Wall Street Kills the TurboQuant Bear Case on SanDisk With One Word: Jevons

The Cantor Note

Cantor Fitzgerald reiterated Micron as Overweight with a $700 price target on April 1, calling it a top pick after meetings with management. Analyst C.J. Muse pointed to an "elongated cycle supported by leading edge wafers, growing HBM trade ratios, and lack of clean room space" and dismissed concerns about declining DRAM spot pricing — which represents less than 5% of the total market — as noise. The call lifted the entire memory complex: Micron gained roughly 9% and SanDisk moved 8.5% in sympathy, extending a recovery that has now fully erased the 18.5% TurboQuant drawdown from late March.

This is the second major analyst endorsement in a matter of days. BofA reiterated its Buy on SNDK with a $900 target, citing AI-driven NAND demand and an accelerating mix shift toward enterprise SSDs. Bernstein similarly dismissed TurboQuant's impact on NAND as negligible. The Wall Street consensus is hardening: the selloff was a panic, not a thesis change.

The Jevons Paradox Defense

The most interesting part of the Cantor note wasn't the price target — it was the intellectual framework. Cantor invoked the Jevons Paradox to argue that TurboQuant, far from destroying memory demand, would likely increase it. The logic: when a resource becomes more efficient to use, total consumption tends to rise because efficiency unlocks new use cases and scales existing ones.

Morgan Stanley's Shawn Kim made the same argument in the days following the selloff, citing historical precedents — JPEG compression didn't reduce storage demand, it enabled digital photography at scale. Video codecs didn't shrink the need for bandwidth, they enabled streaming. BofA's Vivek Arya added that Google itself raised its 2026 capex guidance to $180 billion despite publishing TurboQuant, suggesting the company expects efficiency gains to translate into capability expansion rather than procurement cuts.

TrendForce's technical analysis added the sharpest wrinkle: TurboQuant targets only inference-phase KV cache compression. It has zero direct impact on NAND flash storage, which is the actual business underpinning SanDisk. The selloff conflated DRAM — where the efficiency impact is at least debatable — with NAND, where it is essentially nil.

The Setup Into April 30

SanDisk announced on March 31 that its fiscal Q3 earnings call is set for April 30. The guidance is striking: $4.4–4.8 billion in revenue with non-GAAP EPS of $12–14 and gross margins of 65–67%. For context, Q2 already demolished expectations at $3.025 billion revenue (+61% YoY) and $6.20 EPS against a $4.85 consensus, with free cash flow of $843 million. The sequential acceleration implied by Q3 guidance reflects raw pricing power as NAND contract prices rose 38% in Q1 2026.

There is also a geopolitical dimension hardening the bull case. The Iran conflict has driven up energy costs in South Korea, where two-thirds of global memory chips are produced. Cantor specifically flagged this as favoring Micron over Korean competitors — and by extension SanDisk, which operates fabs in the US and Japan — at a time when Korean chipmakers face rising input costs and uncertain helium supply chains. The average Wall Street price target sits near $790 with BofA's $900 target implying roughly 30% upside from here. Whether that gap closes depends on whether NAND pricing power sustains through the April 30 print — and so far, every data point suggests it will.

Trading on Hyperliquid

Trade SNDK on Hyperliquid with up to 10x leverage.