries none; the other five units passed the September 2025 test. But no headroom percentages are disclosed for any unit — so the market cannot judge how close Industrial Controls and Sensors, which was impaired only a year earlier, is to a further charge.
3. What return threshold governs acquisitions? With ~$2.7bn to deploy and no ROIC metric in compensation, what hurdle rate does the board apply, and over what horizon?
4. What is the terminal shape of the power-semiconductor business? Management says rationalisation will cost 1–2% of revenue annually through 2030 and promises margin expansion “over the mid to long term.” Is the intention to shrink it to a profitable protection-adjacent core, or eventually to divest it?
5. Why did the Chief Legal Officer / EVP of M&A depart in January 2026, immediately after the Basler close and immediately before a five-year plan built on acquisitions?
6. Will Basler’s controls be clean in the FY2026 ICFR assessment? 9% of total assets was excluded from the FY2025 evaluation, one year after a material weakness in a non-US manufacturing entity.
7. How much of the 38% Q1 2026 incremental flow-through is price versus volume, and how durable is the price component if silver and copper reverse?
8. What is the pension and footprint-restructuring cash cost of the ongoing manufacturing consolidation? Charges are disclosed; cash outflows are less visible.

14. What Must Be True

14.1 For the bull case

14.2 For the bear case

15. Source Appendix

See Appendix B below for the full source list with URLs and access dates.

The analysis above contains no investment recommendation and no price target; the Claude's Take block is explicitly identified as the author’s own subjective opinion. This article is general information and not investment advice. The author may or may not hold a position in any security mentioned; readers should do their own research and consult a licensed adviser before making any investment decision.

APPENDIX A — Standard Diligence Questionnaire

Littelfuse, Inc. (NASDAQ: LFUS) — as of 2026-07-25

Supplemental to the analysis above. Fact / Interpretation / Assumption labels are applied where the distinction matters.

General

What thoughtful questions have other investors asked about this company?

From the Q1 2026 earnings call and Investor Day, the questions institutional investors are actually pressing on cluster into four areas:

1. Sizing the data-centre exposure. David Williams (Needham) asked directly: “is there a way to size that TAM or Littelfuse exposure across the entire data center footprint?” Management deferred to the Investor Day, where it provided SAM figures and percentage-of-market mix but still did not disclose data-centre revenue in dollars. (FACT.) This remains the single most-asked and least-answered question.
2. Whether Electronics margins can return to 2022 levels. Also Williams: “do you think you could ultimately get back to where you were maybe in 2022 in the lower 30% range?” The CFO explicitly declined to commit to the prior number, instead separating the segment into Passives (“we love the business and its margin profile”), the Protection franchise (“one of the most profitable”), and power semiconductors, where “that work takes time given factory consolidation.” (FACT.) The refusal to reaffirm the 2022 margin is informative.
3. Order durability. Christopher Glynn (Oppenheimer) probed book-to-bill sequencing repeatedly. Management confirmed sequential improvement Q4→Q1 and within Q1, book-to-bill “north of 1.0,” bookings +20% y/y. (FACT.)
4. M&A bandwidth. Glynn asked whether the company has capacity to pursue deals while integrating Basler, reorganising go-to-market and fixing power semis. Management said the pipeline is active and “you should expect to continue to see us doing acquisitions.” (FACT.)

Questions investors are not asking loudly enough, in my view (INTERPRETATION): why the incentive plan contains no return-on-capital metric; why the Dortmund fab acquired in December 2024 became the named driver of a $301.2M impairment twelve months later; and how a company earning 8.4% on invested capital reaches “mid-teens ROIC” while adding ~$2.7bn of goodwill-heavy acquisitions.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Neither extreme, but closer to a cyclical upswing than a trough. (INTERPRETATION.) FY2025 revenue of $2,386.3M remains 5.1% below the FY2022 peak of $2,513.9M, so revenue is not at a cycle high. But margins and order rates are inflecting sharply upward: Q1 2026 adjusted EBITDA margin of 22.9% (+180bp), GAAP operating margin 15.4% (+270bp), incremental flow-through 38% against a 30–35% long-term guide, book-to-bill “well above 1.0” and bookings +20%. A book-to-bill well above 1.0 with bookings growing at 20% is arithmetically unsustainable — that is a mid-upcycle reading, not a trough. (FACT for the metrics; INTERPRETATION for the cycle position.)

