[Stock Analysis] Kioxia (TSE: 285A): Net Profit Up 46x, So Why Is the Stock Swinging Wildly?

Hey everyone, Hirokichi here.

On July 31, 2026, Kioxia Holdings (TSE: 285A) released earnings for the first quarter of its fiscal year ending March 2027. Net profit came in at 46.1 times last year’s level, an eye-popping number. At the same time, the stock has been extremely volatile over the past month and a half. Since this has been all over the news and social media, let’s break down what’s actually in the earnings report and look at it from an individual investor’s point of view.

What kind of company is Kioxia?

Kioxia Holdings makes NAND flash memory, the type of memory chip used in SSDs for smartphones, PCs, and data centers. The company was originally Toshiba’s memory chip business, spun off and later listed on the Tokyo Stock Exchange Prime Market in December 2023.

Looking at major shareholders, Toshiba holds 30.5%, while entities tied to the US private equity firm Bain Capital (the BCPE Pangea Cayman group) hold more than 50% combined. In other words, this is effectively a company co-managed by Toshiba and Bain Capital. Kioxia had 15,042 consolidated employees as of the end of March 2025. Almost all of its revenue comes from NAND memory, and demand from generative-AI data centers has become the single biggest swing factor for its results.

Share price and key valuation metrics

First, let’s look at the stock’s recent moves. Kioxia shares trended steadily upward through the first half of 2026 and hit an intraday all-time high of JPY 112,700 on June 22, 2026 (the closing price that day was JPY 108,700).

Then things flipped. On July 17, 2026, South Korea’s SK Hynix announced a massive capital investment of KRW 80 trillion (roughly JPY 8.7 trillion) to build new NAND plants. That triggered fears that NAND supply would loosen up, and Kioxia’s stock dropped 16.10% in a single day to JPY 52,110. The stock kept swinging after that, falling to around JPY 39,500 by July 30 — a decline of more than 60% from the June 22 high in about a month.

Looking at this chart, the stock went from a June surge to a July crash, losing more than half its value in a little over a month. Then, on earnings day, July 31, 2026, the stock jumped 17.72% to close at JPY 46,500. On top of the earnings numbers themselves, the buyback and stock split announced the same day (details below) appear to have driven the rebound.

On valuation, as of May 22, 2026, Kioxia traded at a P/E ratio (how expensive a stock is relative to its earnings) of 56.49x and a P/B ratio (price relative to book value per share) of 22.39x. By the June 22 peak, those had climbed to a P/E of 106.97x and a P/B of 42.4x — a sign the stock had run well ahead of actual earnings growth. The dividend yield, for reference, is 0%: Kioxia currently pays no dividend, so this is purely a capital-gains story, not an income play.

Earnings check

Next, let’s look at the numbers. Here’s Kioxia’s revenue and operating profit over the past five fiscal years, based on its financial statements and securities reports.

In FY March 2022, Kioxia posted revenue of JPY 1,526.5 billion and operating profit of JPY 216.2 billion. But FY March 2023 and FY March 2024 both fell into operating losses (JPY -99.0 billion and JPY -252.7 billion respectively) as memory prices collapsed. The company swung back to profit in FY March 2025 (revenue JPY 1,706.5 billion, operating profit JPY 451.7 billion), and in FY March 2026 posted record results: revenue of JPY 2,337.6 billion (+36.99% YoY), operating profit of JPY 870.4 billion (+92.67% YoY), and net profit attributable to owners of the parent of JPY 554.49 billion (+103.62% YoY). The chart makes the V-shaped recovery from the loss years very clear.

Then, on July 31, 2026, Kioxia reported Q1 FY March 2027 (April-June 2026) results: revenue of JPY 1,767.1 billion (+415.5% YoY) and quarterly net profit attributable to owners of the parent of JPY 842.165 billion (46.1x YoY). The main driver was a sharp rise in NAND selling prices on the back of surging demand from generative-AI data centers. For Q2 FY March 2027 (July-September), the company guided for net profit of JPY 1,270.0 billion, 31 times the year-earlier level. On the other hand, Kioxia did not disclose full-year guidance for FY March 2027, which is worth keeping an eye on.

