I checked my phone this morning the way I do most mornings — half out of habit, half out of curiosity — and the numbers stopped me for a second. Korea’s KOSPI index closed around 6,516, down roughly 304 points, or about 4.46%, in a single session. That’s not a typo. That’s one trading day.
For a minute, my stomach did the thing it always used to do back when I owned individual stocks — that quick mental math of “wait, how much of my money is sitting in that one company right now?” Except this time, the answer was easy: none. Not directly, anyway. And that gap between my old reaction and my current reality is exactly what I want to write about today.
What actually happened
The sell-off wasn’t really a Korea-only story, even if Korea felt it hardest. The proximate trigger, based on what I’ve read, is intensifying competition in AI chips — reports of new, cheaper AI models out of China putting pressure on the assumed demand growth for the kind of high-end memory and logic chips that companies like Samsung Electronics and SK Hynix make. Both stocks opened down more than 5% before paring some of the losses by the close. When investors get nervous about future demand for a specific product category, the stocks most tied to that category get hit hardest, and hit fastest.
It wasn’t isolated to Korean tech either. The S&P 500 itself had already closed down about 1% a few sessions earlier, with U.S. futures little changed heading into the new week. There’s also a separate layer of nerves running through markets right now around the Iran-U.S. military tension that’s been pushing oil prices higher. None of these threads are unrelated — a jumpy market tends to sell first and ask questions later, and semiconductor stocks in particular have a habit of amplifying whatever mood the broader market is already in.
Why this doesn’t touch my actual strategy
Here’s the thing I keep coming back to, and it’s not new — I’ve written about versions of this before. I don’t own Samsung. I don’t own SK Hynix. I don’t own Nvidia or TSMC or any individual chipmaker, Korean or otherwise. What I own, through my monthly automatic contributions into my pension account, is a mix of S&P 500-tracking funds and a Nasdaq 100 fund, bought the same way every single payday regardless of what happened in the news that week.
That distinction matters more on a day like today than it does most days. An index doesn’t rise or fall based on one sector’s bad week the way a single stock does. The S&P 500 holds roughly 500 companies across technology, healthcare, financials, energy, industrials, and more — so even a rough patch for semiconductors gets diluted by everything else sitting alongside it. I wrote a while back about what the S&P 500 actually is and how ETFs spread risk across many companies at once, and today is a pretty clean, live example of exactly that mechanism doing its job. My Nasdaq 100 fund will feel more of today’s tech weakness than my S&P 500 fund does — that’s the tradeoff of holding both — but neither one lives or dies on Samsung’s earnings call.
I’ll be honest: I have no idea whether AI chip demand slows down from here or roars back next quarter. I have no idea whether today’s drop is the start of something bigger or a one-day overreaction that gets mostly erased by the end of the week. I’m not going to pretend I do, and I’m suspicious of anyone who says they do with real confidence. That’s precisely why I never tried to build a strategy around correctly picking which chipmaker wins the next AI cycle. I’d have needed to be right about the technology, right about the competitive landscape, and right about the timing, all at once, over and over again. That’s a much harder job than just owning the index and letting it average out.
What I’m actually doing about it
Nothing. That’s the honest, slightly anticlimactic answer. My next contribution goes in on schedule, into the same S&P 500-heavy mix it always does, whether today’s headlines say “tech stocks crumble” or “markets hit new highs.” I’m not adding money early to “buy the dip” in chip stocks, and I’m not pulling back my regular contribution because the KOSPI had an ugly day. The whole point of dollar-cost averaging, which I’ve leaned on since day one of this journey from around $96,000 in savings toward a long-term goal of $100,000 a year in passive income, is that I don’t have to have a strong opinion about today’s headline to keep showing up.
Days like this are uncomfortable to watch, even from the sidelines of not owning the individual names. But they’re also a decent reminder of why I built my portfolio the way I did in the first place — not because I’m confident about any single company’s future, but because I’m not, and I’d rather own the whole board than bet on one square.
The number I did not know when I wrote this
There is a hole in the argument above, and I only found it later.