Driven by the external environment or internal actions? Both, in roughly equal measure, and the split is unusually legible. External: data-centre and grid demand, the end of distributor destocking, and FX (a 3pp tailwind in Q1 2026). Internal: genuine self-help — Transportation segment operating margin went 5.0% (FY2023) → 8.7% (FY2024) → 12.5% (FY2025) on “favorable price, volume leverage, cost reduction initiatives,” in an end market management itself calls “soft”; and the deliberate exit of low-margin automotive sensor and marine products. (FACT.) The Transportation improvement is the strongest evidence that the current management team can execute operationally. (INTERPRETATION.)

How stable are revenues? Not very. Revenue fell 12.5% in FY2019, 3.9% in FY2020, rose 43.9% in FY2021, rose 20.9% in FY2022, fell 6.0% in FY2023, fell 7.3% in FY2024, and rose 8.9% in FY2025. (FACT.) There is essentially no contracted recurring revenue; arrangements are “predominately short term.” Stability comes instead from design-in persistence — once specified, a part stays on the bill of materials — which produces stable share but not stable volume, because volume follows the customer’s build rate and the distribution channel’s inventory position.

Outlook for products/services? Favourable for protection content, mixed for power semiconductors. Higher-voltage architectures (800V in both vehicles and data-centre racks) structurally increase protection content per system. Against this, management guides to 1–2% of annual revenue drag through 2030 from deliberate product rationalisation in power semiconductors. (FACT.)

How big will this market be — growing, shrinking, domestic or international? Management estimates its serviceable addressable market at ~$22bn in 2025 growing to ~$30bn by 2030, a ~6.4% CAGR (FACT — company estimate, not independently verified). Growth, not shrinkage. Thoroughly international: the company sells across the Americas, Asia-Pacific and Europe, manufactures predominantly outside the US (the FY2024 material weakness arose at “one of its non-U.S. manufacturing legal entities”), and acquired assets in Germany and Ireland in the last three years.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? It depends entirely on which sub-industry. (INTERPRETATION, well supported.) Circuit protection: stable competitive intensity — fragmented, specification-gated, and not attracting new capacity. Power semiconductors: materially more competitive, with Infineon, onsemi, STMicroelectronics, Mitsubishi and Fuji operating at multiples of Littelfuse’s wafer volume and Chinese entrants adding capacity. Grid and data-centre power: currently supply-constrained and therefore benign, but this is exactly where capital is flooding in — the Marathon capital-cycle warning applies. Peer own-history valuations across the complex (Vishay 98.1st percentile, nVent 94.9th, Amphenol 94.0th, Hubbell 91.1st) are the clearest evidence that capital is being attracted. (FACT for the percentiles.)

How profitable is the business (ROIC, ROE)? Poorly, relative to the multiple being paid. (FACT for figures.)

Even the flattering ex-impairment figure of 8.4% is below any plausible WACC for a business with a 1.54 beta. The reason is the denominator: $1,806.3M of goodwill and intangibles, or 74.5% of book equity. The operating business earns respectable returns on tangible assets; shareholders earn poor returns on the capital committed, because much of it was spent buying other companies.

How profitable is the industry — how many competitors, what barriers to entry? Barriers in circuit protection are real but modest in absolute terms: safety-agency qualification (UL 248, IEC 60269, AEC-Q), design-in incumbency, and application engineering depth. They are sufficient to protect share but not to generate monopoly economics — Littelfuse’s best-ever consolidated operating margin was 20.3% (FY2022, adjusted). In power semiconductors, barriers are capital scale, which works against Littelfuse.

Can the business be easily understood? Yes. This is one of its attractions. The company makes physical components that protect circuits; revenue equals units times price; the drivers are end-market build rates, content per unit and distributor inventory. There is no complex accounting beyond the impairments and the ship-and-debit distributor rebate estimate. The one genuine complexity is the divergence between GAAP and adjusted operating income, which is not obvious to a casual reader and is misreported by major data services. (FACT — ROIC.ai reports FY2025 operating income of $357.6M against the audited GAAP figure of $37.5M.)

Can it be undermined by foreign low-cost labour? Partially, and it already has been in the commoditised segments. Automotive sensors have shrunk from $88.5M to $62.2M over three years on exit of “lower margin products”; power semiconductors face Chinese competition directly. The protection franchise is better insulated — not by labour cost but by qualification lock-in and the trivial cost of the part relative to the protected system. (INTERPRETATION.)

