Micron's latest earnings beat sent memory-adjacent names including Sandisk sharply higher in after-hours trade, with Fed PCE inflation data now the next macro catalyst on deck. The setup puts memory bulls against a macro overhang — strong fundamental momentum in MU, but SNDK's deep net losses complicate the sympathy trade.
Micron's latest earnings beat sent memory-adjacent names including Sandisk sharply higher in after-hours trade, with Fed PCE inflation data now the next macro catalyst on deck.
MU's blowout print and nearly 49% revenue growth validate the memory upcycle, but the question is whether the rally holds into a PCE print that could flip the macro tape — and whether SNDK's deep losses make it a tradeable sympathy or a trap.
A hotter-than-expected PCE print reverses the tech bid and wipes overnight gains; additionally, any forward guidance commentary from MU pointing to softening NAND pricing or hyperscaler capex moderation would undercut the thesis regardless of the backward-looking beat.
CoverageSource: Investor's Business Daily · Published here THU, JUN 25 · 8:14 AM ET · 2 outlets in this record · latest listed: Yahoo Finance at 8:14 AM ETHow this is decided →
Micron Technology reported results that materially beat expectations, driving a sharp after-hours rally across memory-related names including the recently spun-out Sandisk (SNDK). MU's fiscal-year revenue of $37.4B represents nearly 49% YoY growth, with solid gross margins of 39.8% and diluted EPS of $7.59 — numbers that validate the AI-driven DRAM and NAND demand thesis that has underpinned the memory upcycle narrative.
The earnings catalyst puts MU squarely in focus as the cleanest expression of memory market health. Sandisk, which carries a very different financial profile — 30.1% gross margins but a -22.3% net margin and -$11.32 diluted EPS — is riding the sympathy wave, but its standalone fundamentals are materially weaker and its NAND-heavy exposure introduces different demand dynamics than MU's HBM/DRAM mix.
The looming Fed PCE print is the key macro risk that could cut this rally short. If PCE comes in hotter than expected, rate-sensitive growth and tech names could give back the overnight gains quickly, regardless of MU's fundamental strength. Conversely, a benign inflation print could extend the tech bid and pull forward further multiple expansion for MU.
For MU, the bull case rests on the continuation of AI infrastructure capex driving HBM demand that remains capacity-constrained — a structural tailwind reflected in near-50% revenue growth. The bear case is that memory cycles are notoriously mean-reverting and current pricing may already be peaking, with any capex slowdown from hyperscalers a significant headwind. SNDK's persistent net losses make it a riskier sympathy play — the revenue growth is real but the bottom line remains deeply negative.
MU's FY revenue of $37.4B at +48.9% YoY with 39.8% gross margins and $7.59 EPS demonstrates genuine fundamental strength driven by HBM/AI demand — this is not a guidance-driven beat but a hard revenue inflection. The sympathy lift in SNDK is more speculative given -$11.32 EPS and negative net margins, making MU the cleaner long. The PCE print is the near-term gating event; a soft reading removes the macro headwind and allows the fundamental multiple to expand.
The read above, as written. kept as written · closes shown from JUN 25 on
1-2 weeks into PCE resolution. Follow to be told when one lands.
Price context does not establish that the story caused the move.
MU's near-50% revenue growth with expanding gross margins signals the AI/HBM demand wave is still in early innings, and if PCE prints soft the macro lid comes off a stock with genuine earnings power of $7.59 diluted EPS.
Memory cycles are historically mean-reverting and MU's rapid revenue ramp may already be pricing in peak DRAM/HBM demand, while SNDK's deeply negative net margin (-22.3%) and -$11.32 EPS expose the broader memory trade to a sharp reversal if hyperscaler capex guidance softens.
Kept as written · your side, if you take one, is graded privately against licensed closes after 10 trading days · nothing here is advice · How the Wire is made →