The Bank of Korea fully tightens "borrowing money to speculate in stocks," Goldman Sachs advises clients to hedge against downside risks in the KOSPI.
```
The risk signals in the Korean stock market are no longer just “rising too fast.” More crucially, part of the money fueling the market comes from credit loans and overdraft limits, and this funding channel is now being tightened by banks. Major Korean banks first restricted mortgages, and are now targeting credit loans, overdraft accounts, and online loan channels, with a direct goal: suppressing leverage-driven stock speculation.
Goldman Sachs trader Alvin So stated in a client note that AI and semiconductor narratives remain strong, and valuations are still attractive, but “a new structural dynamic is becoming more important to the market than ever before.” This dynamic refers to the amplifying effects brought by leveraged ETFs and derivative positions. The point is not to immediately turn bearish on Korean tech stocks, but to remind clients: core longs can be kept, but use options to cover short-term pullbacks.
The numbers are already extreme. Korean leveraged ETF assets have reached about $40 billion, around 2.6% of the free float market capitalization; implied volatility in Korean options has risen to about 80%, compared to just 20% a year ago; skew is near highs, and the volatility term structure is inverted. In other words, rising stock prices haven’t suppressed risk pricing; instead, volatility is rising along with prices.
The credit side is cooling in sync. Last month, Korea’s entire financial system saw household loans increase by 9.3 trillion won, of which credit loans in May rose by 3.4 trillion won, while in April they decreased by 900 billion won. After regulators launched emergency management mechanisms, banks set loan limits, suspended certain channels, and cut unused overdraft lines. If stock market positions continue to be based on short-term borrowing and leveraged ETFs, the real danger isn’t a typical pullback, but the simultaneous occurrence of financing contraction and position rebalancing.
Banks first squeezed mortgages, now they're squeezing "stock buying money"
Korean banks previously controlled household loans, mainly focusing on mortgages, such as pausing certain mortgage guarantees and setting branch limits. Now, the scope of restriction has expanded to credit loans and overdraft accounts.
Shinhan Bank has begun restricting non-face-to-face credit loan applications: when offline and online applications exceed internal daily management thresholds, remote applications are tightened. Small loans for financially vulnerable groups and “win-win” refinancing products are not affected by the restrictions.
Handling of overdraft accounts is more detailed. For household credit loans with contract limits over 30 million won, if the overdraft usage rate is below 10% during the contract period and three months before maturity, the limit may be cut by up to 20% at renewal. This means banks are less willing to let clients keep unused credit lines long-term, especially during stock market rises, as these limits can become stock-buying funds at any time.
Kookmin Bank has lowered the maximum credit loan limit to 100 million won and overdraft account limit to 50 million won. Hana Bank has also capped the maximum limit for credit loan applicants at 100 million won, no longer amplifying it based on annual income; and has eliminated certain exceptions in overdraft account renewals, strengthening reductions of unused limits.
Woori Bank is more direct: it has stopped handling new credit loans and refinancing through Toss, KakaoPay, Naver Financial, Finda, Banksalad, and other internet platforms, and has also suspended refinancing products on its own app.
Internet banks were taking over demand, now they’re shutting gates
After major commercial banks tightened, loan demand naturally flowed to internet banks. This outlet was quickly blocked as well.
Kbank, from June 16 to the end of next month, temporarily pauses new overdraft accounts and plans to lower credit loan limits for high-income people. Toss Bank has cut the maximum credit loan limit to 100 million won and overdraft account to 50 million won; if customers already have a 50 million won overdraft account, they can still get another 50 million won credit loan. Kakao Bank, from the 22nd, lowered overdraft account maximum limits to 100 million won; for overdraft accounts with contractual limits of 50 million won or above, if usage in the past six months has not exceeded 20%, the limit at renewal can be reduced by up to 20%.
Korea’s so-called “minus account” is essentially an overdraft credit line tied to the account. Clients can borrow at any time within the approved limit, functioning much like a short-term revolving personal loan. During a stock market rally, it can easily turn into investors’ spare ammunition.
