
Renesas Electronics
Fab-lite semiconductor IDM: in-house mature-node fabs plus TSMC foundry for advanced nodes; sells MCUs/SoCs/analog/power into automotive (~48% of revenue) and industrial/infrastructure/IoT (~52%), increasingly bundled as 'winning combination' reference designs; recurring software revenue via Altium (PCB/electronics design cloud)
Earnings, margins, COGS & capex
FY2025 (CY) revenue fell 2.0% IFRS to JPY 1,321.2B as automotive (-9.0% non-GAAP, to JPY 639.7B) offset industrial/infrastructure/IoT growth (+5.5%, to JPY 671.8B), but non-GAAP gross margin rose 1.6pts to 57.6% on manufacturing cost cuts. GAAP swung to a JPY 51.8B net loss - the first annual loss in six years (vs JPY 139.5B net income in FY2024) - driven by the JPY 236.6B loss (incl. FX effects) from converting the ~$2.06B Wolfspeed SiC wafer deposit into other financial assets (notes, ~16.9M shares, warrants) after Wolfspeed's Chapter 11. Underlying earnings stayed strong: non-GAAP net income JPY 329.3B (25.0% margin), FCF JPY 328.2B. Q1 2026 marked a sharp cyclical inflection: non-GAAP revenue +20.6% YoY with operating margin back to 33.7% on demand from both automotive and industrial/infrastructure/IoT, including AI/data-center infrastructure. Portfolio reshaped in 2026: the IDT-heritage Timing product business was sold to SiTime for ~$2.9B cash-and-stock (announced Feb 5 2026, completed Jul 1 2026) and is excluded from non-GAAP figures from Feb 2026.
Income statement — where each revenue dollar goes
% of revenueOf every $1 of revenue, ~43¢ is cost of goods and ~42¢ operating expense, leaving ~15¢ of operating profit (~25¢ net).
Revenue trend
Margins
up - +1.6pts YoY FY2025 on manufacturing cost reduction despite lower auto utilization; Q2 2026 implied ~57.0% (H1 guide 58.1%) on mix/cost factors
recovering - Q1 2026 operating profit +49.6% YoY to JPY 125.4B; Q2 2026 implied ~29.0% (H1 guide 31.3%) on opex step-up
gap vs non-GAAP is JPY 111.4B acquisition amortization/PPA depreciation + JPY 37.1B SBC + JPY 39.9B other non-recurring items
stable underlying; GAAP net was -3.9% due to the one-off Wolfspeed loss
stable-to-up
COGS structure
COGS ~42% of revenue (57.6% non-GAAP gross margin, FY2025). Cost base is fab-lite: in-house Japanese fabs (Naka, Kofu 300mm power line, etc.) for mature-node MCU/analog/power, TSMC foundry for advanced-node SoCs/MCUs (28nm and below). FY2025 gross margin expanded despite lower automotive utilization because of manufacturing cost reduction; management flags utilization, mix and FX (154 JPY/USD planning rate) as the main swing factors.
Capex
Modest by IDM standards: FY2025 net investing cash outflow JPY 124.7B (~9% of revenue), mainly PP&E and intangibles - versus JPY 1,284.1B in FY2024, which was dominated by the Altium acquisition (~$5.9B). No megafab program; capacity strategy leans on foundry partners and reactivated in-house lines (Kofu 300mm) for power semis. Pure PP&E capex not separately disclosed in the summary release.
Latest earnings
Beat the company's own Feb 5 guide: on a comparable basis (including the Timing business, which was moved out of non-GAAP from Feb 2026) revenue was JPY 379.9B vs the JPY 375.0B guide midpoint (+1.3%), and non-GAAP operating margin of 33.7% exceeded the ~32% guided; some third-party coverage still flagged the GAAP-vs-non-GAAP profitability gap
H1 2026 guide (given at Q1): non-GAAP revenue JPY 752.8-767.8B (+18.9% to +21.2% YoY), non-GAAP gross margin 58.1% and operating margin 31.3% at midpoint - implying Q2 (Apr-Jun) revenue ~JPY 388.0B (+~19.5% YoY), gross margin ~57.0% and operating margin ~29.0%, i.e. QoQ margin fade on mix and opex step-up. Renesas gives quarterly-basis forecasts in lieu of a full-year guide. FY2026 dividend remains undecided per the company (FY2025 paid a JPY 28/share year-end dividend).
