South Korea’s benchmark has surrendered more than 40% from its June peak as forced liquidations turn an extraordinary AI-driven rally into a warning for investors everywhere.
MARKET INSIDER — South Korea’s KOSPI has completed a spectacular—and painful—round trip. After almost doubling from late February and approaching 9,500 points in June, the benchmark tumbled toward 5,000 in late July, losing more than 40% from its peak in little more than a month.
The reversal has not erased all of this year’s gains: the KOSPI remained up 41.5% in US-dollar terms as of July 29. But investors who entered near the top—or borrowed heavily to buy the initial decline—face an entirely different outcome. The episode shows how leverage can transform an ordinary correction into a self-reinforcing liquidation event.
Key highlights
- The KOSPI has fallen more than 40% from its June peak after almost doubling from late February. The index plunged nearly 11% on July 28 and as much as 12.6% on July 29.
- Samsung Electronics and SK Hynix together represent more than half of the KOSPI’s market value. More than 1.2 million leveraged retail accounts had reportedly received margin calls by July 13.
- An estimated 320,000 to 360,000 accounts were fully liquidated, according to Goldman Sachs data cited by Reuters. Domestic margin-loan balances reached a record 38.63 trillion won in June.
- Regulators are tightening rules for single-stock leveraged products, but forced deleveraging may not have finished.
From world-beating rally to record rout
The speed of the KOSPI’s reversal is almost as remarkable as the rally that preceded it.
South Korean equities surged as investors chased artificial-intelligence beneficiaries, particularly memory-chip producers SK Hynix and Samsung Electronics. Both companies enjoyed rapidly rising earnings expectations as demand increased for advanced memory used in AI accelerators.
The benchmark nearly doubled in approximately six months and, at its peak, had more than tripled over the preceding year. That performance attracted increasingly speculative capital, including retail money financed through broker margin loans and leveraged exchange-traded products.
Then the cycle reversed.
The KOSPI dropped nearly 11% on July 28 and fell as much as another 12.6% the following session. The cumulative decline exceeded 40% from the high reached just over a month earlier, putting the market on course for a record two-day fall, according to Reuters
This was not simply a reassessment of earnings or valuation. It became a market-structure event in which falling prices triggered margin calls, forced sales and further declines in collateral values.
Buying the first dip became another leveraged trade
The first wave of forced liquidations did not necessarily mark the end of the selling.
When the KOSPI initially fell toward 7,000, some brokers, fund managers and retail investors may have concluded that positions established near 8,000 had already been cleared. That encouraged another round of dip-buying.
But a market does not necessarily bottom when the first leveraged cohort is liquidated. If new buyers enter with borrowed money, leverage is merely transferred to a lower price level.
A further decline can then trigger a second liquidation cycle:flowchart TD A["Rapid rally"] --> B["Leveraged buying"] B --> C["Initial correction"] C --> D["Margin calls"] D --> E["New dip buyers at lower prices"] E --> F["Further decline"] F --> G["Second liquidation wave"]
This is why identifying a supposed “margin-call bottom” is exceptionally difficult. Investors cannot observe every account’s collateral, borrowing terms or liquidation threshold. Apparent exhaustion in selling can be temporary if new leverage enters during the rebound.
Five-times leverage leaves almost no room for error
Reuters documented the experience of 24-year-old university student Lee Seung-ho, who used a trading application offering five-times leverage.
Lee reportedly turned 20 million won saved during mandatory military service into a portfolio briefly worth nearly 300 million won. Within four weeks, forced liquidations erased the gains and reduced his account below its original value.
His experience illustrates the arithmetic of leverage. With five-times market exposure, a 10% decline in the underlying position translates into an approximate 50% loss of investor equity before interest, fees and changes in margin requirements. A 20% decline can theoretically eliminate the entire starting capital.
In practice, a broker will normally liquidate the position before equity reaches zero. During a fast or illiquid market, however, execution can occur at much lower prices, potentially leaving the borrower with additional debt.
Leverage therefore does not merely multiply the final return. It changes the probability that an investor can remain in the position long enough for the investment thesis to work.
Important correction: the apartment was not reported as collateral
Some online accounts have claimed that Lee mortgaged a studio apartment and parking space to fund his trades. That is not what Reuters reported.
The report said Lee was interviewed inside a studio apartment “barely larger than a parking space.” It did not say that he owned or pledged the apartment or a parking space as collateral.
Reuters did report that broader investor debt—including borrowing outside conventional brokerage margin facilities—surpassed 60 trillion won at the end of May. This indicates that some investors may have used bank loans or other credit to fund trading, but it does not establish that Lee mortgaged real estate.
This distinction matters. The documented story is already a powerful example of excessive risk without adding an unsupported property-backed borrowing claim, Reuters reported.
The leverage was concentrated in an unusually narrow market
South Korea’s vulnerability was intensified by index concentration.
