Why did Kioxia's stock price crash? The fall from top market capitalization and the shock of 920 billion yen in margin buying
Kioxia Holdings, which captured market attention as a prime candidate for AI semiconductors and at one point surged to the top tier of market capitalization rankings.
However, the stock price plummeted after its peak in June 2026. By October, the decline from its high had briefly exceeded 66%, and margin buying had swelled to an unprecedented scale of approximately 920 billion yen.
On the other hand, looking at performance, the environment surrounding Kioxia has not completely deteriorated. Against the backdrop of rising NAND flash memory prices, there are even bullish market forecasts for net profit of approximately 5.6 trillion yen for the fiscal year ending March 2027.
Even so, why was the stock price sold off to this extent?
In this article, we will organize the background of the stock price decline, the issue of margin buying, the stock price level considered from fundamentals, future dividend policy, and three possible future scenarios.
*Data on stock prices and margin balances in this article are based on information confirmed as of October 23, 2026, unless otherwise noted. This does not guarantee future stock prices or performance.
1. The fall from top market capitalization
Looking back at the price movement of Kioxia stock, the excitement in the first half of 2026 and the intensity of the subsequent decline are clear.
In late June, the stock price hit a high, and against the backdrop of expectations for AI-related stocks, it showed a presence as one of the top companies in Japan in terms of market capitalization.
However, the market changed completely from there.
In addition to the correction in AI-related stocks as a whole, the outlook for the semiconductor memory market and investor supply and demand were taken into account, and the stock price turned to a downward trend.
As of October 19, it was reported that the decline from the high in late June had briefly exceeded 66%.
This is not just a correction of a few percent.
If a stock price falls by 66% from its high, the valuation of shares held at the high for 10 million yen would be approximately 3.4 million yen.
For the stock price to return to its original level, an increase of approximately 194% from there would be required.
In other words, for stocks that have fallen significantly, one must consider not only the rate of decline but also the rate of increase required to return to the original price.
Of course, the past high was not necessarily the fair price.
Rather, what should be considered from Kioxia’s current price movement is the question of to what extent the market had priced in future growth at the time the stock price was rising.
Stock prices are not determined solely by current performance.
If expectations for future growth expand, the stock price can rise beyond actual profits.
Conversely, if those expectations recede even slightly, the stock price will fall significantly, even if performance is strong.
Kioxia’s stock this time can be said to be a symbolic case for considering this very nature.
2. Why was it sold off so heavily?
There is not just one reason for the stock price decline.
Broadly speaking, the following five factors can be considered.
1. The sense of overheating in AI-related stocks has receded
The rapid spread of generative AI is the background to the attention Kioxia received.
AI learning and inference require handling massive amounts of data, and in data centers, not only high-performance semiconductors but also large-capacity storage are important.
NAND flash memory, which is Kioxia’s mainstay, is a storage semiconductor used in SSDs and the like.
The expansion of AI data centers is a major growth opportunity for the company.
However, what we want to be careful about here is the point that the growth of the AI market and the rise of Kioxia’s stock price are not the same thing.
There is the issue of how much future growth is priced into the stock price.
Even if AI demand is growing, if the market was expecting even more growth, the stock price will fall due to a correction of expectations.
Also, if US semiconductor-related stocks are sold, selling is likely to spread to Japanese semiconductor-related stocks as well.
The flow of funds in the market as a whole, which cannot be explained by Kioxia’s specific performance alone, is also having an impact.
2. Even if the NAND market is strong, it is not necessarily strong forever
What is particularly important when analyzing Kioxia is the cyclical nature of the memory market.
The price of NAND flash memory fluctuates significantly depending on the balance between supply and demand.
If demand exceeds supply, prices rise and manufacturers’ profit margins improve.
On the other hand, when manufacturers increase production capacity and supply exceeds demand, prices can fall and profits can drop sharply.
In short, the business performance of memory manufacturers is different from the type of business that grows steadily every year.
Profits can surge during boom periods and plummet during recessions.
This is why the method of valuation differs from that of companies that consistently earn high profits.
For example, even if a company earns a net profit of 5 trillion yen in a certain fiscal year, it does not necessarily mean that this will continue every year.
The appropriate stock price changes significantly depending on whether the market evaluates it as “the 5 trillion yen profit will continue in the future” or “this is close to the peak of profits.”