- Q1 2026 non-GAAP revenue
- JPY 372.3B, +20.6% YoY (JPY 380.3B IFRS, +23.2%)
- Q1 2026 non-GAAP operating margin
- 33.7% (operating profit +49.6% YoY)
- FY2025 GAAP net result
- JPY -51.8B, first annual loss in six years (Wolfspeed JPY 236.6B conversion loss incl. FX); non-GAAP net income JPY 329.3B
- FY2025 free cash flow
- JPY 328.2B (~25% of revenue)
- Net debt (Dec 2025)
- ~JPY 930.9B; debt/equity 0.50; interest-bearing debt down JPY 196.0B YoY
- Timing business sale to SiTime
- ~$2.9B cash-and-stock; announced Feb 5 2026, completed Jul 1 2026
Growth drivers
- AI data-center infrastructure — power management, connectivity and memory-interface content per server rack - a visible engine of the Q1 2026 rebound (the clock/timing product line itself was divested to SiTime in Jul 2026)
- Automotive semiconductor content growth (EV/hybrid powertrain, ADAS, software-defined vehicle gateways) recovering off the 2024-25 inventory correction
- New R-Car/RH850 automotive SoC-MCU roadmap and RA/RX 32-bit MCU share in industrial and edge-AI endpoints
- Altium: recurring cloud software revenue and a design-in funnel that seeds Renesas silicon into boards designed on the platform
- Winning-combination cross-selling of the Intersil/Dialog analog-mixed-signal and power portfolio attached to its MCU/SoC sockets
Bull & bear
A top-quality embedded franchise exiting a two-year downcycle with 20%+ growth, record-class margins, an AI data-center power kicker, a portfolio actively high-graded (Timing sold to SiTime at ~$2.9B), and a one-off Wolfspeed loss obscuring JPY 329B of underlying earnings power - at a multiple below US analog peers.
- Cyclical inflection is confirmed, not hoped-for: Q1 2026 non-GAAP revenue +20.6% YoY with operating margin back to 33.7% (a beat vs the ~32% guide), and H1 guided +19-21% - operating leverage is compounding off a lean cost base
- AI data-center infrastructure is a structural new leg: power-management and memory-interface content rides rack power density, largely uncorrelated with the auto cycle - and management showed discipline by selling the sub-scale clock/timing line to SiTime at a rich ~$2.9B rather than subsidizing it
- Wolfspeed hit is one-off and already expensed: JPY 236.6B booked in FY2025, converted into Wolfspeed notes, ~16.9M shares and warrants with recovery optionality; cleaner GAAP comparisons from 2026
- Balance sheet self-repairs fast: JPY 328B FCF and JPY 196B debt paydown in a DOWN year, plus ~$2.9B of SiTime consideration landing from Jul 2026 - net debt of ~JPY 931B can be largely neutralized within a few years at this run-rate
- Valuation: ~26x trailing non-GAAP EPS (JPY ~182 on FY2025), falling to high-teens forward if the Q1 +49.6% operating-profit growth carries into FY2026 - a discount to TI/ADI/MPWR for comparable gross margins, with Altium's recurring software not separately valued
A cyclical automotive-heavy supplier whose rebound is partly inventory restocking, whose SiC strategy just destroyed ~$2B of capital, which sold one of its best AI-levered product lines, and whose GAAP earnings, China exposure and undecided dividend argue the non-GAAP multiple flatters it.
