Sandisk (SNDK) rebounded after a three-day selloff, gaining 5.6% through 1:45 p.m. ET. Morgan Stanley argues the memory chip supply deficit persists, with Q3 memory prices projected to rise 25% vs. Q2 and shortages not easing before 2028—supporting near-term profit visibility for a cyclical memory name.
The real edge here is not that memory is tight; it is that pricing is now becoming a second-order AI capex beneficiary, which makes SNDK more than a cyclical rebound story. If pricing is still accelerating while data-center demand is the only demand engine, the market should start capitalizing forward earnings with a higher multiple for longer than normal, especially for the cleanest exposure to the shortage.
The losers are likely downstream hardware buyers rather than the chip names themselves: server OEMs, PC/handset assemblers, and any platform with weak pricing power that cannot pass through a memory cost step-up. TSM is a useful tell rather than a direct beneficiary; if foundry commentary stays strong while memory tightness worsens, it confirms AI infrastructure spend is broadening, but it also raises the risk that component inflation starts eating into customer margins and eventually slows unit demand.
The key risk is timing. In the next 1-3 months, the trade is driven by analyst revisions, pricing prints, and capex commentary; over 6-18 months, the market will start looking for capacity response from Samsung/Micron/SK Hynix, which is usually the point where peak-cycle multiples compress before earnings actually roll over. The contrarian miss is assuming a 2028 shortage automatically means an easy multi-year long: if supply responds faster than expected or cloud demand normalizes for even one quarter, these names can de-rate sharply despite still-firm fundamentals.
AI-powered research, real-time alerts, and portfolio analytics for institutional investors.
Overall Sentiment
mildly positive
Sentiment Score
0.30
Ticker Sentiment
