Korea Doubled, Then Fell 39%. The Fund Americans Can Buy Is Still Up 56%.

South Korea's stock market set a record in June, then lost more than a third of its value — 19% in October alone. It is now the cheapest trillion-dollar market in twenty years. And the US-listed fund that tracks it is still up about 56% this year. Every one of those statements is true at the same time, and understanding why is worth more than any of them.

What actually happened

The KOSPI, Korea's main index, closed at a record 9,114.55 on 22 June 2026 after a rally so steep it roughly doubled inside the year. By late October it had fallen about 39% from that high, including a drop of roughly 19% during October alone. At one point it was the worst-performing market in the world for the month.

That is not a wobble. A fall of that size, that fast, in a developed market of that size, is rare.

The number that makes professionals stare

Korea's benchmark now trades at a forward price-to-earnings ratio of 4.2.

Price-to-earnings is roughly how many years of profit you are paying for. A P/E of 4.2 means the market is priced at about four years of expected earnings. For comparison, the long-run US average sits in the high teens, and we noted recently that Tesla trades near 407.

Of the 18 markets worth more than $1 trillion at the start of this year, none had its main index fall below a forward P/E of 5 in nearly two decades. Korea is the first since records began in 2006.

Before you read that as an opportunity, hold on — the last section deals with exactly that instinct.

Why it fell

Several things gave way at once, which is usually how this works.

• The AI trade cooled. Korea makes the memory chips the AI build-out runs on. When investors started questioning returns on AI spending, Korea was directly in the path — the same chip rout that hit the whole region.
• Two stocks carried the damage. Of all the shares foreign investors sold, roughly 80% was concentrated in Samsung Electronics and SK Hynix: 5.2 trillion won of Samsung and 12 trillion won of SK Hynix between September and late October.
• The buyers stopped. Samsung and SK Hynix ended their share buybacks, removing a large and reliable source of demand.
• Foreigners left for five straight months, May through September, with net selling reaching 21.6 trillion won from September onward.
• Liquidity drained. Trading turnover fell about 80% from its May peak. Thin markets fall faster, as we saw at the opposite extreme with a Ghanaian micro-cap.
• Leverage unwound. Losses on leveraged products drained retail buying power, and Korea's National Pension Service had less capacity to buy after its own valuation losses.

Notice how little of that is about company earnings. One analyst quoted on the de-rating put it plainly: the plunge is "not because the buybacks ended, but because the market has entered a bear market." Prices fell because owners sold, not because the businesses suddenly became worth a third less.

The currency did the opposite of what it should have

This is the part almost nobody outside Korea has written about, and it is the most interesting thing in the whole story.

Normally a market and its currency move together. Foreign investors selling Korean shares convert won back into dollars, which pushes the won down. That is the textbook link.

It broke. While foreigners dumped tens of billions of dollars of stock, the won went the other way:

A lower number means a stronger won. It strengthened roughly 14% against the dollar during the same stretch in which the stock market collapsed.

The reason is chips again, from the other direction. Korea's semiconductor exports are at records — $121.1 billion in September, with the cumulative total reaching $814.7 billion, already above last year's full-year record of $709.3 billion. Those exports flood the currency market with dollars that get converted into won. The selling pressure from departing investors was simply outweighed by the dollars rolling in from selling memory to the world.

So why is the American fund still up?

Here is where beginners get caught. As of 23 October, EWY, the main US-listed South Korea fund, was up about 56% for 2026 and about 73% over twelve months — while the headlines said Korea had crashed.

Both are accurate, because they measure different things.

• Different starting point. "Down 39%" is measured from the June record. "Up 56%" is measured from 1 January. The market doubled on the way up, so it can give back a third and still be far above where the year began.
• Different currency. The KOSPI is quoted in won. EWY is quoted in dollars. A won that strengthened roughly 14% cushions a dollar investor against part of the fall — the exact reverse of what we found in African markets, where local gains shrank once converted.
• Different baskets. EWY tracks an index of Korean shares, not the KOSPI itself, so the holdings and weights differ.

We are not going to hand you a precise bridge between the two numbers, because an honest one would need data we do not have. The point is the principle: whenever you see a market's return, ask over what period, in what currency, and measured by which basket. Change any of the three and the same market can look like a triumph or a disaster.

On the temptation to call it cheap

A forward P/E of 4.2 on a developed market is the kind of number that makes people reach for a buy button. Three things are worth sitting with first.

Cheap can get cheaper. A low valuation is a description of price against expected earnings, not a floor. If those expected earnings fall — and memory is a famously cyclical business where profits swing violently — the ratio rises again without the price ever recovering. A cyclical company at its cheapest-looking is often at the top of its cycle, which is the trap we described in the Micron piece.

Korea has been structurally cheap for decades. There is a long-standing discount on Korean shares, usually attributed to governance and ownership-structure concerns. A market being cheaper than its peers is not automatically a mispricing waiting to correct.

You would be buying two companies. Samsung and SK Hynix dominate both the index and the selling. "Investing in Korea" through a broad fund is, in practice, substantially a bet on memory chips.

We do not publish price targets or buy and sell calls, and we are not making an exception here — our editorial policy explains why. What we will say is that "the P/E is 4.2" is the beginning of a question, not the end of one.

Sources

1. BigGo Finance, "South Korean Stocks Hit Valuation Floor: KOSPI Forward P/E Falls to Just 4.2x," 23 Oct. 2026 — valuation, the 18 trillion-dollar markets comparison, October decline, causes and analyst quote — finance.biggo.com
2. Bloomingbit, "Kospi Stays 25.3% Below Record, Struggles to Reclaim 7,000 as Foreigners Sell," Oct. 2026 — record close of 9,114.55 on 22 June, foreign net selling, Samsung and SK Hynix concentration — en.bloomingbit.io
3. The Investor (Korea Herald), "Why Kospi and Korean won have parted ways" — won levels, semiconductor export figures, foreign flows and the broken market-currency link — theinvestor.co.kr
4. The Asia Business Daily, "Foreign Investors' Selling Blitz: KOSPI Is the World's Worst Performer This Month," 19 Oct. 2026 — asiae.co.kr
5. StockAnalysis, country fund comparison including EWY — year-to-date and one-year returns as of 23 Oct. 2026 — stockanalysis.com