STMicroelectronics N.V. — 5.4/10
STMicroelectronics is a broad-line European semiconductor maker at the bottom of a deep automotive/industrial downcycle that is now inflecting on the top line. Revenue YoY has swung from -27.4% (Q1'25 trough) to +23.0% (Q1'26), with volume (+17%) and ASP (+6%) both turning positive and three of four product segments growing 23-34% YoY. After the Q1'25 reorganization the company reports four segments: AM&S (analog + MEMS, ~43% of revenue), Embedded Processing (STM32 MCUs, ~31%), RF & Optical Communications (~13%), and Power & Discrete (incl. silicon carbide, ~13%).
The core tension: this is a genuine, improving cyclical recovery whose thesis is still a consensus promise funded by debt against negative free cash flow. STM passes the oligopoly gate — it is the #1 silicon-carbide maker at ~29-32.6% share — but that leadership sits in only ~13% of revenue, while its two largest pools are markets where it is a sub-20% #3-#4 follower behind TI, ADI, and Infineon. Reported Non-GAAP FCF turned negative (-$249M FY25, -$723M Q1'26 on the NXP MEMS acquisition), the single most important quality flag. Held to 5.4/10.
| CEO | Jean-Marc Chery (since 2018) | Revenue Growth | Re-accelerating (+23% Q1'26) |
| Secular Exposure | AI DC / Silicon photonics / LEO | FCF Trajectory | Negative (-$249M FY25) |
| SiC Position | #1, ~30% share (oligopoly) | FYE | December 31 |
| Quality Gate | PARTIAL PASS (1 NO: FCF) | Margin Trend | Stabilizing, ~1,600bps below peak |
| Dimension | Score | Weight | Weighted |
|---|---|---|---|
| Financial Trends | 4 | 25% | 1.00 |
| Thematic Exposure | 5 | 35% | 1.75 |
| Management Quality | 7 | 20% | 1.40 |
| Investor Sentiment (Inverted) | 7 | 5% | 0.35 |
| Concerns / Risks | 6 | 15% | 0.90 |
| Composite | 100% | 5.4 |
A textbook cyclical-trough-to-recovery setup: revenue YoY re-accelerated five straight quarters from -27% to +23%, with volume (+17%) and ASP (+6%) both inflecting in Q1'26 and three of four segments growing 23-34% YoY. Layered on top are three structural AI growth vectors — data-center silicon photonics, the AWS engagement, and LEO satellites. Held to 5.4/10 by four drags: (1) fails the positive-and-growing-FCF gate — reported Non-GAAP FCF turned negative (-$249M FY25, -$723M Q1'26); (2) the oligopoly leadership (#1 SiC) sits in only ~13% of revenue while STM is a sub-20% follower in its two largest pools (Thematic 5/10); (3) margins are merely stabilizing ~1,600bps below peak with op margin barely positive (Financial 4/10); and (4) a real >30%-China / export-control / government-ownership / negative-credit-outlook overhang (Concerns 6/10).
Quality gate: PARTIAL PASS (1 NO). Oligopoly YES (narrow — #1 SiC). Positive & growing FCF NO. Management track record YES. One NO → no composite cap; the negative-FCF gap is the single material quality blemish and is flagged prominently.
STM's recovery is genuine and improving, and management quality (7/10) and the management-street divergence (7/10) are the standout positives — a stable, beat-and-raise team putting capex (~$2.2B for 2026) and OpEx ahead of a specific AI-datacenter / silicon-photonics / LEO thesis the street has not yet underwritten (targets sit below where the recovery math implies). That is the correct contrarian shape.
But the bull case lives entirely in the second-derivative. Trailing financials do not yet support a high score: full-year operating income is -96% from the FY23 peak, gross margin is stuck ~33-35% (underutilization, SiC price erosion, mix, a $140M capacity-reservation-fee headwind), and free cash flow has gone negative even as revenue rips +23%. The structural read is the binding one — STM is the #1 player in silicon carbide, but SiC is only ~13% of revenue; in the ~74% of the business that is analog and MCUs it is a #3-#4 follower behind TI, ADI, and Infineon. "Don't settle for the 2nd, worse the 3rd."
This is a watch-list name: a real recovery with real AI optionality, priced against negative cash generation, where the catalysts are credible but unconverted to cash and the leadership does not extend to the businesses that drive the P&L.