This is what regulators worry about most: stock market gains drive anxious buying, bank credit becomes a source of stock buying funds, and loan growth further supports stronger risk appetites. Now banks are tightening limits, entry points, and renewal conditions together, effectively cutting off the part of the financing chain most likely to inflate during a rally.
Leveraged ETFs make late-day trading riskier
Tightening credit is just the first layer of risk. The second layer comes from the mechanisms of leveraged ETFs themselves.
Leveraged ETFs restore their leverage ratio daily; the greater the market’s ups and downs, the more concentrated late-day rebalancing flows become. According to calculations, if the market swings 5%, the Korean market could trigger around $4.7 billion in dealer delta rebalancing, about 13% of daily trading volume.
The pressure is more obvious in large-cap stocks. Samsung Electronics, SK Hynix, and TSMC are index core components, and each has single-stock leveraged ETF products. At the individual stock level, related flows can exceed 20% of average daily trading volume.
This means a 5% swing is not just a price change but triggers systemic passive buying and selling. When prices rise, dealers are forced to buy; when prices fall, they’re forced to sell. The market thus is prone to “overheating rises” and “fast declines.”
Volatility rises along with stock prices—an unhealthy signal
Normally, during strong stock market rallies, implied volatility usually drops, because the market’s fear of declines lessens. But Korea now presents another picture: spot prices rise, and option implied volatility rises too.
Korean option implied volatility has risen from about 20% a year ago to about 80%. Skew is near highs, meaning the market is willing to pay more for options in certain directions; term structure is inverted, meaning near-term risk pricing is higher than long-term, with short-term uncertainty pushed up sharply.
This kind of combination often points to crowded trades. Investors chase the upside, while the options market simultaneously signals defensiveness against short-term violent volatility. When dealers are short gamma, if the market moves, they control exposures by buying as prices rise and selling as prices fall, further amplifying price movements.
So the real issue isn’t whether the AI and semiconductor stories fail, but whether positions are already fragile enough. Fundamentals remain, but trading structure might break first.
Hedging is not retreat, it’s adding guardrails to long positions
This framework doesn’t ask investors to liquidate Korean tech stocks. On the contrary, the core idea is to keep key positions, and use derivatives to cover short-term pullback risks.
Compared to directly buying puts, put spread collars emphasize cost control: by selling higher strike calls, you trade off part of extreme upside for downside buffer, buying a lower strike put spread for protection.
Examples of zero-cost structures given include: selling SK Hynix 153% calls, while buying 85/55% put spreads; selling Samsung Electronics 140% calls, while buying 85/55% put spreads; selling Kioxia 155% calls, while buying 85/55% put spreads. All expiring December 10, 2026.
The message is clear: if stocks keep rising 40–55%, longs remain profitable; if there’s a technical pullback, the structure provides protection. This isn’t betting on a fundamental collapse, but acknowledging trading risk has already risen.
True breaking point: credit and positions reverse together
The trickiest part of the Korean market now is that two variables are changing at the same time.
On one side, banks are tightening credit loans and overdraft lines, raising the difficulty of fresh leveraged buying; on the other, leveraged ETFs, options, and dealer gamma positions make the market more sensitive to price swings. The former impacts funding sources, the latter impacts price transmission.
If it was only banks restricting loans, the market might not immediately fall; if only leveraged ETFs were growing, the index could keep rising for a while. But together, market tolerance for a pullback decreases. Loan valves are narrowed, fresh buying declines; leveraged positions need rebalancing, selling pressure might be magnified.
This is why “being bullish on AI and semiconductors” isn’t contradictory with “putting insurance against KOSPI pullbacks.” The rally narrative remains, but funding structures are no longer clean. What Korean stocks now need to watch most isn’t the bad news itself, but who will be forced to sell when bad news strikes.
Risk Warning & DisclaimerThe market has risks, investment requires caution. This article does not constitute personal investment advice nor does it consider individual users’ unique investment objectives, financial situations, or needs. Users should consider whether any opinions, views, or conclusions herein are suitable to their circumstances. Investing based on this, responsibility is yours. ```