On Sept. 23, the last trading day before the Chuseok (Korean Thanksgiving) holiday, individual investors were net sellers of about KRW 1.44 trillion worth of stocks in the KOSPI market (they sold more than they bought). Yet KOSPI actually closed 0.90% higher at 7,080.92. Foreigners also sold about KRW 510 billion, but institutions bought about KRW 320 billion, and buying by ‘other corporations’ (ordinary companies that are neither institutions nor foreigners) is seen as having lifted the index.
Anyone holding stocks going into a holiday faces one question: sell and rest, or hold through the break? Heavy selling by individual investors before a holiday is often read as a sign that ‘something feels off.’
But Chuseok data going back to 2000 tell a different story. KOSPI rose after the holiday in more years than not, but that pattern was weaker in years like this one, when the market had already climbed going into the break. And the Chuseok pattern that Korea Investment & Securities pointed to was not in stock prices but in trading value.
- What retail investors sold before the holiday was their gains and the ‘risk of the holiday’
This year's Chuseok holiday runs from Sept. 24 to 27. The Korean stock market is closed on the 24th and 25th and reopens on Monday, Sept. 28. Meanwhile, U.S. stocks trade as usual, and a U.S.-China summit is scheduled in Washington on the 24th (local time). Korean investors can hear the news from abroad but cannot buy or sell Korean stocks during the holiday.
Past memories also play a part. During the 2019 Chuseok holiday, drones attacked oil facilities of Aramco, Saudi Arabia's state oil company, and global oil prices soared. In 2021, the bankruptcy crisis at Evergrande, a major Chinese property developer, shook stock markets worldwide.
Interestingly, selling before the holiday doesn't even put cash in your hands during the break. Stock sale proceeds arrive two business days later, so shares sold on the 23rd won't pay out until the 29th. Selling before the holiday is less about needing money and more about locking in gains after the rally and shedding ‘risk you'd rather not hold.’
- 26 Chuseok holidays: KOSPI rose after 19 of them
According to the Korea Exchange, of the 26 Chuseok holidays from 2000 to 2025, KOSPI rose over the five trading days after the holiday in 19 years, or 73.1%. Over the five trading days before the holiday, up years and down years were split exactly in half, at 13 each (50%).
The numbers vary depending on how they are counted. In an analysis by Korea Investment & Securities (a major Korean brokerage), KOSPI rose in the five trading days after the holiday 16 times (61.5%) out of the same 26, and in the five trading days before it 14 times (53.8%), with average returns of -0.3% before the holiday and 0.7% after. Korean brokerage Daishin Securities analyzed 22 Chuseok holidays from 2004 to 2025 and also found a 68% chance of gains in the five trading days after the holiday (15 of 22), with an average return of 1.34%, better than the average for the five trading days before (-0.05%).
Whichever count you use, the period after the holiday beat the period before. But the Chuseok effect shows up more clearly in a different number than returns.
- What really changes is not stock prices but trading value
According to Korea Investment & Securities, over the past five years KOSPI's average daily trading value (the total value of shares bought and sold in a day) around Chuseok fell from KRW 10.7 trillion to KRW 9.9 trillion just before the holiday, then rose to KRW 11.6 trillion after it. Kim Dae-jun, an analyst at Korea Investment & Securities, said, “Statistically, Chuseok has a calendar effect in trading value, not in returns.” A calendar effect is when the stock market shows a similar pattern at a particular time each year.
Think of the empty roads just before a holiday. Fewer cars doesn't mean the city has collapsed. People have simply left the roads for a while, and the cars come back when the holiday ends. By Korea Investment & Securities' count, in the 26 years from 2000 to 2025, trading value rose in the five trading days after the holiday in 17 years (65.4%).
This year, too, the roads have started to empty. In the week of Sept. 14 to 18, KOSPI's average daily trading value was KRW 20.1 trillion, down KRW 3.7 trillion from the week before.
- When the statistics weaken (1): heading into the holiday after a rally
This year starts from a different place. KOSPI rose 5.40% over the five trading days through Sept. 23. Unlike the usual ‘weak before, strong after’ pattern, where stocks are weak before the holiday and strong after it, the market went into the holiday already up.