Do brands matter? Not as consumer brands, but the Littelfuse name carries genuine specification weight with design engineers, and acquired brands (Bussmann is the competitor analogue; Carling, Basler on the Littelfuse side) carry incumbency in their niches. Notably, the company impaired a tradename intangible in Industrial Controls and Sensors in FY2024 — a direct admission that at least one acquired brand was worth less than paid for. (FACT.)

What is the nature of competition? Design-in competition at the engineering level, then multi-year incumbency; price competition at contract renewal in transportation; and pure cost/capacity competition in power semiconductors. Management’s stated strategy is to move up the value chain by selling integrated sub-systems rather than discrete parts — the explicit Basler rationale (“allows us to move up the value chain, offering more comprehensive and higher power solutions”).

Customers’ switching costs? High relative to the part cost, which is the whole moat. Requalification and re-certification cost engineering months against per-unit savings measured in cents. This is genuine Greenwald customer captivity — the strongest single element of the investment case. (INTERPRETATION, supported by 18.3% FY2025 passive-products growth with simultaneous margin expansion.)

Financial Condition & Balance Sheet

Assets not fully recognised on the balance sheet? Yes, modestly. Internally-developed intellectual property, engineering know-how, safety-agency qualifications and the design-in positions on customer platforms are all expensed as incurred (R&D was $106.9M in FY2025, ~4.5% of revenue) and carry no balance-sheet value. The installed base of specified parts is the company’s most valuable asset and appears nowhere. (INTERPRETATION.) Conversely, and more importantly, the balance sheet carries $1,806.3M of goodwill and intangibles that are demonstrably worth less than stated — $430.4M has already been written off since 2018.

Off-balance-sheet liabilities? Nothing unusual. Operating leases are capitalised ($82.5M of right-of-use assets at Q1 2026). Pension liabilities are modest at $38.7M. Capital lease obligations total $83.2M. There is no securitisation, no material guarantee structure, and no litigation accrual of consequence disclosed. (FACT.)

How conservative is the accounting? Mixed, and this is a genuine concern. (INTERPRETATION.) Points against: (i) a material weakness in internal control over financial reporting as of 28 December 2024, arising from inventory cycle-count failures at a non-US plant, which the company said originated “in periods prior to 2023” and which produced $13.5M of out-of-period adjustments hitting FY2024 gross margin by 60bp; (ii) SEC comment-letter correspondence in October–November 2023; (iii) a GAAP-to-adjusted operating income gap that widened from $16.5M (FY2023) to $108.4M (FY2024) to $320.0M (FY2025). Points in favour: the material weakness was disclosed promptly via Form 12b-25, remediated within twelve months, and Deloitte & Touche LLP issued an unqualified ICFR attestation as of 27 December 2025; the impairments were taken decisively rather than deferred; and the annual incentive sales metric was defined to exclude the Basler acquisition. One residual caveat: Basler’s internal controls, representing ~9% of total assets, were excluded from the FY2025 ICFR assessment — a standard first-year exclusion, but material given the recent control history.

How CapEx-hungry is the business? Moderately, and less so recently — which bears watching. Capex has fallen from $104.3M (FY2022) to $86.2M, $75.9M and $67.6M in FY2025 — just 2.8% of revenue. (FACT.) For a company that owns wafer fabs and has committed to nearly doubling revenue by 2030, 2.8% is a light reinvestment rate, and part of the recent free-cash-flow strength is attributable to it. (INTERPRETATION.) Depreciation of $134.7M runs at roughly 2x capex, which is not sustainable indefinitely.

Capital Allocation & Management

How much FCF does the business generate, how does management use it, what is the philosophy? Free cash flow was $366.1M in FY2025 (15.3% of revenue), $291.7M in FY2024 and $371.2M in FY2023 — consistently 13–16% of revenue and genuinely strong. (FACT.) Uses in FY2025: $407.7M on acquisitions (Basler), $72.0M dividends, $27.6M buybacks, $67.8M debt repayment. The philosophy is explicitly acquisition-led: the 2030 plan calls for +6–8% annual revenue CAGR from M&A alongside 7–9% organic.