- Restock, not demand: FY2025 automotive fell 9.0%; a +20% Q1 against a depressed comp says little - the H1 guide already implies Q2 margin fade (GM to ~57.0%, OM to ~29.0%) as costs step up
- Wolfspeed was not just a write-down - it was the SiC supply strategy: Renesas now lacks a secured SiC wafer anchor precisely as Infineon, onsemi and STMicro scale EV power share; the JPY 236.6B loss is sunk but the roadmap gap is not
- Selling the Timing unit to SiTime for ~$2.9B hands away a genuinely AI-levered, high-margin franchise (IDT heritage, cloud/data-center clocking) right as that demand accelerates - proceeds are real but the growth mix is now more auto-cyclical
- GAAP reality: 15.2% operating margin and a JPY 51.8B net loss in FY2025; JPY ~148B/yr of acquisition amortization and SBC is a recurring cost of the M&A-built portfolio, not noise
- Capital-allocation overhang: serial large M&A (IDT $6.7B, Dialog ~$5.9B, Altium ~$5.9B) plus the Wolfspeed loss, and an FY2026 dividend still undecided at the last print - shareholder-return visibility ranks behind the next deal; China localization and export controls squeeze the largest growth region on both ends
What it is worth
Comps on non-GAAP earnings plus reverse-DCF sanity check; peer set NXPI / IFNNY / STM / TXN / MCHP
~JPY 3,300 (-30%)
auto relapse + China pricing pressure hold EPS near JPY 170-175 and the multiple compresses to ~19x cyclical-trough treatment
~JPY 5,000 (par)
EPS ~JPY 210 on mid-teens FY2026 growth fading to high-single-digit, ~24x
~JPY 6,500 (+35%)
FY2026-27 EPS ~JPY 240 on sustained 15-20% growth + AI-infra mix, re-rated to ~27x toward US analog peers
At JPY 4,817 (market cap ~JPY 8.72T / ~$53.7B at ~162 JPY/USD), Renesas trades at ~26x trailing non-GAAP EPS (~JPY 182 on FY2025 non-GAAP net income of JPY 329.3B) and roughly high-teens-to-20x forward if Q1 2026's ~+50% non-GAAP operating-profit growth carries into FY2026 - broadly in line with NXP/Infineon, below TI/ADI/MPWR despite comparable gross margins. Forward comparisons must adjust for the Timing divestiture (excluded from non-GAAP from Feb 2026; ~$2.9B consideration received). Reverse-DCF read: the current price implies roughly high-single-digit long-run revenue growth with ~30% sustained non-GAAP operating margin - achievable if the AI-infrastructure leg is structural, demanding if the rebound is restock. Scenario values are illustrative, non-GAAP-EPS x multiple, 12-18mo horizon. Not financial advice.
SWOT
Strengths
- Top-tier global share in automotive MCUs (roughly #1-2 with NXP/Infineon) with decade-long design-in stickiness at Toyota/Denso, Honda, Bosch, Continental and global OEMs
- Structurally high profitability for the sector — 57-59% non-GAAP gross margin, ~30%+ non-GAAP operating margin at the Q1 2026 run-rate, ~25% FCF margin
- Broadened portfolio via disciplined M&A (Intersil, IDT, Dialog, Altium, Transphorm) - analog, power, connectivity and design software around the MCU core - with proven willingness to prune: the IDT timing unit was monetized to SiTime at ~$2.9B
- Fab-lite balance of in-house mature-node capacity (cost + supply control) and TSMC advanced nodes (no megafab capex burden)
- Rapid deleveraging — interest-bearing debt down JPY 196B in FY2025 to 0.50 debt/equity while still generating JPY 328B FCF in a down year
Weaknesses
- FY2025 GAAP net loss of JPY 51.8B — the first in six years; the Wolfspeed ~$2.06B deposit conversion (JPY 236.6B loss incl. FX) showed real capital-allocation risk in the SiC strategy
- SiC/EV power roadmap set back — the Wolfspeed wafer-supply anchor is now restructured paper (notes/equity/warrants) rather than secured supply
- Automotive, ~48% of revenue, fell 9.0% in FY2025 and remains hostage to OEM production schedules and inventory cycles
- Large gap between IFRS (15.2%) and non-GAAP (29.3%) operating margin - JPY 111.4B acquisition amortization plus JPY 37.1B SBC per year makes statutory earnings look thin