The fact that current profits are large and the fact that future profits are stable are separate issues.
(3) Movements that anticipate and price in the peak of business performance
The stock market moves by anticipating future performance, not just past financial results.
Therefore, it is not uncommon for stock prices to have already fallen by the time a company’s performance reaches its highest level.
Even if the net profit for the fiscal year ending March 2027 reaches approximately 5.6 trillion yen as expected by the market, that alone does not guarantee a rise in the stock price.
If the market expects even higher profits, the financial results as expected may be judged as insufficient.
Conversely, even if profits grow significantly, if a subsequent decline in memory prices is expected, investors will anticipate the future decrease in profits and evaluate the stock price accordingly.
What investors are looking at is not just the current period’s profit.
It is the point of whether the same profit can be earned in the next period and the period after that.
(4) Selling by investors who bought at high prices triggers further selling
For stocks that have plummeted, the balance of margin buying is also a significant issue.
Investors who bought on margin at high prices will hold unrealized losses when the stock price falls.
Some may try to reduce their losses by selling if the stock price recovers later. On the other hand, if it falls further, they may be forced to sell to provide additional margin or to realize losses.
When such selling increases, a vicious cycle can occur where the recovery of the stock price is suppressed, and more investors end up holding unrealized losses.
What is particularly viewed as a problem for Kioxia is the size of this margin buying balance.
I would like to examine this in detail in the next chapter.
5. Short-term capital that supported the stock price increase has departed
In phases where stock prices are rising rapidly, capital from short-term investors aiming for price gains tends to gather.
If price movements are large and trading is active, it attracts attention as a target for short-term trading.
However, after the stock price falls and price movements settle down, short-term capital may move to other stocks.
In fact, an article reported by Bloomberg on October 19, 2026, states that the trading value of Kioxia stock decreased from a peak of approximately 4.7 trillion yen to approximately 770 billion yen on October 16.
When trading value decreases, there is a possibility that enough buyers will not appear to absorb the selling necessary to unwind margin buying.
In other words, the decline in the stock price can be explained not only by disappointment in performance, but also by changes in investor interest and capital flows.
3. The abnormal situation of 920 billion yen in margin buying
When considering Kioxia stock this time, what I personally want to focus on most is the margin buying balance.
As of the application on October 15, 2026, the margin buying balance was approximately 920 billion yen.
According to Bloomberg reports, it was the largest among stocks listed on the Tokyo Stock Exchange.
The margin buying balance refers to the balance of long positions that have not yet been settled among stocks purchased using margin trading.
Margin buying itself is not bad.
However, if margin buying is extremely inflated, it can become future selling pressure when the stock price falls.
Why is a high margin buying balance a problem?
For example, let’s consider a stock whose price has fallen significantly from its high.
Investors who bought on margin at high prices may continue to hold, hoping that the stock price will return.
However, if the stock price does not easily return and the settlement deadline for margin trading approaches, they may be forced to consider selling.
If many such investors exist, selling is likely to occur as soon as the stock price rebounds slightly.
As a result, there is a possibility that the stock price will continue to struggle to rise.
Furthermore, if the stock price falls again, it is conceivable that selling to avoid further losses will increase.
This is the supply-demand risk that is cautioned against in stocks with high margin buying balances.
Margin ratio is also at a high level
In addition to the margin buying balance, I would also like to pay attention to the margin ratio.
The margin ratio is a figure calculated by dividing the margin buying balance by the margin selling balance.
As of the application date of October 15, 2026, the margin ratio is reported to be approximately 38 times.
This indicates that the margin buying balance is extremely large relative to the margin selling balance.
However, just because the margin ratio is high does not necessarily mean that the stock price will fall.
If the stock price rises, there are cases where a high margin buying balance does not immediately become a problem.
The important point is whether the unwinding of long positions will proceed, or whether new buying will enter to absorb the selling.
Margin buying balance can be a factor that suppresses the upside of the stock price
The figure of approximately 920 billion yen in margin buying balance is by no means small.
Of course, this entire amount will not be sold off all at once. The margin buying balance is not a figure that directly indicates future selling amounts.
Even so, the fact that a large amount of margin buying remains during a phase where the stock price has fallen can become a burden on supply and demand.
