KIOXIA’s June 2026 investor day dropped some bold claims about NAND market sustainability, and the data is generating serious debate among Korea-focused analysts. With demand projected to outpace supply through end-2027 and operating margins holding at historically elevated levels, the pure-play NAND bellwether is essentially telling investors this isn’t your father’s memory cycle. We asked three of our analysts to weigh in — and predictably, they disagree.
Let me be clear about what KIOXIA actually said at their investor day, because the market seems to be hearing what it wants to hear. Yes, they projected demand exceeding supply through late 2027. Yes, they used the phrase “New Phase” to describe what they believe is a structurally elevated revenue and margin environment. But a conservative Japanese industrial company deploying optimistic forward guidance deserves more scrutiny, not less. When management at a cyclical company tells you the cycle has changed, that’s precisely the moment to reach for your historical data set.
The macro backdrop here is not benign. We are operating in an environment where the won remains under pressure, NDF markets have attracted speculative positioning, and Korean regulators literally convened an emergency joint market monitoring session in early June to address currency volatility. Foreign investors rotating out of Korean equity positions — even partially — creates a valuation headwind that fundamental NAND bulls tend to ignore. When global liquidity tightens or dollar strength reasserts itself, Korean memory-adjacent names get hit regardless of underlying chip fundamentals. KIOXIA is a Japanese entity, but its supply chain and competitive dynamics are deeply intertwined with SK Hynix and Samsung, meaning Korean investors cannot separate NAND market optimism from their own macro exposure.
The Greenspan parallel is worth invoking here. In 1996, irrational exuberance didn’t mean the tech sector had no future — it meant valuations had disconnected from the pace at which fundamentals could realistically materialize. NAND demand from AI workloads is real. But the timeline and magnitude of that demand, and more importantly the pricing discipline of the supply side, are being extrapolated with a confidence that historical NAND cycles simply do not support. The moment one major player blinks on capex discipline, the supply-demand math changes fast.
I’ll grant the bear one point: history is not kind to NAND bulls who overstay their welcome. But let’s look at what the actual financials are telling us, because there’s something genuinely different in this data set. KIOXIA’s second-quarter guidance came in at roughly four times the revenue base from prior cycle troughs. Four times. That isn’t a rounding error or a favorable comparison period — that’s a structural shift in the revenue floor. And when you overlay operating margins holding at levels that would have been considered peak-cycle in any prior period, you’re looking at a company that has repriced its earnings power meaningfully upward.
The key question for value-focused investors isn’t whether NAND is cyclical — it obviously is. The question is whether the trough of the next cycle will be higher than the peak of the previous one. The comparison between KIOXIA and SanDisk (WDC’s NAND spinout) is instructive here. Two companies sharing joint venture fab capacity, essentially operating with similar cost structures, yet trading at materially different valuation multiples for periods of time. That kind of pricing dislocation is where genuine value lives. The HDD versus NAND valuation divergence is similarly interesting — Seagate’s long-run revenue chart is essentially a flat line interrupted by the 2011 Thailand flood anomaly, whereas the NAND revenue trajectory has an unmistakably different slope since AI training demand inflected.
What I’m watching is capex discipline, specifically whether the JV fab structure between KIOXIA and WDC creates natural constraints on supply-side aggression. If both parties need consensus to meaningfully expand wafer starts, the supply response to elevated pricing is structurally slower than in prior cycles when each player acted unilaterally. That’s not a guarantee of sustained margins, but it’s a credible mechanism for why this cycle might behave differently. I want to see two more quarters of data before making a strong call, but the earnings power case is more compelling than the bears are acknowledging.
Here’s what I know from watching this market daily: KIOXIA’s investor day was the most bullish major NAND read-through we’ve gotten from a primary source all year, and the street largely agrees. Demand greater than supply through end-2027 is not a small claim — that’s eighteen-plus months of runway on a cycle that typically flips in twelve. And this is coming from a company whose corporate culture is notoriously conservative. When a buttoned-up Japanese industrial says “New Phase,” they’ve war-gamed that statement extensively. They’re not in the business of embarrassing themselves with investor relations hyperbole.
The practical NAND market read right now is straightforward: DRAM spot prices have been in a corrective phase after a massive run, but NAND spot has been lagging the recovery for much longer and now looks like it’s catching up. The AI training wave that dominated 2023 and early 2024 was primarily a DRAM story — HBM, high-capacity server DRAM, all the SK Hynix headlines. But AI inference at scale, edge deployment, and the explosion in SSD storage demand from data centers building out warm and hot storage layers — that’s the NAND demand wave that’s still in earlier innings. Seagate’s recent results confirmed that even the HDD side of the storage equation is seeing strong data center demand, and HDD and NAND are complementary, not purely competitive.
For Korean investors specifically, the actionable angle isn’t necessarily a direct KIOXIA trade — it’s the upstream equipment and materials ecosystem. Companies levered to NAND fab investment cycles, particularly those with structural exposure to advanced stacking and layer count increases, are seeing demand that doesn’t follow the same spot price volatility as the chip makers themselves. The laser process equipment story in Korea, for example, becomes more compelling the higher layer counts go, because physics demands more precision at each step. The KIOXIA investor day didn’t just tell us NAND is healthy — it told us the entire supply chain has runway.
KIOXIA’s investor day presented the most explicit industry-source confirmation yet that the NAND supply-demand balance remains constructive well into 2027, and the debate among our analysts reflects the genuine tension in how to interpret that signal. The Macro Bear is right that macro headwinds — won volatility, global rate dynamics, and speculative currency flows — create noise that can overwhelm even solid fundamental stories for Korean market participants. The Value Hunter correctly identifies that the JV fab structure and the step-change in KIOXIA’s revenue floor represent credible reasons why this cycle’s trough may be structurally higher than prior cycles. And the Street Pragmatist is simply right that the trade has been on for a while and the NAND ecosystem read-through — from pure-play chip makers to upstream Korean equipment suppliers — remains one of the cleaner risk-reward setups in the semiconductor space right now. As always, the truth likely lives somewhere between the bear’s skepticism and the bulls’ enthusiasm, but the KIOXIA data has materially shifted the burden of proof toward those betting against NAND market sustainability.