STMicroelectronics at 374x: The Arithmetic of a 60 Euro Print

The stock prints 59.84 EUR against trailing diluted EPS of 0.16 EUR. That is a 374x trailing multiple. The consensus target average sits at 46.72 EUR; the high end at 58.38 EUR. Price has cleared even the most generous sell-side anchor. So before anything else, I want to do the arithmetic the multiple is asking me to accept.

Start with the company’s own forward signal. Q1 2026 revenue came in at 3.10 billion dollars, up 23% year-on-year, and Q2 guidance midpoint is 3.45 billion, implying a roughly 24.9% YoY run-rate into mid-year. Management has put explicit numbers on the AI and datacenter ramp: more than 500 million dollars in 2026, more than 1 billion in 2027. Annualize the Q2 guide and you land near 13.8 billion dollars in revenue, against TTM revenue of 12.38 billion. So the top-line story is real but not yet transformative on the print — roughly 11-12% growth above trailing, with management implying acceleration.

Now the multiple. At 59.84 EUR on 0.16 EUR TTM EPS, the market is implying one of two things: either trailing earnings power is wildly understated by the cycle, or forward earnings need to compound at a rate that makes 374x collapse into something defensible inside five years.

If I want this to land at a mid-cycle multiple of, say, 20x on forward earnings five years out, EPS has to reach roughly 3.00 EUR. From 0.16 EUR, that is a compound annual growth rate of about 79% for five consecutive years. Even if I grant that 0.16 EUR is a cyclical trough rather than normalized earnings power, and I reset the base to a mid-cycle 2.00 EUR, getting to 3.00 EUR in five years still requires about 8.4% annual EPS growth on top of full margin recovery. Two assumptions, both contestable: that margins fully normalize from the current 4.93% operating margin back toward the high-teens the company used to print, and that the AI/datacenter ramp compounds on top without cannibalizing the legacy automotive and industrial mix.

The operational picture does not yet support either leg. TTM operating margin is 4.93% on 12.38 billion dollars of revenue. Inventory days stand at 140 as of Q1 2026 quarter-end, which tells me the channel is still working through stock; utilization at the fabs cannot be at the level needed to drive operating leverage back to historical norms. Net income growth TTM is negative 86.64%. The shrink in earnings is not noise; it is the cycle. And the cycle has not turned in the operating data, only in the guide.

The bull-case arithmetic does have a spine. Q1 beat on personal electronics and CECP. The NXP MEMS sensor acquisition closed. The AWS multi-year engagement, with warrants issued, gives the datacenter number a named anchor rather than a hopeful slope. Stack a successful design-win cycle for SiC power discretes in automotive, qualify the new MEMS portfolio at scale, sample the AI-adjacent products into hyperscaler tape-outs, and the 2027 EPS trajectory can plausibly step to 2.50-3.00 EUR. At that earnings level, 60 EUR is a 20-24x forward multiple, which is defensible for a structurally growing IDM with automotive exposure above 40% of revenue.

The assumption that breaks is margin.

To clear 2.50 EUR of EPS by 2027, operating margin has to recover from 4.93% to something in the 15-18% zone on a revenue base around 15 billion dollars. That requires utilization to climb back to the 80%+ range, pricing to hold against a customer base that has spent eighteen months destocking, and the SiC ramp to deliver gross-margin accretion rather than dilution as new capacity qualifies. Three variables, all moving, none de-risked by the Q1 beat alone.

The discount-rate environment compounds the math problem. Long-duration earnings get penalized as policy rates stay restrictive, and European semiconductor names are now being priced on AI-adjacent growth multiples drawn from a US rate curve that has not actually softened. The US Section 232 investigation into semiconductor imports is active, with 25% duties on certain advanced chips already shaping cost pathways, and the company explicitly excluded further tariff impacts from its Q1 outlook. The EU Chips Act rollout offers some offset through state aid for European fabs, but that subsidy stream is slow and tied to a contested restructuring of the legacy French and Italian sites. None of this lowers the discount rate the 3.00 EUR EPS target needs to clear.

One accounting note. The reported figures are in US dollars on the operating lines but the ADR trades against a Euro share price, and the company is Dutch-incorporated with Swiss headquarters. That means any reader pulling the 0.16 EUR EPS figure against the 59.84 EUR price is comparing a dollar-translated earnings number to a Euro-quoted price at a 1.1633 EUR/USD print. The 374x is real, not an FX illusion, but the margin recovery path will be reported in dollars while the multiple compresses in Euro.

To justify 59.84 EUR on any reasonable five-year forward multiple, the market is pricing roughly 70-80% EPS compound growth from the trough, or a full reversion to mid-cycle earnings power plus roughly 8-10% structural growth on top. That is not impossible. It is the kind of arithmetic that has historically corrected hard when a single quarter of utilization data disappoints.

If the company prints 2026 operating margin below 12% on the full year, the EPS path to 2.50 EUR by 2027 mathematically fails and the 60 EUR print becomes indefensible on any mid-cycle frame. That is the trigger I am watching.

© Mathstock