What I want to note here is that even if business performance improves, the supply-demand problem will not necessarily be resolved immediately.
Even if business performance is strong, if there is a lot of selling on rallies, the stock price will be difficult to rise.
Conversely, if the margin buying balance decreases and selling pressure weakens, there is a possibility that the stock price will rebound even if the evaluation of business performance does not change.
Going forward, I would like to pay attention not only to financial results but also to the trends in the margin buying balance.
4. From a fundamentals perspective, what is the fair stock price?
This is where we get to the main point.
Now that the stock price has fallen significantly, is Kioxia undervalued?
In conclusion, if the company can achieve the profit for the fiscal year ending March 2027 that the market expects, the current stock price is undervalued. On the other hand, if the profit is merely a temporary peak, it cannot be judged as undervalued simply because the P/E ratio is low.
We need to consider these two points separately.
First, let’s check the actual financial results.
Kioxia’s consolidated financial results for the fiscal year ending March 2026 were as follows.
Indicator: Fiscal Year Ending March 2026 Results; Revenue: Approx. 2.34 trillion yen; Operating Profit: Approx. 870.4 billion yen; Net Profit: Approx. 554.5 billion yen; Year-on-Year Net Profit Growth Rate: Approx. 103.6%
Revenue increased by 37.0% year-on-year, and net profit increased by 103.6% year-on-year.
The significant growth in profit can also be confirmed from the financial statements.
However, the net profit for the fiscal year ending March 2026 was approximately 554.5 billion yen, and based on the number of shares after the stock split, the earnings per share is approximately 337 yen.
When simply compared to the closing price of 12,755 yen on October 23, 2026, the P/E ratio based on past performance is approximately 38 times.
In other words, looking only at past performance, it cannot necessarily be called undervalued.
On the other hand, regarding the net profit for the fiscal year ending March 2027, a figure of approximately 5.6 trillion yen has been reported as a market forecast.
If this profit is realized, the earnings per share, calculated with approximately 1.646 billion shares outstanding after the stock split, would be approximately 3,400 yen.
The expected P/E ratio against the current stock price of 12,755 yen is approximately 3.7 times.
Looking at this figure alone, it certainly looks undervalued.
However, the important point here is that the profit of approximately 5.6 trillion yen is not a confirmed earnings forecast by the company, but a figure reported as a market forecast.
Also, the second-quarter financial results for the fiscal year ending March 2027 are scheduled to be announced on October 30, 2026, and at the time of writing this article, verification based on the latest quarterly results is necessary.
If the P/E ratio is 3 to 4 times, is it unconditionally undervalued?
I do not think so.
For companies with high cyclicality like memory manufacturers, the P/E ratio can become extremely low when profits reach their peak.
For example, if a company with an annual profit of 5 trillion yen has a market capitalization of 20 trillion yen, the P/E ratio is 4x.
However, if that profit decreases to 2 trillion yen the following year, the effective P/E ratio becomes 10x.
If profits further decrease to 1 trillion yen, the P/E ratio becomes 20x.
Even if the stock price remains unchanged, the perception of value changes significantly based solely on changes in profit.
Therefore, when considering the fair value of Kioxia’s stock, it is necessary to assume not only current earnings but also profit levels several years into the future.
So, where does the fair stock price lie?
Here, we will calculate based on the stock price of 12,755 yen on October 23, 2026, assuming approximately 1.646 billion shares outstanding after the stock split.
The following is a scenario analysis combining future profit levels and valuation multiples, and is not a company forecast or a brokerage’s target price.
Scenario A: Profits grow to 5.6 trillion yen
Assuming a net profit of 5.6 trillion yen and a P/E ratio of 5-8x.
-
Earnings per share: approximately 3,400 yen
-
P/E ratio 5x: approximately 17,000 yen
-
P/E ratio 8x: approximately 27,200 yen
In this case, the current price of 12,755 yen may be evaluated as undervalued relative to the profit level potential.
However, this assumes that a net profit of 5.6 trillion yen is achieved and that this profit level continues for a certain period.
Scenario B: Profits remain at around 3 trillion yen
Assuming a net profit of 3 trillion yen and a P/E ratio of 6-8x.
-
Earnings per share: approximately 1,820 yen
-
P/E ratio 6x: approximately 10,900 yen
-
PER 8x: approx. 14,600 yen
In this case, the estimated fair stock price is approximately 10,900 to 14,600 yen.