- Dividend policy for FY2026 explicitly undecided at the FY2025 print - weaker shareholder-return visibility than US analog peers
Opportunities
- AI data-center buildout — power management and memory-interface content per GPU rack is a fast-growing line, visible in the Q1 2026 infrastructure-led rebound
- Software-defined vehicle transition raises per-car silicon + software content; Renesas' R-Car Gen5 and crossover MCU roadmap targets it, including the Honda high-performance SDV SoC co-development
- Altium + silicon — a cloud design-to-BOM ecosystem that can convert the electronics-design funnel into recurring revenue and silicon pull-through
- Edge-AI MCUs/MPUs (RA8, RZ/V with DRP-AI accelerators) in industrial automation and robotics
- Japan semiconductor policy tailwind (subsidies, supply-chain onshoring) supports domestic fab economics; ~$2.9B of SiTime deal consideration adds balance-sheet firepower
Threats
- Chinese MCU and analog localization (e.g. GigaDevice-class 32-bit MCU vendors) eroding pricing in mid-range auto/industrial sockets in the China market - competitive context, not a call
- US-China export-control and tariff regime could fragment automotive/industrial demand and its China revenue exposure
- Auto cycle relapse — Q1 2026's rebound is partly restocking; a renewed OEM cut would hit the highest-margin franchise
- Competition from scale analog/embedded peers (TI's 300mm cost curve, Infineon's power leadership, NXP's S32 auto platform)
- FX: yen appreciation versus the 154 JPY/USD planning rate would compress reported revenue and margins (spot was ~162 in early Jul 2026, currently a tailwind - the risk is a reversal)
Moats, dependencies & bottlenecks
Moats
Safety-qualified MCUs/SoCs are designed into vehicle platforms for full model cycles; requalification cost keeps incumbents locked in
RA/RX/RH850 toolchains, AUTOSAR stacks and 30+ years of customer firmware create high migration friction
MCU + Intersil/Dialog power/analog + connectivity sold as validated reference designs raises attach rates and deal size (timing devices exit the bundle with the SiTime sale)
Owned Japanese fabs give cost and supply-assurance advantages on 40nm+ auto/industrial parts where capacity is scarce in shortages
Leading PCB-design tool creates a funnel from board design to Renesas BOM placement; silicon pull-through still unproven at scale
Dependencies
Advanced-node (28nm and below) MCUs/SoCs, including R-Car, are foundry-sourced; Taiwan concentration risk shared with the whole industry
Honda HMC, Denso DNZOY, Bosch, Continental) ~48% of revenue; OEM production cuts flow through with a lag via Tier-1 inventory
supplier-turned-investee SiC wafer supply deal restructured in Wolfspeed's 2025 Chapter 11; ~$2.06B deposit converted to notes, ~16.9M shares and warrants (JPY 236.6B loss incl. FX) - SiC sourcing must be rebuilt
Cortex cores across RA MCUs and R-Car; partially hedged by proprietary RX/RH850 cores and RISC-V options
Plans at 154 JPY/USD (spot ~162 in Jul 2026); yen appreciation compresses reported revenue and gross margin
Meaningful China revenue in auto/industrial; localization pressure and US-Japan export rules both bite
Advantages
- Sector-leading non-GAAP gross margin (57-59%) rivaling US analog leaders
- Deep multi-decade automotive relationships in Japan plus global reach from Intersil/IDT/Dialog
- FCF conversion ~25% of revenue funds deleveraging and R&D step-up simultaneously
- Portfolio breadth spanning MCU, SoC, analog, power, connectivity and design software - one-stop embedded supplier
- AI data-center power-management position now scaling with rack power density, plus ~$2.9B of SiTime deal consideration from monetizing the timing line
Weaknesses
- GAAP profitability materially below non-GAAP optics (15.2% vs 29.3% operating margin, FY2025)