Significant acquisitions recently? Yes — six since 2018, totalling $1,882.1M of cash: IXYS (2018), Hartland Controls ($111.0M, 2021), Carling Technologies ($315.5M, 2021), Western Automation (~$162M, 2023), Dortmund Fab (~€94M, closed 31 Dec 2024) and Basler Electric ($350.3M, 11 Dec 2025, ~$130M sales, ~2.7x sales). (FACT.) Of that spend, $430.4M — 22.9% — has been impaired. The successes cluster in protection and control businesses adjacent to the core; the failures cluster in semiconductor manufacturing. The Dortmund fab is the starkest case: acquired December 2024, and named in the FY2025 10-K as the primary driver of the $301.2M impairment twelve months later.

Buying back shares? Barely, and not at all recently. Buybacks were $40.9M (FY2024), $27.6M (FY2025) and $0 in Q1 FY2026. (FACT.) Declining to repurchase at $400+ having bought at lower levels is defensible discipline — or simply cash preservation for the acquisition pipeline. Either reading is more favourable than repurchasing at the top. (INTERPRETATION.)

Issuing large amounts of new shares to insiders? No. Dilution is minimal: shares outstanding went 24.49M (FY2020) → 24.93M (FY2025), roughly 0.35% annually. Stock-based compensation was $27.3M in FY2025, or 1.1% of revenue — low for a technology-adjacent company. (FACT.) Total securities outstanding under equity plans are 706,945, about 2.8% of shares.

Compensation policy of directors/management? This is the weakest element of the entire investment case. (INTERPRETATION, on strong factual footing.) The 2026 DEF 14A contains zero occurrences of “ROIC,” “return on invested capital,” or “EBITDA.” FY2025 annual incentives for the CEO and CFO were weighted 30% AIP Corporate Sales, 30% AIP EPS, 20% AIP Cash Flow from Operations and 20% individual performance — no return-on-capital measure of any kind. Outcomes: sales $2,383M vs $2,265M target = 153%; AIP EPS $10.69 vs $10.05 target = 143%; operating cash flow $434M vs a $416M maximum = 200%.

The critical fact: management earned a 143% payout on “AIP EPS” of $10.69 in the fiscal year GAAP diluted EPS was −$2.89 — a $13.58 gap consisting of the impairment and related charges added back. In the year the company wrote off $301.2M of shareholder capital, the scorecard registered an excellent year.

Two credits are owed. AIP Corporate Sales was explicitly defined to exclude the Basler acquisition — a real and unusual discipline preventing management from buying its way to a sales bonus. And from FY2025 the long-term incentive shifted from stock options to PSUs earned on relative total shareholder return, in response to shareholder feedback. But a relative-TSR award pays out on a re-rating; combined with no ROIC metric, the scorecard now rewards revenue scale, adjusted EPS, operating cash flow and share price — exactly the four outputs a debt-funded acquisition programme delivers regardless of whether it earns its cost of capital.

Motivations of management? Hard to assess charitably on the evidence. Directors and executive officers together own 1.0% of the company (260,164 shares, at 27 February 2026); CEO Greg Henderson holds 5,917 shares and CFO Abhishek Khandelwal holds zero. Both are recent hires, which explains much of it, but the executives underwriting a five-year multi-billion-dollar capital plan have almost no personal capital at risk. (FACT.)

Across the complete five-year insider record (421 of 421 Form 3/4/5 filings parsed): 4 open-market purchases totalling $4,094,575 — all by one director, Anthony Grillo, on a single day, 7 February 2025 — against 235 sales totalling $98,020,560, a 24:1 ratio. Selling by year: $23.5M, $5.8M, $11.7M, $11.7M, $13.0M and $32.4M in 2026 year-to-date, the heaviest year on record with five months still to run, executed into the re-rating. Much of this is routine option-exercise-and-sell by executives with concentrated positions, and several large sellers were departing. But no operating executive has bought a share on the open market in five years. (FACT.)

Management turnover has been extraordinary: CEO, CFO and Chief Legal Officer all replaced within roughly fifteen months (Henderson Feb 2025, Khandelwal Jun 2025, D’Angelo May 2026). The FY2025 impairment was taken in the new CEO’s first full year and the 10-K explicitly frames it “in the context of a recent leadership transition and strategic reassessment” — consistent with a decks-clearing exercise, which is a favourable rather than unfavourable reading of that specific event. (INTERPRETATION.)

Valuation & Market Data

Is the stock an ADR, MLP, or K-1 issuer? No. Littelfuse, Inc. is a Delaware corporation listed on NASDAQ, filing 10-K/10-Q/8-K and DEF 14A. Ordinary common stock, 1099 reporting, no K-1. (FACT.)

