I heard shares fell because Samsung Electronics and SK hynix finished buying back their own stock. Could next week's hynix earnings be a turning point? I'm wondering when the KOSPI might bottom.
2026-10-23
What the end of buybacks really means
Samsung Electronics and SK hynix bought back about KRW 55 trillion worth of their own shares this year. The problem is that foreign investors sold the same stocks heavily during that period. From the start of the year through October 8, foreign investors were net sellers of about KRW 86.1 trillion of SK hynix and about KRW 85.5 trillion of Samsung Electronics, and SK hynix's foreign ownership fell from 53.83% to 49.59%, its lowest of the year (Sidae Ilbo, 2026-10-12). In effect, foreigners sold into the shares the companies were buying. With that supporting bid gone, selling pressure showed up directly in prices, and that was the 10/22 decline.
There is also a counterargument
Some argue the end of buybacks need not be seen only as negative. Lee Sang-heon of iM Securities viewed foreign selling during the buyback period as rebalancing to match weightings (SBS Biz, 2026-10-06).
"Because they responded with selling as part of rebalancing" — Lee Sang-heon, iM Securities (SBS Biz, 2026-10-06)
When a company buys and cancels shares, the free float shrinks, and foreign investors tracking an index must sell a corresponding amount to keep their weightings. When the buyback ends, this mechanical selling ends too. The same article concluded that Q3 earnings are ultimately the key to filling the gap.
Expectations for hynix's 10/28 earnings are already high
According to FnGuide, SK hynix's Q3 operating profit estimate is about KRW 77–79 trillion, up about 4% over the past month (Financial News, 2026-10-05; Sidae Ilbo, 2026-10-12). Samsung Electronics reported preliminary Q3 operating profit of KRW 107.4 trillion (Sidae Ilbo, 2026-10-12). Yet Samsung Electronics' shares fell sharply after it posted record profit. It is a precedent showing that "good numbers" are not enough, and that what is said about next-quarter prices and customer investment plans drives shares more than the numbers themselves.
Three possible paths
| Path | Conditions | Expected flow |
|---|---|---|
| Flow reversal | Earnings beat with HBM and DRAM prices staying strong and mention of expanded investment next year | With foreign rebalancing selling over, a turn to net buying. Won strength (1,324 won) adds expectations of FX gains |
| Time correction | Earnings in line, but hyperscaler funding worries unresolved | Index moves sideways around 5,500 with thin turnover. Repeated pattern of rebounds fading within a day |
| Further decline | Conservative guidance or signs of peak memory prices | KOSPI lows keep falling, retesting the 5,400s. A further BOK hike on 11/26 adds pressure |
Whichever path, it won't end on 10/28 alone. That's because Amazon's earnings on 10/29 will again test the funding source for end demand for memory. Even if hynix's numbers are good, if Amazon signals "we're short on cash," foreign investors are likely to wait.
How to read it alongside won strength
If foreign investors were leaving Korea, the won should weaken. Yet the won is at 1,324.2 per dollar, its strongest this year. This means foreign selling is concentrated on reducing weightings in two chip stocks rather than Korea as a whole, and that the door is open for them to return once the reason for selling disappears. On the other hand, won strength has already made Korean stocks less cheap for foreigners measuring in dollars.
So what to do now?
- Don't buy all at once before earnings. As the Samsung Electronics precedent shows, volatility is high around the release date. If you want to buy, it is better to buy one-third before earnings and the rest after confirming earnings and Amazon's release.
- The confirming signal is consecutive net foreign buying. Watch whether foreign investors buy SK hynix several days in a row after earnings and whether foreign ownership rebounds from the 49% range. That is the evidence that rebalancing selling is really over.
- Korea ETFs like EWY are not yet showing an entry signal. On the signals, as of 10/20 it was already about 10% below the exit line (trail_buf -10.29%) and -30.29% from its high, and on 10/22 it fell further to 149.06.
- Treat stabilization fund activation only as a secondary signal. Some analysis shows past mentions of activation came near bottoms, but a fund the size of 2020's (KRW 10.76 trillion) would struggle to offset the scale of foreign selling in the two chip stocks.
Detailed charts and signals for other tickers are available on the /signals dashboard.