Market crashes come unannounced. But there are always signs. One domino collapses across the world, sets off a reaction, and it carries lessons for the rest of the world.
In the last few weeks, Korea’s stock market has undergone its fastest and largest drop in history — it fell by more than 20% in the last few weeks, and then recovered a bit. But the signals that showed up before the market crashed in Korea are now showing up right here in the United States.
We’re seeing:
Extreme concentration in a few stocks
A massive rally being carried by AI
Record amounts of retail leverage
The market being priced at its most expensive level since the Dotcom bubble
The term “Market Bubble” has reached a search volume that’s at its highest level since 2004, in the United States. So the question on everyone’s mind is, if it can happen there, can it happen here?
The goal isn’t to predict the crash and time the market — if I could do that, I’d be the most sought-after guy on the planet — but rather to make sure you’re prepared before it breaks. To get the latest updates that protect and grow your money, join 41,000+ investors for free here:
While a stock market bubble builds up over months or years, there are things that could change your life in seconds: Like leaving your information up on the Internet where anybody can find it. Data brokers legally collect and sell your personal information to anyone with a credit card. Your name, address, and phone number are all out in the open and this isn’t just a nuisance. It can lead to risks like identity theft, phishing, and doxing — unless you do something about it.
Signing up to DeleteMe is the simplest way to help protect yourself – and your business – from scams and identity theft. Instead of scrolling away, just take a second to click here:
You get an additional 20% off when you use the code GRAHAM.
Just like we have the S&P 500 in the United States, Korea has the KOSPI, an index that tracks the largest companies in the country. For a while, this was one of the best performing markets on the planet!
2026 has been a wild year for KOSPI. At one point, it was up more than 60% in 2026 alone. It overtook London in total market value. One of their chipmakers, SK Hynix, makes the high bandwidth memory that goes into Nvidia’s AI chips. In June, SK Hynix nearly tripled in value, becoming the most valuable company in Korea for the first time in over 25 years!
To put into perspective how insane the situation is, the S&P 500 has 500 companies, and people are worried that Nvidia makes up 7% of the index. The top 10 companies make up about 40% of the index. But in Korea, just two chip stocks made up more than 50% of the KOSPI. So for Koreans, the “diversified national stock market” was a huge bet on memory chips.
Another ingredient adding to this problem was leverage.
In May, Korea approved a product that had never existed before: Single-stock leveraged ETFs, which is just a fancy way of borrowing money to double the daily moves of one single company. In this case, the ETFs were targeting the two companies that already dominated the market — Samsung and SK Hynix. 16 ETFs launched around the same time, and within just the first two months, retail investors poured in more than $9.6 Billion. They were chasing a rally that had already tripled in size.
Then the regulators decided to step in. On June 22nd, they warned the public that the rally and the leveraged products were becoming dangerously overheated. Korea’s top financial regulator even said that he wished he’d blocked those ETFs from ever launching. Of course this led to a frantic selloff, and it took the entire Korean stock market down with it. As the market fell, a feedback loop kicked in.
As chip stocks fell, leveraged ETFs lost double.
They were forced to rebalance by selling more chip shares.
This pushed the chip stocks down further.
And that created a death spiral. After this brutal sell-off, regulators banned new single-stock leveraged ETFs entirely and tripled the minimum deposit required to trade the existing ones.
But that would just be an interesting incident in a foreign country if it didn’t have some ingredients that we are seeing in our market.
Korea’s crash required five ingredients:
Extreme concentration
A massive preceding rally
Leveraged retail speculation
Foreign selling
Doubts about AI
Where does the US sit on these parameters? As we speak, the S&P500 is sitting around 7500, about 8% higher than it was at the beginning of the year. On the surface, it all looks fine. Once you dig deeper, things start to get really interesting.