Dividend policy? A growing, well-covered, low-yield dividend. Per share: $2.02 (2021), $2.26 (2022), $2.50 (2023), $2.70 (2024), $2.90 (2025), with $0.75 per quarter in Q1 2026 — an approximate 9% CAGR. (FACT.) Current yield ~0.74%. FY2025 dividends of $72.0M were covered 5.1x by free cash flow. Note the GAAP payout ratio is meaningless in FY2025 (net loss); on cash it is comfortable.

How profitable is the business? On margins, respectably: FY2025 gross margin 38.0%, adjusted operating margin 15.0%, adjusted EBITDA margin 20.6%; Q1 2026 gross margin 38.7% and GAAP operating margin 15.4%. On returns on capital, poorly: 8.4% ex-impairment and 0.9% GAAP in FY2025. The distinction between margin and return is the crux of this investment case — Littelfuse converts revenue to profit acceptably and capital to profit badly, because the capital base is inflated by acquisitions. (FACT / INTERPRETATION.)

Is net income diverging from cash from operations? Yes, dramatically, and in the favourable direction. FY2025 net income was −$71.7M against operating cash flow of +$433.8M — a ratio of −6.0x. (FACT.) This divergence is fully explained by the $302.1M of non-cash asset impairments plus $134.7M of D&A and $27.3M of stock compensation. It is not a red flag; it is the correct signal that the impairment is a write-down of historical purchase price rather than a deterioration in current cash economics. The genuine red flag lies elsewhere — in the fact that the adjusted earnings measure used for compensation and for most sell-side models excludes a charge that reflects real, permanently destroyed shareholder capital.

Where does the stock trade? At $404.86 (close, 2026-07-24) on 24.926M shares: market cap ~$10,092M, net debt $239.2M, EV ~$10,331M. EV/TTM sales 4.15x; EV/TTM EBITDA 19.70x; EV/TTM adjusted EBIT 26.69x; P/FY2025 FCF 27.6x (3.63% FCF yield); P/tangible book 16.3x. The P/E is not meaningful on negative trailing GAAP EPS. AZI own-history percentiles: P/S at the 93.7th percentile, P/B at the 86.2nd, composite 89.9th — with the P/E percentile null and therefore excluded, since trailing GAAP EPS is negative. (FACT.)

Risks & Downside

What factors would cause the stock to decline? In descending order of probability-weighted impact (INTERPRETATION):

1. Order normalisation. Book-to-bill falling below 1.0 as data-centre and distributor customers digest. Bookings +20% with book-to-bill “well above 1.0” cannot persist arithmetically. The 2023–24 analogue took Electronics revenue down 12.1% and segment margin from 22.3% to 14.3%.
2. Multiple compression. At the 93.7th percentile of its own price-to-sales history, in a peer complex where six of seven names sit above the 79th percentile, a thematic de-rating requires no company-specific news. The stock is already 17.2% below its 22 June high with a negative three-month return.
3. A disappointing acquisition. ~$2.7bn must be deployed to hit the 2030 plan; the historical impairment rate on deal spend is 22.9%.
4. Further impairment. $1,211.4M of goodwill remains, $238.5M in the unit just written down; no headroom percentages are disclosed.
5. Commodity or tariff shock. Silver and copper inflation is being managed to “price-cost neutral”; the April 2025 tariff selloff took the stock to a five-year low of $147.74.

Risk of a catastrophic loss? Low. (INTERPRETATION, well supported.) The balance sheet is the reason: net debt of $239.2M, net leverage ~1.0x, an $800M undrawn-capacity revolver refinanced in March 2026, $563.4M of cash and $366.1M of annual free cash flow. There is no refinancing wall, no covenant stress, no customer concentration disclosed as material and no single-product dependency. A cyclical downturn would compress earnings substantially — FY2024 showed GAAP operating margin falling to 7.2% — but would not threaten solvency.

Chance of a total loss? Negligible. A 99-year-old, cash-generative, investment-grade-profile component manufacturer with a defensible protection franchise, $2.4bn of revenue and modest leverage does not go to zero absent fraud. The realistic bear case is a de-rating toward $210–270, not impairment of the equity.