Years like this have done worse than average. According to a tally by Korean financial daily Edaily, of the 14 times since 2000 that KOSPI rose in the five trading days before the holiday, it also rose after the holiday in 8 years. In Daishin Securities' tally, too, gains continued in only 7 of the 11 years in which the market rose before the holiday. In the same Daishin Securities tally, in years when the market was weak before the holiday, KOSPI rose an average of 1.65% over the five trading days after it, with a 73% chance of gains. In other words, the post-holiday rebound came more often in years when the market was weak before the holiday.
- When the statistics weaken (2): when the world moves while the market is closed
The direction on the first day after the holiday has more often matched overseas markets than the calendar. According to a tally by The Financial News, in four of the past five years, the direction of the S&P 500 (the benchmark index of large U.S. companies) during the Chuseok holiday matched the direction of KOSPI on the first day the Korean market reopened.
This year's holiday is also packed with big events. A U.S.-China summit is scheduled for the 24th (local time), and on the 30th, right after the holiday, U.S. memory chipmaker Micron is set to report earnings and the U.S. August Personal Consumption Expenditures (PCE) price index is due. If something big happens abroad during the holiday, as in 2019 and 2021, the 26-year average won't be that year's answer.
- BITPRESS Insight
The ‘stocks rise after Chuseok’ formula worked better in years when the market fell before the holiday. In Daishin Securities' tally, the chance of gains after the holiday in years that were weak beforehand (73%) was higher than in years that had already risen, like this one (7 of 11). The post-holiday rebound was often less about new good news and more about recovering what had been lost before the holiday.
So there is little reason to read individual investors' KRW 1.44 trillion in net selling before the holiday as a sign that ‘bad news is coming.’ Proceeds from shares sold on the 23rd don't arrive until the 29th. Rather than raising cash to spend over the holiday, the selling looks more like taking profits after a gain of more than 5% in five trading days and shedding the risk of a few days in which they can't act.
What Chuseok clearly changes is how much is traded, and in four of the past five years, KOSPI moved on the first day after the holiday in the same direction as U.S. stocks did while the Korean market was closed. There is an average of 19 out of 26, but in years like this one, when the market had already risen before the holiday, it was 8 out of 14. The calendar alone is a weak reason to buy or sell.
Sources
https://www.hankyung.com/article/202609218300i
https://edaily.co.kr/News/Read?mediaCodeNo=257&newsId=04149206645583072
https://weekly.hankooki.com/news/articleView.html?idxno=7184785
https://www.mt.co.kr/stock/2026/09/23/2026092315445242797
https://www.mt.co.kr/stock/2026/09/23/2026092310090072468
https://www.dailian.co.kr/news/view/1694220/
https://www.news1.kr/finance/market-exr/6300685
https://biz.heraldcorp.com/article/10883232
https://biz.heraldcorp.com/article/10879962
https://www.fnnews.com/news/202609211646264641
https://www.gukjenews.com/news/articleView.html?idxno=3705132
https://www.insnews.co.kr/news/articleView.html?idxno=92849
https://www.sedaily.com/article/20093457
https://www.wowtv.co.kr/NewsCenter/News/Read?articleId=A202609210161
https://www.seoul.co.kr/news/economy/securities/2026/09/16/20260916500208
https://news.jkn.co.kr/post/995199
Glossary
Net selling — When the amount sold over a given period is greater than the amount bought; the difference is the net selling amount.
Trading value — The total value of all shares bought and sold in the market in a day, showing how actively money is moving through the market.
Calendar effect — A pattern in which the stock market behaves similarly at certain times each year, such as year-end, January, or major holidays.
Weak before, strong after — A pattern in which stock prices are weak before a certain point in time and strengthen after it.
Other corporations — Ordinary companies that are neither institutions, such as securities firms and asset managers, nor foreigners; companies buying back their own shares also fall into this group.
KOSPI — The Korea Composite Stock Price Index, the main index of Korea's stock market, tracking all common stocks listed on the Korea Exchange's main board. It works like a Korean counterpart to the S&P 500.
Chuseok — Korea's autumn harvest holiday, often called Korean Thanksgiving. Families travel home for several days, and the Korean stock market closes while markets abroad keep trading.
Individual (retail) investors / ‘ants’ — Korean market data split trading into individuals, foreigners, institutions and other corporations. Individuals are ordinary people trading with their own money, nicknamed ‘gaemi’ (ants) in Korea because they are many and each trades small amounts.