Samsung Electronics and SK Hynix together represented more than half of the KOSPI’s market capitalisation by late July. Retail borrowing was also heavily directed toward these companies and leveraged products tracking their shares.
That created several layers of concentration:
- One investment theme: artificial intelligence.
- One industry: semiconductors.
- Two dominant companies: Samsung and SK Hynix.
- One investor group: highly active domestic retail traders.
- One source of instability: leverage and forced liquidation.
When both chipmakers declined together, index funds, leveraged ETFs and margin accounts came under pressure simultaneously.
Retail investors purchased 42.4 trillion won of KOSPI shares in June and another 13.2 trillion won during the first half of July, even as foreign investors reduced exposure. Borrowed investment in KOSPI shares remained near a record at 28 trillion won on July 14. Reuters citied.
The concentration meant that selling in two companies could destabilise the broader national benchmark.
More than one million accounts received margin calls
The scale of the retail deleveraging was extraordinary even before the latest two-day plunge.
According to Goldman Sachs figures cited by Reuters, more than 1.2 million leveraged retail accounts had triggered margin calls by July 13. An estimated 320,000 to 360,000 accounts were fully liquidated.
That was equivalent to approximately one in every 30 South Korean adults receiving a margin call, although the figures refer to accounts rather than necessarily an equal number of unique individuals.
The latest fall toward 5,000 may have exposed another group: investors who survived the first correction or added leverage around the 7,000 level.
South Korea’s market decline could consequently affect more than brokerage balances. Large household losses may reduce consumer confidence and discretionary spending, particularly if investors funded positions with income or unsecured borrowing that must still be repaid.
Why young Koreans accepted such extreme risks
The speculative boom also reflects a deeper problem in South Korea’s economy.
Reuters reported that average Seoul apartment prices were equivalent to roughly 14 years of salary. For young graduates who feel excluded from property ownership and other conventional routes to wealth, leveraged stock trading can appear to offer economic mobility that employment income cannot provide.
During a rising market, that belief receives powerful reinforcement. Investors see peers making rapid gains, brokers make leverage available through a simple application button, and recent performance makes extreme risk appear rational.
For several months, the strategy seemed to work. Young traders who believed they had discovered a faster route to wealth were, temporarily, correct.
The difficulty is that gains produced by leverage are conditional. They exist only while prices move in the right direction and the investor retains sufficient collateral. A profitable position that cannot survive normal volatility is not necessarily a successful investment strategy.
Could actual leverage have exceeded five times?
Online commentators have suggested that some Korean traders may have combined margin borrowing with derivatives or leveraged ETFs, producing effective exposure greater than five times their capital.
That is financially possible. An investor could theoretically borrow against an account and then purchase a product that itself uses swaps or futures to magnify daily returns. Such layering can create exposure far above the headline margin multiple.
However, there is no reliable public evidence establishing how widespread this practice was among Korean retail investors. Reddit discussions should therefore be treated as hypotheses about market structure, not verified data.
Another complication is that leveraged ETFs generally target a multiple of the underlying asset’s daily return. During volatile periods, compounding and repeated rebalancing can cause longer-term performance to diverge sharply from the simple stated multiple.
For an investor already using borrowed money, those path-dependent losses can accelerate the depletion of collateral.
Regulators are attempting to contain the damage
South Korean authorities have acknowledged that leveraged single-stock products were introduced too quickly.
Regulators banned new listings of leveraged ETFs tied to individual shares only two months after approving the products. They also brought forward a rule requiring retail investors to maintain at least 30 million won in cash before initiating or increasing certain leveraged ETF and exchange-traded note positions.
Finance Minister Koo Yun-cheol said the government was reviewing additional market-stabilisation measures as the sell-off deepened.
Those interventions may reduce new speculative positions, but they cannot remove leverage already embedded in existing accounts. Nor can regulatory restrictions immediately restore buyers to a market in which potential investors are waiting for forced selling to finish.
The lesson for investors: survival matters more than identifying the bottom
The KOSPI remains substantially higher than at the beginning of 2026, particularly when measured in dollars. Long-term investors who entered before the rally may still hold considerable gains.
But the outcome is very different for investors who bought near 9,000, used five-times leverage, or borrowed again to average down at 7,000.
The Korean experience provides three broader lessons.
First, a fundamentally attractive investment can still produce disastrous results when purchased with excessive leverage. Samsung and SK Hynix may continue generating strong profits, but those profits cannot prevent forced liquidation in the meantime.
Second, a large decline does not automatically mean leverage has cleared. New borrowing during the first rebound can create another layer of vulnerable positions.
Third, the price at which an investor is forced to sell matters more than the price at which the asset eventually recovers. An unleveraged investor can wait through volatility. A leveraged investor may lose that option.
South Korea’s market could eventually recover, especially if the long-term outlook for AI memory remains strong. Yet for investors whose accounts have already been liquidated, the destination is irrelevant. In leveraged markets, the ability to survive the journey is the investment decision that matters most.