Recent News & Events

Has the business environment changed recently? Yes, materially and favourably on demand. Data centre and grid/utility infrastructure have emerged as the fastest-growing end markets: 2025 design wins in data centre were “up more than double year over year,” Q1 2026 delivered “strong double-digit growth” in both, book-to-bill was “well above 1.0,” bookings rose more than 20% y/y with sequential improvement through the quarter, and the Q2 guide of $690–710M implies 14% growth (8% organic). (FACT — management statements, unaudited.) Offsetting: soft residential HVAC, a soft global vehicle production environment, and silver/copper inflation.

Significant acquisitions? Basler Electric, closed 11 December 2025 — $350.3M net of cash for ~$130M of annualised sales in grid, utility, power-generation and data-centre electrical control and protection. It is performing ahead of underwriting: it contributed 39pp of Industrial segment growth in Q1 2026, management raised expected 2026 adjusted EPS accretion from $0.10–0.15 to $0.15+, and it carries high-teens adjusted EBITDA margins. (FACT.) Note that high-teens margins are below the 24–26% corporate 2030 target, so the deal is accretive to EPS but dilutive to the margin goal.

Change in accounting policies? No change in accounting policy, but a significant change in control environment. A material weakness in ICFR existed as of 28 December 2024 (inventory cycle counts at a non-US manufacturing entity), causing a late 10-K filing on Form 12b-25 dated 27 February 2025 and $13.5M of out-of-period adjustments in FY2024. Remediation was completed and attested as of 27 December 2025. The company adopted ASU 2025-05 with no material impact. (FACT.)

Recent changes — new markets, facilities, management?

• Markets: strategic pivot toward grid/utility infrastructure, data centre and aerospace & defence, formalised as three “high-growth opportunities” at the 14 May 2026 Investor Day, with a 2030 ambition of $4.5bn revenue, $1.1bn adjusted EBITDA, $600M FCF, >$25.00 EPS and mid-teens ROIC.
• Facilities: Dortmund 200mm wafer fab acquired (closed 31 Dec 2024); ongoing footprint optimisation and factory consolidation in power semiconductors; exit of the marine business in commercial vehicle.
• Management: CEO Greg Henderson (ex-Analog Devices) from 10 February 2025; CFO Abhishek Khandelwal from 18 June 2025; Chief Legal Officer Anne-Marie D’Angelo from 1 May 2026, replacing Ryan Stafford (who departed 30 April 2026 and has since become General Counsel of UFP Technologies). Holly B. Paeper (President, Commercial HVAC, Trane Technologies) joined the board 4 March 2026, expanding it to nine; the appointment 8-K disclosed ~$19.0M of sales to Trane and affiliates since 29 December 2024.
• Financing: amended and restated $800M senior unsecured revolving credit facility entered 12 March 2026. (All FACT.)

This appendix is supplemental to the analysis above. It contains no investment recommendation and no price target.

APPENDIX B — Source Appendix

Littelfuse, Inc. (NASDAQ: LFUS) — as of 2026-07-25

All sources accessed 2026-07-25 unless otherwise noted. Primary sources (SEC filings and company disclosures) are listed first, with primary sources prioritised. Third-party quantitative feeds are labelled as aggregated data and were reconciled to filings wherever a material number depends on them.

1. SEC Filings — Primary

Entity: Littelfuse, Inc., CIK 0000889331, Delaware. Fiscal year ends on the Saturday nearest 31 December (FY2025 ended 2025-12-27; FY2024 ended 2024-12-28).

1.1 Insider transaction corpus (Forms 3, 4, 5)

Complete parse: 421 of 421 filings covering 2021-07-25 → 2026-07-25, enumerated from the EDGAR filing index and retrieved directly (stylesheet path segment stripped from each URL to avoid local filename collisions). Base index: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0000889331&type=4

Aggregate non-derivative transaction census:

The four code-P purchases were all filed by director Anthony Grillo for transactions dated 2025-02-07 (3,307 @ $232.92; 8,704 @ $233.87; 5,306 @ $234.75; 183 @ $235.63).

