ON Semiconductor's stock looks expensive at current levels — until investors factor in how much fuller its factories could run, according to an analysis from Trefis.

The investment research firm's central claim is straightforward: the gap between ON's present valuation and its underlying worth narrows considerably once you assume its wafer fabs return to higher utilization rates. Factories that sit partially idle carry heavy fixed costs; the same factories running near full load convert a much larger share of each revenue dollar into gross profit. Trefis frames this operating-leverage dynamic as the key variable that the market, at current prices, has not fully priced in.

That framing matters for a company with ON's manufacturing footprint. The supplier, which built its business around silicon carbide power devices, image sensors, and power management ICs for automotive and industrial customers, operates its own fabs alongside a hybrid manufacturing model. Fixed-cost absorption is therefore a direct lever on margins: when utilization falls, depreciation and overhead spread across fewer shipped wafers, and gross margin compresses even if average selling prices hold. When utilization climbs, the arithmetic reverses.

The debate Trefis puts forward is essentially about timing and trajectory. On one side sits the bear case implicit in the headline — the stock trades at a multiple that already assumes a recovery, leaving little cushion if demand stays soft. On the other side sits the argument that a utilization-driven margin expansion, once it arrives, would justify the premium investors are currently paying. Between those two positions lies the actual question buyers of the stock must answer: how quickly and how completely do ON's fabs refill?

The automotive and industrial end markets that dominate ON's revenue have been the swing factor. These customers historically carry long design-in cycles and sticky sockets, which supports the durability of ON's content per vehicle and per factory — the strategic logic behind its silicon carbide investments in particular. But the same stickiness that protects long-term share does little to smooth short-term order patterns, and it is those order patterns that determine fab loading quarter to quarter.

For semiconductor investors more broadly, the Trefis piece is a reminder that utilization is one of the most underappreciated drivers of reported profitability in the analog and power segments. Unlike leading-edge logic, where pricing and node leadership dominate the narrative, mature-node power and analog suppliers live and die by fixed-cost leverage. A ten-point swing in fab loading can move gross margins by amounts that would take years of product mix improvements to replicate.

Trefis does not present the stock as cheap on today's numbers. The judgment is conditional: pay up only if you believe the factories get fuller. Investors weighing the name will need to form their own view on automotive demand, industrial inventory digestion, and the pace at which ON's silicon carbide and power-management capacity fills — and decide whether the market's current price already assumes the best of those outcomes.