I said an index dilutes single-company risk — that a bad week for semiconductors gets watered down by everything sitting alongside them. That is true of the S&P 500. It is much less true of the index I was watching that morning.
According to Korea Exchange data, as of mid-June 2026 Samsung Electronics and SK Hynix together were worth about 4,162 trillion won — roughly 55% of the entire KOSPI’s market value. Two companies. More than half the index.
For comparison, the ten largest companies in the S&P 500 combined make up somewhere around 36–40% of that index, with the single biggest holding at about 7%.
So when the KOSPI fell 4.46% that morning on chip-demand fears, it was not really an index falling on sector news. It was two companies falling, and the index reporting it. Anyone in Korea who bought “the market” to avoid concentrating on Samsung had, without quite meaning to, bought roughly half a position in Samsung and SK Hynix anyway.
That is the part I would want a reader to take away, and it is more useful than my original point: “buy the index” is not one strategy. It means something completely different depending on which index. The word diversification does a lot of quiet work in personal finance writing, and it is worth checking, for any fund you own, how much of it is actually two or three names.
What happened next
I wrote this on July 20. It is worth saying plainly how the rest of that month went, because a post that stops at the scary morning and never comes back is exactly the kind of writing I do not want on this site.
It got much worse before it got better. On July 28 the KOSPI fell 10.84% in a single session and triggered a market-wide circuit breaker. It triggered another one the next day — back-to-back circuit breakers, which had never happened before in the index’s history. Then, on July 31, it rose 17.91% in one day, the largest single-day gain ever recorded for the index.
July finished down 22.19%, the third-worst month in the KOSPI’s history.
And here is what I actually did across all of it: my scheduled contribution went in, on the same day, into the same funds. I did not add extra on the way down. I did not stop on the way down either. I wrote about what that month did to people who used leverage instead — because the difference between their outcome and mine had almost nothing to do with being right about semiconductors, and almost everything to do with structure.
Where my own argument is weakest
I should apply the same scrutiny to my own holdings that I just applied to the KOSPI.
The S&P 500 is far less concentrated than the Korean market, but it is more concentrated than it used to be. Its top ten holdings have grown from roughly a quarter of the index around 2000 to more than a third today, and the largest single position is now worth more than some entire sectors of the index. When I say I own 500 companies, the honest version is that I own 500 companies with a heavy tilt toward about eight of them, most in the same industry.
I also hold a Nasdaq 100 fund alongside it, which makes that tilt bigger, not smaller. I knew that when I bought it. It is a deliberate choice, not an accident — but it does mean my portfolio is less diversified than the phrase “index funds” makes it sound.
So the fair conclusion is not “I am diversified and Korean investors were not.” It is that concentration exists on a spectrum, everyone sits somewhere on it, and the useful habit is knowing where rather than assuming a label protects you. I have written elsewhere about how little of this machinery can actually be predicted; this is the same lesson from a different direction.
Frequently asked questions
Does buying an index fund protect me from a single company collapsing?
Partly, and it depends entirely on the index. In a broad index where no single holding exceeds a few percent, one company failing is absorbed. In a highly concentrated index — where two companies are half the market value — it is much closer to owning those companies directly.
How do I check how concentrated a fund is?
Every fund publishes its top holdings and their weights on its official fact sheet, usually updated monthly. Look at the top ten and add them up. If that number is large, the label on the fund is describing less diversification than you might assume.
Should I have bought more during the July 2026 sell-off?
With hindsight, buying at the July 28 low would have looked brilliant three days later. That is precisely the problem with hindsight. I had no way to know on July 28 whether it was the bottom or the middle, and neither did anyone telling you otherwise. My schedule bought some of that dip automatically, which is the only version of “buying the dip” I trust myself to execute.
Why hold a Nasdaq 100 fund if it increases concentration?
Because I wanted more exposure to that part of the market than the S&P 500 alone gives, and I decided that deliberately rather than drifting into it. The honest cost is a less diversified portfolio than the words “index fund” suggest, and I would rather name that than pretend otherwise.
This is not investment advice.