The 10 largest stocks now make up roughly 37% of the S&P 500. Nvidia and Apple take up 15% between themselves, and the IT sector is 38% of the index. If you buy the index, you’re hoping to spread your money across 500 companies so that any shift in the market doesn’t hit you suddenly, but more than a third of the money is now riding on just 10 companies.
It’s not as bad as Korea’s two stocks at 50%, but if the top 10 fell by 25%, that would drag the index down by 9% along with it.
The S&P 500’s Forward Price-to-earnings ratio is about 20.7. Compared to the 10-year average of 19, that doesn’t sound too bad. But there’s another measure, the Shiller CAPE ratio that’s around 41 now — during the Dot-com bubble, this measure went up to 44 before the crash, and we’re dangerously close now.
Apart from this:
The price-to-sales ratio is at a record high
Dividend yield is near record lows
The Buffett indicator, which measures the total market value versus GDP, is at 219%
By every long-term measure, this is one of the most expensive stock markets in American history.
FINRA just reported that margin debt hit $1.5 Trillion in June, 49% up from last year. When you subtract the actual cash in those accounts, net margin debt is up 51% YoY.
If those numbers sound complicated, here’s the bottomline: Investors have been borrowing record amounts of money to buy stocks, and even if we don’t have leveraged ETFs like Korea, the margin debt acts the same way. It’s bullish fuel that crashes on the way down, because falling prices trigger margin calls. This forces selling and falling prices.
To buy the dip, you need cash, and right now almost everyone’s gone all-in on equities. Professional money managers are only holding 3.6% of their portfolio in cash. The VIX is around 19, below its long term average. Positioning has been maxed out, and nobody’s hedged for what’s ahead.
US Corporate Insiders are selling stocks rapidly, and they’re buying at half the normal rate. Insiders sell for all sorts of reasons — taxes, diversification, pre-scheduled plans, etc. But combined with everything else, this isn’t exactly a vote of confidence.
So if retail investors are going all-in and are maxed out on stocks, where are the institutions investing?
The institutions are telling a completely different story. Morgan Stanley believes that the S&P 500 should continue to rise this year, ending at 7800. CitiBank sees 7700, and both JP Morgan and Wells Fargo see a relatively flat market for these next few months. Basically, all major banks are aligned that it’s more of the same — a market that will continue to go higher instead of collapsing. The earnings data backs this up. The most recent quarter has been incredible:
Blended earnings growth is near 25%
Revenue growth is near 13%
88% of companies are beating estimates
Analysts expect this momentum to continue through the rest of the year. Compared to the dot-com bubble, where companies literally made no money, this is a very different market. Taking this context into consideration, the “expensive” tag that we saw earlier doesn’t automatically mean that today’s giants are in a bubble: Nvidia, Microsoft, Apple, Amazon, and Alphabet are some of the most profitable businesses in human history.
Have you been holding any of these giants in your portfolio for a long time? How has it worked out for you?
If we look at the broader market indices, they are actually doing better than the S&P 500. Take the Russell 2000 for example, made of small caps — it’s up over 16% this year, nearly double the S&P 500. The equal-weight S&P 500 where every company counts the same recently hit record highs. This is no longer about seven stocks carrying everything: instead, the average company is now participating in the market.
In the case of bubbles, corporates are usually the first to smell trouble and offload the stocks.
But on the demand side, corporate buybacks are already coming in over $400 Billion, announced through late April. This is up 20% from last year. Index funds are seeing automatic inflows as usual, every two weeks, from 401k contributions. Credit markets, which usually smell trouble first, are showing high-yield spreads near historic tights, meaning that bond investors are seeing almost no stress in the system.
Even if 95% of AI experiments fail, the winners in cloud, semiconductors, and software are already monetizing. Hyperscaler revenue is real unlike the Internet companies in the dotcom era that had no business to show, and this buildout is being funded primarily from profits rather than junk debt and IPO money. If AI delivers even modest productivity gains across healthcare, logistics, finance, and manufacturing, we may simply “grow into” today’s valuations instead of overdelivering. When Amazon grew into its dotcom-era price and became one of the most valuable companies on earth, this is what happened.
