Neither the Securities Market Stabilization Fund nor a blanket short-selling ban has an automatic trigger in Korea. No statute says the switch flips at a given index level; both have always been discretionary calls by the Financial Services Commission. That is why trying to compute the trigger from an index number usually fails. What is left is to tabulate the size, duration and stated purpose of the four past episodes, and then check whether those conditions are recurring now.
Start with the scale of the drop. On 28 July the KOSPI closed at 6,023.66, down 732.09 points or 10.84% — the fourth-largest single-day percentage decline on record. Foreign investors sold a net 4.99 trillion won and institutions 633 billion won that day, while retail investors bought a net 4.33 trillion won, Newsis reported. The July decline came to 28.9%, the worst monthly fall since 1990, and 33.5% from the 19 June peak (Hankyung). The KOSDAQ fell 7.72% to 705.85, its first close below 700 since 16 April 2025.
The fund has been assembled four times
The stabilization fund is not government money. Banks, brokerages and insurers, together with market infrastructure bodies such as the exchange and the depository, pool capital as a civil-code partnership and buy listed shares. Because the size is only fixed once the contribution talks close, there is always a lag between the phrase "reviewing activation" and an announced total. And being assembled does not mean being deployed.
| Episode | Size (KRW) | Trigger event | Outcome |
|---|---|---|---|
| 1990 (stabilization fund) | 4.85tn | Market collapse | Dissolved Aug 1996 |
| 2003 (1st) | 400bn | Credit-card debt crisis | Recovered within a year |
| 2008 (2nd) | 515bn | Global financial crisis | Rebound in early 2009 |
| March 2020 (3rd) | 10.7tn | COVID-19 | Never deployed |
The 2020 vehicle combined 10 trillion won from policy banks and other financial firms with roughly 700 billion won from market infrastructure bodies. It was the largest ever, yet no actual buying was executed. The market moved anyway: after the plan was announced, the KOSPI rose 8.6% on 24 March and 5.89% the following day, according to Business Post.
Comparing it to one day of flows gives a sense of proportion. 10.7 trillion won is about 2.1 times the 4.99 trillion won foreigners sold on 28 July alone. Fully deploying the largest fund ever assembled would, at the current pace, absorb barely two days of foreign selling. That arithmetic is why many read the 2020 effect as a signal — "the authorities will intervene" — rather than as the weight of the money. By the same logic, a signal repeated without follow-through burns out fast.
What actually triggers a short-selling ban?
There is no threshold. Since 2000 there have been four blanket bans; three targeted crisis stabilization and one had a different aim. Hwang Se-woon of the Korea Capital Market Institute notes that the 2008, 2011 and 2020 measures were about calming a crisis, whereas the November 2023 ban was aimed at eliminating naked short-selling through systemic reform (KCMI). Different purposes imply different exit conditions: the crisis type lifts when markets settle, the reform type lifts when the plumbing is fixed.
Computing the durations directly from the start and end dates:
| Episode | Start | End | Days | Months | Purpose |
|---|---|---|---|---|---|
| 1st | 2008-10-01 | 2009-05-31 | 243 | 8.0 | Global financial crisis |
| 2nd | 2011-08-10 | 2011-11-09 | 92 | 3.0 | Euro debt crisis |
| 3rd | 2020-03-16 | 2021-05-02 | 413 | 13.6 | COVID-19 |
| 4th | 2023-11-06 | 2025-03-30 | 511 | 16.8 | Naked short-selling reform |
| Total | — | — | 1,259 | 41.4 | — |
Months are days divided by 30.44. From the first start date (1 October 2008) to the last lift date (30 March 2025) is 6,024 days, of which 1,259 fell under a blanket ban — 20.9%. In other words, in the Korean market a full short-selling ban has been less an exceptional emergency measure than a state that was switched on roughly one day in five. The fourth ban applied to every listed stock from its first day, 6 November 2023 (Newspim).
Did the index rise while bans were in force?
