Stock Price: ¥11,250
Market Cap: ¥158.3bn ($1bn USD)
MegaChips Corp (TYO 6875) is one of several companies in Japan with a large stake in a publicly listed company in the U.S. that dwarfs its own market cap. In this case, MegaChips owns ~10% of SITM. Off the top of my head there’s also Sanken Electric (TYO 6707) which owns 32% of ALGM.
But there’s a reason I’m writing about MegaChips instead of Sanken Electric: the discount is wider AND the remaining operating business is both simpler and better.
But before we get to that, here’s the Megachips thesis, a simple valuation framework where we only value NCAV + the listed stake in SiTime (NASDAQ: SITM) and subtract taxes owed if sold today:
MegaChips has 14 million shares outstanding, so that’s ¥16,920 per share of after tax SiTime value. NCAV sits at ~¥40.9bn or about ¥2,923 per share.
Our NCAV calculation adds back the SiTime-related deferred tax liability because we already deducted the estimated tax.
So for ¥11,250, you get ¥19,842 of after-tax SiTime value and current assets net of all liabilities. This gives zero credit for the remaining operating business and other financial assets. What are those other assets?
That’s another ¥2,604 per share of value, which is not insignificant given that MegaChips trades at just ¥ 11,250. To remain conservative we can haircut this by the same 31.47% tax rate which gets us to ¥1,784 per share.
The Macronix stake can be marked to market as it is listed in Taiwan, but Morse Micro is private and carried at book value. MegaChips has stated that Morse Micro raised funds in the past at a price above its carrying value of the stake. The company is considering an IPO on the Australian Stock Exchange, targeting a valuation of $600m AUD. It’s hard to know exactly how much of the company MegaChips owns or at what valuation it invested, so using book value here is fine.
Lastly, what is the MegaChips standalone operating business worth? The company forecasts the following for FY27:
Revenue: ¥42.0bn
Operating profit: ¥2.5bn
Ordinary profit: ¥2.0bn
This appears conservative as Q1 already generated ¥1.058bn of adjusted operating profit, but let’s assume they do ¥2.5bn. The reported Q1 operating profit of ¥448m includes a SiTime related tax charge.
MegaChips is a fabless semiconductor company that designs and sells custom application-specific integrated circuits (ASICs). Nintendo accounted for a whopping 77% of MegaChips’ revenue last year, up from 72.1% the prior year. MegaChips’ dependence on Nintendo is significant, but this is not new. Megachips has supplied Nintendo since the Nintendo 64 era in 1994, a relationship spanning more than 30 years and multiple game console generations.
The company is targeting ¥10bn in operating profit in FY2030, but for now let’s value the operating business conservatively at 5x FY27’s conservative operating profit target, which gives us ¥12.5bn, or ¥892 per share. If they hit their FY2030 targets, this would be worth ¥3,568 per share. A 5x multiple on a fabless semiconductor business seems low, but we’re trying to be conservative. Heck, let’s just round the entire operating business to 0. It doesn’t move the needle too much anyway.
The entire company has only 328 employees, which makes sense since MegaChips designs chips rather than manufacturing them.
MegaChips is buying back stock and recognizes this discount
Unlike many value traps and other sum of the parts stories, MegaChips is actively selling down its SiTime stake and repurchasing its own shares. Since May 2025, the company has done 3 buyback programs:
With ~14 million shares outstanding (market cap of ¥158 billion), these are significant buybacks. The company specifically mentions that this most recent repurchase is being funded with proceeds from its May 2026 SiTime sale when they sold 400,000 shares at $779.5 per share. Since May 2025, share count has fallen from 17.07 million to 14 million shares.
Perhaps most interestingly, management’s language surrounding these buybacks has evolved from boilerplate “improve capital efficiency and return profit to shareholders” back in May 2025 to something much more explicit on August 6. The exact phrasing was “the current share price is below the level MegaChips itself considers appropriate”
The company went from targeting ¥20bn in repurchases by FY2030 (announced in May 2026) to ¥20bn in FY27 alone while the messaging changed from “flexibly buying back shares based on market conditions” to “our shares are too cheap”.
MegaChips isn’t a pure capital return story. They’ll be making new investments with some of their SiTime proceeds
While MegaChips is buying back stock, the company also mentioned that they will allocate SiTime proceeds into growth investments, startups, strategic alliances, and M&A. Given that MegaChips had the foresight to acquire SiTime back in 2014 for $200 million (current market cap $21 billion), management has earned some credibility in making investments and doing M&A.
MegaChips plans to reduce its SiTime stake from roughly 10% today to 5% by FY2030.
The key variable: SiTime Performance
While MegaChips is undoubtedly cheap and doing all the right things, whether this investment works out or not is mainly dependent on SiTime’s stock performance. If things stay steady, MegaChips recycling SiTime shares into its own stock will continuously increase NAV per share, but if SiTime collapses, our discount could quickly vanish.
SiTime is a growth stock and it isn’t exactly cheap. The market is excited about it as they have very strong margins and exposure to AI. I’m not going to opine on SiTime as I don’t know much about it. While their technology is real and exciting, the $21 billion valuation looks rich to me, but who knows. One way to mitigate this risk would be to short SiTime, but since I don’t have a strong view, I remain unhedged.
I bought MegaChips after Sitime’s 30% rise earlier this month after Q2 results. MegaChips stock was barely up that day.
Why I like MegaChips more than Sanken Electric
Sanken Electric is another interesting sum of the parts story with its U.S. stake in ALGM being worth more than its market cap, but unlike MegaChips, the remaining business at Sanken Electric is capital intensive, has thousands of employees, and has recently been loss making. Their stake in ALGM is also much larger (32%+ of shares outstanding) vs MegaChips’ remaining 10% stake in SITM.
Conclusion:
This is a cheap stock and they seem to be doing everything right. Given that I don’t have much AI exposure in my portfolio, MegaChips is a discounted way to get some (via SiTime). I’m betting that MegaChips keeps selling SiTime stock and buying back more stock. In the meantime, we collect a 2.3% dividend.
This is a slightly above average basket sized bet.
Disclosure: I own shares in MegaChips Corp (6875). The security could be sold at any point in time without prior notice. This is a small position as part of a broader basket of cheap Japanese companies so I haven’t dug too deep into this name. If I missed anything important, feel free to share in the comments. None of this is investment advice. Everything in this post is my own opinion and I could be wrong. Do your own due diligence.
I used Claude to transform some boring text tables / bullet points into pretty graphics.
The bull thesis for SITM is interesting. Their product is superior to existing quartz for timing and their average selling prices have been soaring. According to their CEO SiTime had ~$200 worth of product in a fully populated rack, but today it's $500-$700. This number seems to be increasing and as AI compute prices soar. Growth + margins seemingly inflected last quarter, but this thing is NOT cheap today.
My only Japanese position at the moment