1.2 SEC XBRL structured data

SEC “companyfacts” API, CIK 0000889331, retrieved 2026-07-25: https://data.sec.gov/api/xbrl/companyfacts/CIK0000889331.json

Tags relied upon for figures that differ from third-party presentations: us-gaap:OperatingIncomeLoss; us-gaap:RestructuringSettlementAndImpairmentProvisions; us-gaap:GoodwillImpairmentLoss; us-gaap:GoodwillImpairedAccumulatedImpairmentLoss; us-gaap:AssetImpairmentCharges; us-gaap:ImpairmentOfIntangibleAssetsExcludingGoodwill; us-gaap:Goodwill; us-gaap:IncomeTaxExpenseBenefit; us-gaap:NetIncomeLoss; us-gaap:IncomeLossFromContinuingOperationsBeforeIncomeTaxesExtraordinaryItemsNoncontrollingInterest.

Note on a material discrepancy: SEC XBRL reports FY2025 GAAP operating income of $37,528,000. Widely-used commercial data services report FY2025 “operating income” of $357,578,000, an ex-restructuring-and-impairment presentation. The memo reports both, labelled, and treats the XBRL/audited figure as authoritative.

2. Company Disclosures — Primary

Management commentary from calls and presentations is treated throughout as hypothesis requiring external validation, not as evidence. Company estimates (SAM sizing, 2030 targets, end-market growth rates) are labelled as such wherever cited.

3. Quantitative Data Feeds — Third-Party (Aggregated, Not Primary)

All third-party statistical and aggregated figures are reportable as facts about the data; forward inferences drawn from them are labelled interpretation and are subordinate to the fundamental analysis. No figure from any of these feeds is used as a price target or entry/exit level.

4. Peer and Industry References

Peer own-history valuation percentiles used above (AZI valuation_index, 2026-07-24): Amphenol (APH) composite 94.0th; nVent Electric (NVT) 94.9th; Hubbell (HUBB) 91.1st; Vishay Intertechnology (VSH) 98.1st; TE Connectivity (TEL) 79.8th; Sensata Technologies (ST) 59.8th. Peer market capitalisations and relative-strength data from FactorsToday /stock-info/, 2026-07-25.

Competitor set identified from the FY2025 10-K, the 2026 Investor Day presentation and sector knowledge: Eaton (Bussmann), Mersen, Bourns, TDK, Vishay and Amphenol in circuit protection and passive components; Infineon, onsemi, STMicroelectronics, Mitsubishi Electric and Fuji Electric in power semiconductors; Sensata and Aptiv in transportation electrical and sensing; Schweitzer Engineering Laboratories, ABB, Schneider Electric and GE Vernova in protective relays, excitation and grid equipment.

5. Financial Media and Secondary Commentary

Used only for market-positioning and consensus context, never as evidence for a financial fact:

6. Analytical Frameworks Applied

• Competition Demystified (Bruce Greenwald & Judd Kahn) — barriers to entry as the dominant consideration; the three genuine advantage types (supply/cost, demand/customer captivity, economies of scale reinforced by captivity); the market-share-stability and ROIC tests. Applied to name the moat type as customer captivity in circuit protection and to conclude its absence in power semiconductors.
• Capital Returns (Edward Chancellor / Marathon Asset Management) — supply-side capital-cycle analysis; high returns and high valuations attract capital and mean-revert. Applied to locate the electrification and grid supply chain in the capital-attracting phase, evidenced by uniformly extended peer own-history valuations.

7. Data Integrity Notes

1. Complete insider parse. 421 of 421 Form 3/4/5 filings in the five-year census were retrieved and parsed. No insider conclusion rests on a partial sample. An initial automated mirror captured only 245 of 421; the shortfall was identified by counting the EDGAR census rather than the files on disk, and corrected.
2. GAAP vs. adjusted operating income. The FY2025 divergence ($37.5M vs $357.6M) is the single largest source of potential error in reading this company. All operating-margin and ROIC figures in the memo state which basis is used.
3. Unreconciled management statement, excluded. On the Q1 FY2026 call the CFO stated the company “returned $90 million to shareholders through our dividend.” The Q1 FY2026 10-Q reports $18.836M of dividends paid and zero share repurchases for the quarter. The $90M figure cannot be reconciled to the filing and is not used anywhere in this report; the filed figure governs.
4. Own-history percentile caveat. The AZI composite percentile for LFUS is computed from only two components (P/B and P/S) because trailing GAAP EPS is negative, making the P/E percentile null. This is disclosed wherever the percentile is cited.
5. Sources are public. Every source cited in this article is publicly available: SEC filings, company presentations and press releases, and public market-data services. No non-public or proprietary information was used.

This appendix lists the public sources underpinning the analysis. It contains no investment recommendation and no price target.