KOSPI closes 3.73% lower: how to read an index fall without guessing its cause

1. What changed since the morning?

This morning covered China's LPR and ongoing Asian trading. We now have a completed Korean session and a close-to-close comparison. We do not attribute the KOSPI fall to unchanged LPR simply because both appeared on the same day.

European reports also described weaker semiconductor shares and firmer oil; currencies and crypto had their own clocks. This is context from today's Reuters report, not synchronized prices or proof of one common cause.

2. The index and its companies are different numbers

During the same Korean session, Samsung Electronics fell 2.98%, SK hynix 4.23% and KOSDAQ 3.33% from their previous closes, according to Seoul Economic Daily. Individual shares did not all move exactly like KOSPI.

Our educational context: an index is a basket, and its construction and weights affect its result. Before applying a headline to a particular stock or ETF, check the product's holdings, costs, possible leverage and currency. An index move is not automatically every portfolio's return.

3. What is known about the background, and what is not proven?

Seoul Economic Daily describes net selling by foreign and institutional investors and commentators' concerns about the profitability of AI investment. The named groups' flows are participant data; AI concerns are interpretation reported by the publication, not a measurement of one exclusive cause.

A possible mechanism: changing expectations for future earnings or financing costs can change company valuations. That does not prove one headline caused the whole decline. An alternative scenario remains possible: new data or different expectations may change the next session's response. This is neither a rebound forecast nor a prediction of further losses.

4. Market clocks: a close, current trading and delay

At 15:00 CEST, Korea's regular session has ended, European trading is underway and the regular US open is still ahead: 09:30 ET, or 15:30 CEST, under Nasdaq's schedule. Premarket and futures are not the regular cash session. Crypto follows a separate trading rhythm.

We do not claim a zero-delay Investing.com feed. We use a dated, completed-session result consistent with independent reporting. We do not combine it with an unrelated price from another hour in a chart that implies a common quote time.

A learning exercise with Max: check the denominator first

Our calculation: (5,552.74 / 5,768.01 − 1) × 100 rounds to −3.73%. The 215.27 difference is in index points, not percent or percentage points. Record the reference level, date and session close separately.

In Prodigy, learn by playing with virtual money and simulated trades. This is general education, with no buy or sell recommendations, promises of profit or real financial prizes.

Watch with Max

Video in English. Illustrations are not live market charts.