The current stock price of 12,755 yen falls within this range.
Even if profits fall below market expectations, the current stock price is not necessarily significantly overvalued unless there is an extreme deterioration in performance.
Scenario C: Profits decrease to around 1.5 trillion yen
Assuming a net profit of 1.5 trillion yen and a PER of 8-12x.
-
Earnings per share: approx. 910 yen
-
PER 8x: approx. 7,300 yen
-
PER 12x: approx. 10,900 yen
In this case, the current stock price may be overvalued relative to the profit level.
If memory prices fall and profit margins deteriorate, the current low expected PER will lose much of its meaning.
Comparing the three estimates
Scenario Assumed Net Profit Assumed PER Estimated Stock Price Optimistic 5.6 trillion yen 5-8x approx. 17,000-27,200 yen Neutral 3 trillion yen 6-8x approx. 10,900-14,600 yen Pessimistic 1.5 trillion yen 8-12x approx. 7,300-10,900 yen
*All figures are estimates using the number of shares after the stock split. This does not predict or guarantee future performance or stock prices. The valuation multiples are assumptions set for explanation and are not definitive fair PERs.
What I want to emphasize is which scenario the current stock price is factoring in.
The level of 12,755 yen per share looks undervalued in a case where profits grow significantly, but it is not necessarily undervalued in a case where profits return to normal levels.
To evaluate Kioxia’s stock price, it is necessary to comprehensively check not only the low PER, but also the sustainability of the NAND market, capital investment, profit margins, and cash flow.
5. Will Kioxia pay dividends in the future?
For investors after the stock price drop, the presence or absence of dividends is likely a concern.
This is an important point, especially for those who want to expect not only capital gains from stock price increases but also dividend income from long-term holding.
First, let’s organize the verifiable facts.
Kioxia’s annual dividend for common stock for the fiscal year ending March 2026 was 0 yen.
Furthermore, regarding the fiscal year ending March 2027, the dividend forecast for common stock is listed as undecided in the financial results summary released on May 15, 2026.
In other words, at this moment, it is not a situation where one can definitively say that dividends will definitely be paid this fiscal year.
If performance improves, will dividends increase?
That is not necessarily the case.
This is because even if a company earns profits, it does not necessarily return all of them to shareholders.
In Kioxia’s case, there are options to allocate funds toward future capital investment, research and development, and strengthening its financial position.
In fact, in August 2026, the company announced a plan to invest approximately 5 trillion yen in the Japanese memory sector together with SanDisk.
For companies pursuing such large-scale investments, an increase in profit does not necessarily mean an immediate increase in dividends.
Of course, if investments proceed smoothly and cash flow stabilizes, there is a possibility that room for future shareholder returns will expand.
However, it is necessary to check the company’s official announcements regarding the timing and amount of dividend payments.
Is it rational to buy in anticipation of dividends?
At this point, one should be cautious about using dividends as the primary basis for investment decisions.
As long as the dividend forecast is undecided, it is difficult to build an investment strategy based on dividend yield.
If you are considering an investment in Kioxia, I believe it is appropriate to first evaluate it based on the sustainability of its performance and its stock price level, and then check for future official announcements regarding dividends.
6. Three scenarios expected in the future
Based on the content so far, I would like to consider three scenarios for Kioxia stock going forward.
1. Optimistic scenario: Performance growth is re-evaluated in the stock price
The most bullish scenario is one where demand for AI data centers continues to expand, and profits grow significantly due to rising NAND prices.
In this case, investors may re-evaluate Kioxia’s profit growth.
In addition, if margin buying decreases and supply and demand improve, the factors that have been hindering the stock price’s recovery may also weaken.
If the market begins to value sustained profit growth rather than temporary earnings improvements, a significant rebound in the stock price is possible.
However, for the stock price to reach new highs, it may be necessary to not only report good earnings but also provide results and outlooks that exceed market expectations.
② Objective scenario: Even if earnings are strong, stock price recovery will take time
This is the scenario I want to focus on the most.
The NAND market does not collapse significantly, and Kioxia’s earnings maintain a certain level.
On the other hand, it will take time to clear the margin buying, and the stock price will face heavy upward pressure.
Even if good earnings are announced, there may be times when gains are suppressed by selling pressure from those looking to exit their positions.