When we ask “Are we Korea?”, instead of comparing the symptoms in both cases, we need to look at the structures of the two markets.
The US market is structurally different in a few ways:
Our biggest single stock is 7.6% of the index. Their top two stocks are at over 50%.
Our market has the most diverse set of buyers in the world — pensions, foreigners, corporations, retail, index funds, etc. Korea was dominated by leveraged retail chasing two companies.
Single-stock leveraged ETFs exist in the US but are a rounding error compared to their market share in Korea.
Now, the bull case assumes everything goes to plan. But earlier this year, Goldman Sachs also published a bear case: How things could actually go the other way, when an oil shock from the Iran conflict drags the S&P 500 down to 5,400. This would be a 38% decline from today’s levels. The wild part is that the spread between the bull case and bear case was the widest range they’d published since 2020. The range of outcomes is just enormous, right now.
Why could the bear case happen?
The bull case depends on 24-25% earnings growth happening. The market is already priced for near perfect execution and double digit growth. But if earnings miss by just 5%, that could cause a 25% decline in the S&P500 even without a catastrophe.
AI spending math might catch up. The five biggest hyperscalers are expected to increase CapEx by $534 Billion through 2027 while operating cash flow only grows by about $340 Billion (spending $1.57 for every dollar). If AI revenue catches up, that could work. If it doesn’t, all the major companies could get repriced at the same time just like in Korea, triggering a crash.
The wealth gap: Though smaller companies are outperforming right now, bigger companies pull their weight. On more than half the trading days this year, the S&P 500 has moved in the opposite direction of the majority of its own stocks. That level of divergence is comparable to the year 2000.
Leverage: $1.5 trillion in margin debt, up 49% in a year, is completely fine until a crisis. Then it becomes the American version of Korea’s leveraged ETF doom-loop: prices drop, margin calls hit, investors are forced to sell, and then prices drop further triggering more margin calls.
Cash reserves: Money managers are already 96% invested. So who’s left to buy? Credit spreads at 2.7% mean that bond investors are being paid almost nothing for extra risk. Oil prices swinging up and down threaten the inflation progress that will let the Fed cut rates.
The bear case isn’t that the world ends but rather that everything goes back to normal. Considering how far we’re flying in the clouds right now, normal is a long way down! With so much uncertainty about the future, we need a plan that’ll help us prepare for both the best and worst case scenario.
Will the United States pull a Korea any time soon? I think the answer is no.
Korea’s market is structurally different from ours and it was dangerously unbalanced. On top of that, billions were pumped in all of a sudden into highly leveraged funds in an already unbalanced market. We have decades of circuit breakers and a much more balanced market, and unlike Korea’s regulators, our regulators aren’t going to trash talk the market (hopefully).
But the lesson here isn’t to avoid Korean stocks. We have a problem too: Concentration + Leverage + Euphoria can lead to a disastrous outcome quickly. If you just own the index, a 20-40% drop is tough to swallow, even if not catastrophic. What is necessary is the resilience to overcome that dip.
So it’s the same boring playbook that’s worked for a hundred years that I’m going to recommend:
Keep a 6 month emergency fund
Diversify as much as possible throughout the United States, International Markets, Bonds, Alternative Assets, Real Estate, etc.
No matter what, don’t use leverage to buy stocks.
Nobody knows if Goldman’s right or the bears are right. But in the entire history of the S&P 500, a 20-year holding period has never once lost money. The people who’ve followed that for the 9 years I’ve been making content online have done quite well.
At the end of the day, the truth is that neither does the market have to be a bubble to be dangerous nor does it have to suddenly rally for you to come out ahead. As long as you’re consistently buying in, patient, and investing regularly, you’ll do okay.
If you found this useful, please like, restack, and share it with a friend.
I’ll see you next week.
— Graham
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.