This is where readings diverge. During the fourth ban (6 November 2023 to 30 March 2025) the KOSPI gained 11.35%. Taken alone that looks like evidence the ban worked, but over 17 months the number only means something next to global index returns for the same window. Hwang argues that short selling is a price-neutral trading mechanism, much like margin buying, and that performance during a ban is not evidence that restricting it lifts prices.
Post-lift moves point in no single direction either. After the 2008 ban was lifted the KOSPI fell 0.4% over one month but was up 14.0% at three months; on the day the 2011 ban ended the index dropped as much as 4.94%, yet three months later it was up around 10%. Lifting does not map onto decline, nor banning onto advance. What a ban actually changes is one axis of volatility rather than direction. In a market where panicked retail selling overlaps with leveraged-product rebalancing flows, removing one leg of the selling pressure can compress intraday drawdowns.
Having no trigger line means it is always possible — and equally that it is guaranteed at no index level at all.
The four cards being discussed now
Lee Kyung-soo of Hana Securities argues that meeting any one of four conditions could quickly calm excessive fear: foreign investors turning net buyers; a shrinking balance in single-stock leveraged ETFs; a temporary short-selling ban; or a stated willingness to activate the stabilization fund. On flows, his point is that Korea's weight in the MSCI Emerging Markets index has fallen sharply enough that a passive-rebalancing return is the thing to watch.
The second card carries a number. Roughly 20 trillion won sits in single-stock leveraged ETFs against a government-cited target of 5 trillion won or less — a required 75% reduction in balance. Because that unwind takes time, the third card, a temporary short-selling ban, is framed as the realistic bridge in the interim. The structural drag inside these products is covered separately in the negative-compounding mechanics of leveraged ETFs.
The authorities remain positioned on the supplementary-measure side. On 29 July, FSC Chairman Lee Eok-won told the National Assembly's National Policy Committee that he takes the volatility around single-stock leveraged ETFs seriously and would pursue further measures decisively, citing pre-trade simulated dealing, prior-experience requirements and per-investor caps as items under review once the raised minimum deposit is assessed (Electronic Times). Steps that touch the products themselves — delisting or cutting leverage ratios — and market-wide steps such as the stabilization fund or a short-selling ban have not been confirmed as formally under review.
What to watch
There is no way to know in advance whether either measure fires, but the indicators that move ahead of it are a fixed set. These change earlier than, or alongside, the index itself.
- Daily foreign net selling — 4.99 trillion won on 28 July is the reference point. Does that pace persist, or flip to net buying?
- Net assets in single-stock leveraged ETFs — the distance from roughly 20 trillion won to the cited 5 trillion won target
- Frequency of circuit breakers and sidecars — triggers and stages are laid out in the difference between circuit breakers and sidecars
- Short-interest balances — per-stock figures can be pulled directly using how to check short-selling balances
- The wording used by the FSC and FSS — whether "assessing" shifts to "reviewing" and then "preparing." Because the fund requires contribution talks first, reports of bank-sector contribution discussions are the de facto leading signal
The conditions under which this framing breaks are also clear. First, all four past episodes rested on discretion, so identical indicator levels can produce a different outcome if the policy judgment differs. Second, if this drawdown is driven by earnings variables — semiconductor demand and the durability of AI capex — then flow measures compress volatility without changing direction. Third, without the simultaneous monetary and fiscal easing of 2020, the case where an announcement alone lifted the index 8% will not repeat.
Sources
- Even circuit breakers were no use: KOSPI's second-largest point drop on China shock (Newsis)
- Worst 29% plunge in 36 years: four cards that could revive the KOSPI (Hankyung)
- Stabilization fund may deploy after 14 years (Business Post)
- Institutional improvements and expected effects for resuming short selling (KCMI)
- Blanket short-selling ban starts today: what changes (Newspim)
- FSC Chairman Lee Eok-won on single-stock leveraged ETF volatility (Electronic Times)