Also, as long as investors are wary of peak earnings, a low expected P/E ratio alone does not necessarily lead to a significant rise in the stock price.
In this case, the stock price is expected to fluctuate within a certain range while confirming changes in earnings, memory prices, and margin balances.
Even if a company’s performance is not bad, the stock price does not necessarily recover immediately.
With Kioxia’s stock this time, we should also consider that possibility.
③ Pessimistic scenario: Deterioration of the memory market and the liquidation of margin buying overlap
What I want to be most cautious about is the case where the NAND market deteriorates and expectations for future earnings recede.
For example, if memory supply exceeds demand and prices fall, profit margins could decline.
Furthermore, if margin buying remains at a high level, selling pressure could intensify due to a decline in the stock price.
If concerns about deteriorating earnings and worsening supply and demand proceed simultaneously, the possibility of a further decline in the stock price cannot be denied.
What I want to be especially careful about is the point that just because the stock price has already fallen significantly does not mean it will not fall further.
It is dangerous to judge that it is ‘probably the bottom’ based solely on the rate of decline from the high.
If the reasons for the stock price decline are not resolved, it may fall further from levels that appear undervalued.
7. What should be checked in the future?
If I were to continuously analyze Kioxia stock, I would want to check the following five items.
1. Quarterly earnings and full-year outlook
Focus not only on sales, but also on operating profit margins, net income, and the company’s earnings outlook. The divergence from market expectations is also important.
2. NAND prices and supply-demand environment
Check whether NAND prices continue to rise and whether the expansion of supply capacity is not outpacing demand.
3. Trends in margin buying balance
Check whether the margin buying balance is decreasing and whether the margin ratio is improving. A decrease in the margin buying balance is a clue to supply-demand improvement, but it does not guarantee a stock price rebound on its own.
4. Operating cash flow and capital investment
Check not only accounting profits, but also how much cash is actually being generated and how much capital is being spent on equipment investment.
5. Dividend policy and financial strategy
In addition to the official announcement of dividend forecasts, check future investment plans and changes in financial health.
What is particularly important is not to look at these in isolation, but to judge them in combination.
For example, if earnings are strong but the margin buying balance does not decrease and the stock price does not rise, there may still be supply-demand issues.
Conversely, if earnings forecasts are maintained, the margin buying balance decreases, and the stock price remains firm, it is possible that the market trend is beginning to change.
Of course, these are merely analytical focal points and do not guarantee any specific price movement.
8. Summary: Is Kioxia undervalued, or will it fall further?
The sharp drop in Kioxia stock did not happen simply because earnings deteriorated.
Multiple factors are overlapping, including the correction of expectations for AI-related stocks, the cyclical nature of the memory market, caution regarding future profits, and the accumulation of margin buying.
On the other hand, against the backdrop of rising NAND prices, there are also very bullish market forecasts for profits in the fiscal year ending March 2027.
Therefore, it is considered that the current stock price is at a level where the valuation will change significantly depending on whether future profits can be realized.
If profits expand as the market expects, the current stock price might be undervalued.
However, if the profit is merely a temporary peak, it cannot be judged as undervalued based solely on a low expected P/E ratio.
And, we must not forget the issue of supply and demand.
As long as the margin buying balance has swollen to a level of approximately 920 billion yen, there is a possibility that it will take time for the stock price to recover, even if business performance is strong.
Moving forward, I would like to carefully monitor the earnings announcement scheduled for October 30, changes in the NAND market conditions, and the trends in the margin buying balance.
Personally, I think it is important not to simply judge Kioxia stock as a ‘buy because it crashed,’ but rather to envision future profit levels under multiple scenarios and then consider to what extent the current stock price has priced in those risks.
Finally.
There are no absolutes in stock investment.
No matter how good a company’s performance is, its stock price can fall. Conversely, even if performance deteriorates, the stock price may rise if the bad news has already been priced in.
For stocks like Kioxia in this instance, where expectations for performance are high and stock price volatility is intense, particularly cautious judgment is required.
I want to think in advance about which scenario I am assuming when investing, and how I will respond if things turn out differently than expected.
Final investment decisions must be made based on your own financial situation and risk tolerance. Stock investment does not guarantee principal, and final investment decisions and their results are your own responsibility.