Where the company is headed

On the same day as the earnings release, July 31, 2026, Kioxia made two major shareholder-return announcements.

First, a share buyback: up to JPY 800.0 billion, covering as many as 30 million shares (5.5% of shares outstanding), to be executed between August 3 and October 30, 2026.

Second, a stock split: a 1-for-3 split, with a record date of September 30, 2026 and an effective date of October 1, 2026. Splitting the stock lowers the minimum investment unit, which should make shares more accessible to individual investors who previously found them too expensive.

Management describes these moves as aimed at improving capital efficiency, strengthening shareholder returns, and broadening the investor base. It suggests the company is directing its record profits toward both capital investment (R&D spending of JPY 80.497 billion and capital expenditure of JPY 225.6 billion) and shareholder returns at the same time.

Bull and bear case: what I make of it

Here’s my personal read on this earnings report.

On the bullish side:

(1) NAND prices are on an upward trend driven by generative-AI and data center demand, and that’s flowing straight through to earnings. With Q2 FY March 2027 guidance calling for a 31x jump in net profit, the growth trend looks set to continue, at least for now.

(2) The company paired a JPY 800 billion buyback with a 1-for-3 stock split, pushing both shareholder returns and a broader investor base at the same time. Announcing these at the peak of record profits can be read as a sign of management’s confidence.

(3) Kioxia has already proven it can stage a sharp V-shaped recovery, having gone from steep losses in FY2023-2024 to record profits. Some observers also argue the industry has consolidated into fewer major players, which could make it easier to defend pricing than in past cycles.

On the bearish side:

(1) The stock already looks richly valued. At the June 22 peak, the P/E ratio was 106.97x and the P/B ratio was 42.4x — extremely high for what is, at its core, a manufacturing business. If the assumption of continued strong earnings breaks down, the stock could see a sharp correction.

(2) Memory chips are notorious for the “silicon cycle” — sharp price swings driven by the balance of supply and demand. We just saw this play out in real time: a single announcement from SK Hynix about a new investment plan wiped out more than 60% of Kioxia’s share price in under a month. It’s worth staying level-headed about whether this AI-driven demand cycle is really “different this time.”

(3) Kioxia withheld full-year guidance, and the dividend yield is 0%. That could hint at some uncertainty management isn’t ready to commit to publicly (pricing trends, major customer contracts, and so on), and it’s also a reminder that this stock isn’t suited to investors looking for dividend income.

My personal take is that Kioxia genuinely offers something rare among Japanese stocks: direct exposure to the generative-AI boom. That said, the stock lost more than 60% of its value from its all-time high in just about six weeks, so the volatility here is real. Before jumping in on a strong earnings headline, I think it’s worth pausing to ask yourself honestly whether you can stomach that kind of price swing.

Wrapping up

Kioxia’s Q1 FY March 2027 results delivered a stunning 46.1x jump in net profit, but the stock swung wildly both before and after the announcement. There’s a strong tailwind from generative-AI demand, but the silicon-cycle risk inherent to memory chips, plus an already-rich valuation, both call for caution. I’d approach this one keeping both the bull and bear case in mind, and only within a position size you’re comfortable with. Take it easy and keep building steadily. See you in the next post!

Previous Stock Analysis: [Stock Analysis] Amazon (AMZN): Is It Still a Buy at $235? Record Earnings and a $220B AI Bet

Related post: [Weekly Stock Picks] Murata Manufacturing (6981) Rebounds to Limit-Up After a 55% Plunge & Microsoft (MSFT) Soars 14% on Blowout Azure Earnings

日本語版はこちら → 【銘柄分析】キオクシアHD(285A)決算で純利益46倍!株価はなぜ乱高下?個人投資家が思うこと

* This article is for informational purposes only and does not recommend buying or selling any specific stock. Please make investment decisions at your own responsibility.

Thanks for reading! If you enjoyed this post, a quick click on the banners below would really encourage me.

ブログランキング・にほんブログ村へ

人気ブログランキング







貯蓄の無料相談サイト「ガーデン」

コメント

タイトルとURLをコピーしました