Here’s a notable statistic: A quarter of this year’s blockbuster earnings growth for semiconductors is attributable to margin expansion.
That gives you some idea about the extent to which a wholly unprecedented surge in component prices is accruing to semi bottom lines.
The profitability bonanza is most pronounced for memory companies where gross margins are around 80%, a nosebleed figure that Goldman’s Ben Snider noted is “more than double the historical average.”
But it’s not just memory firms. Industry-wide margins are extraordinarily elevated, and company analysts — who of course take their cues from management in their coverage universe — reckon the good times will continue in perpetuity. The figure below shows you what I mean.
With apologies to whomever they’re due for such a blunt assessment, analysts are accepting, uncritically, a narrative that says the current supply-demand imbalance represents a structural shift rather than an exaggerated version of the boom-bust cycles the space is famous for.
Maybe that narrative’s correct, but I can assure you things won’t be as smooth as consensus projects. This isn’t going to be a clean reset to a higher plateau, which is to say I doubt very seriously the idea that this year’s margins will be next year’s, and next year’s 2028’s. There will be some volatility, even if the supply shortage lasts and hyper-scaler capex holds up.
At the very least, the statistic mentioned here at the outset isn’t replicable. Obviously. As Snider put it, “the rate of incremental margin expansion [will] slow meaningfully next year.” That matters for overall S&P 500 earnings growth, which’ll be up against some very tough comps this time in 2027.
The figure on the left, below, gives you some context for the anomalous boon to semi earnings growth from fatter margins.
To reiterate: The 2026 experience isn’t repeatable, or you better hope it’s not. If it is, your next Mac’s gonna run you $15,000. The figure on the right gives you a sense of just how important this is for index EPS.
“S&P 500 earnings growth in the next few years will be highly sensitive to the trajectory of semiconductor margins,” Snider went on. “In a scenario where slowing AI infrastructure investment, increasing supply and/or technological shift lowers semiconductor prices and profit margins, S&P 500 EPS growth would disappoint.”
On Goldman’s estimates, every 100bps change in semiconductor gross margins in 2027 will shift aggregate S&P 500 EPS growth by about 1ppt. Note from the chart that if semi margins were to revert to the historical mean, earnings growth for the index would be just 4% next year all else equal, roughly 10ppt below both bottom-up and top-down consensus.
