REAL-TIME GLOBAL RESEARCH
WFE/Semi Forecast Update: SPE Setup into Earnings, Capital Intensity, AI Upside
Research evidence excerpt
WFE/Semi Forecast Update: SPE Setup into Earnings, Capital Intensity, AI Upside
USA | Semi. Cap. Equipment EquityJulyResearch14, 2026
WFE/Semi Forecast Update: SPE Setup into
Earnings, Capital Intensity, AI Upside
Raising our C26-C28 WFE forecast to $152B/$200B/$253B on stronger Leading
Edge/DRAM spend. The 20-25% drop in SPE stocks over the last 2 weeks has
created reasonable valuations and a favorable setup into earnings (implied C28
WFE to $250B from $300B). We look for optimism across SPE with DRAM and
LE visibility expanding. AMAT remains our favorite at 25x on $250B WFE, though
$300B is possible in C29 as the AI build continues, representing upside over the
NTM.
As we head into earnings, we are raising our WFE outlook again to reflect the continued strength
but also taking a look at just what level of WFE is priced in, what level of spending will be required
to support the AI forecasts, and what the bottlenecks will be. We are updating our WFE forecast to
$152B/$200B/$253B for C26-28 on stronger Leading Edge and DRAM spend. At the top (6/30/26),
we argued that >$300B was already being priced in for C28, but with the 20-25% pullback in the last
2 weeks, that number looks closer to $250B offering a more comfortable setup into earnings. We
also look at the spending gap vs Cloud capex and implied AI/HBM revenue and WFE vs Foundry
capex. This will take longer to play out, but it is important to keep an eye on just how large total AI
spend has gotten. We also explored the impact from AI/HBM on capital intensity with the answer
that 10% is the new normal vs 15% prior. Overall, we look for a very positive tone from all the SPE
names with DRAM and LE F/L very strong with expanding visibility. While we don’t think $300B is
The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.
Open report viewer