- Capital-allocation credibility dented by the JPY 236.6B Wolfspeed loss
- Auto concentration (~48%) with a -9.0% FY2025 print
- FY2026 dividend undecided at the last report; thinner committed shareholder-return program than US peers
- Serial-acquirer integration load (Altium, Transphorm) alongside a cyclical recovery
Bottlenecks
- Automotive cycle timing — the largest, highest-visibility franchise recovers on OEM schedules Renesas does not control
- SiC power roadmap gap post-Wolfspeed — substrate supply must be re-secured (Transphorm GaN helps but does not replace SiC)
- Quarterly-basis guidance (no full-year frame) limits investor visibility versus peers
- Acquisition amortization (JPY 111.4B in FY2025) suppresses GAAP earnings and constrains statutory payout capacity
- Advanced-node capacity allocation at TSMC during industry upcycles
Top signals & trends
Top signals
First test of whether the rebound is demand or restock
The structural leg of the bull case post-Timing-sale; management color on hyperscaler design wins is the tell
First big capital-allocation decision after the Wolfspeed loss; also watch the mark on the SiTime stock consideration
Explicitly undecided at the Feb 2026 print; setting/raising it would signal balance-sheet confidence post-Wolfspeed
Recovery value on the notes/~16.9M shares/warrants plus any new substrate agreement de-risks the power roadmap
Management and peer (NXP/Infineon) commentary converging on end of digestion
Spot ~162 is currently a tailwind; yen strength back through 154 would be a direct revenue/margin headwind
Trends
More power stages and memory-interface devices per rack; Renesas' power-management portfolio is directly levered (clock/timing devices now belong to SiTime)
Rising MCU/SoC and power content per vehicle over the cycle, even with unit softness
RA8/RZ-V class MCUs/MPUs with NPU acceleration expand ASPs in the II&I segment (+5.5% in FY2025)
Domestic MCU/analog substitution pressures price and share in mid-range China sockets
Subsidy tailwind for domestic capacity, but export-control fragmentation cuts addressable demand
Large opportunity, but Renesas' SiC path was set back by Wolfspeed; GaN (Transphorm) is the nearer-term play
Ecosystem & competitor graph
Suppliers feed the company; customers pull from it. Line thickness shows the strength of each tie (supply-chain dependency, customer earnings contribution). Hover to isolate a tie.
Foundry for advanced-node SoCs/MCUs (R-Car, 28nm-and-below MCUs)
Silicon wafer supply
Silicon wafer supply
Fab equipment for in-house Japanese fabs
Former SiC substrate anchor; deposit converted to notes/equity/warrants in the 2025 Chapter 11 restructuring
OSAT assembly/test partner
OSAT assembly/test partner
Via Denso and direct - flagship auto MCU/SoC relationship
Largest Tier-1 channel; historically also a Renesas shareholder
Auto MCU/SoC customer and SDV SoC co-development partner
Global Tier-1 customer for auto control MCUs
Global Tier-1 customer
Tier-1 customer in vehicle architecture/ADAS
Dell DELL, Super Micro SMCI board makers) End demand for data-center power devices, largely via ODMs
Closest comp: auto MCU/processor (S32) and industrial-edge rival with similar auto mix
Auto semi #1 by revenue; power-semiconductor and SiC leadership where Renesas is weakest
Auto/industrial MCU (STM32) and SiC incumbent; aggressive in the same embedded sockets
Analog/embedded scale player with 300mm cost advantage pressing catalog analog pricing
Broad MCU + analog catalog competitor in industrial/auto mid-range
High-performance analog/mixed-signal competitor in industrial, comms and auto
Direct rival in AI data-center power management - the contested growth leg
Auto power/sensing and SiC competitor
Representative of China MCU localization pressure in mid-range sockets - named for competitive context only, not a recommendation