FX and investor flows
Does a weaker won mean foreign investors will sell? Reading FX and equity flows
'Foreign investors sold because the won weakened' is a simple explanation, but two numbers cannot establish the reason for a sale. A more useful starting point is to calculate the stock's won price, the won's dollar value and foreign investors' actual transactions separately before connecting them. Every return and flow figure in this article is fictional.
TapeFlowPublished / updated: 7 min read
What this guide answers
- A higher USD/KRW rate means more won are needed to buy one dollar.
- An unhedged dollar price return uses the ratio between the stock-price change and the exchange-rate change.
- Foreign net buying measures transactions, not cross-border funding or the motives behind trades.
1. What do prices, exchange rates and net buying each tell us?
If USD/KRW rises from 1,300 to 1,430, buying one dollar requires 10% more won. The won has weakened against the dollar. Check the quotation direction—won per dollar—before interpreting an upward FX chart as you would an upward share-price chart.
KOSPI reports changes in won-denominated share prices. An investor measuring performance in dollars also considers the dollar value of the won received from a sale. The result in a local investor's won account can differ from an overseas investor's dollar valuation. However, not every foreign investor uses dollars as a base currency or leaves currency exposure unhedged.
Foreign net buying is foreign investors' purchase value less their sale value in the selected market and period. Observing a weaker won and net selling on the same day is different from proving that FX caused the selling. Both may respond to common news, while hedging or index-tracking activity may involve information absent from this aggregate.
Sources: Bank of Korea — Exchange-rate quotations and meaning · KRX Data System — Trading by investor
2. Worked example: an 8% share-price gain becomes a dollar loss
Suppose a fictional won price index rises from 100 to 108 while USD/KRW rises from 1,300 to 1,430 over the same period. The stock return is +8% and the exchange-rate increase is +10%. The unhedged dollar price return is 1.08 ÷ 1.10 − 1, or approximately −1.818%. Subtracting 10% from 8% gives −2%, which is only an approximation.
In amounts, an initial holding worth KRW 1,300,000 equals USD 1,000. After an 8% stock gain, it is worth KRW 1,404,000. Translating that amount at KRW 1,430 per dollar produces about USD 981.82. The won valuation rose while the dollar valuation fell. This compares values at two dates, whether or not an actual currency conversion occurred.
The example excludes dividends, taxes, fees, conversion spreads and hedging gains or losses. Reconstructing an actual account return would also require those items and the timing of deposits and withdrawals. The calculation explains a currency valuation effect; it is not evidence that a particular foreign investor earned this return.
Dollar price return = (ending price ÷ starting price) × (starting USD/KRW ÷ ending USD/KRW) − 1 = (108 ÷ 100) × (1,300 ÷ 1,430) − 1 ≈ −1.818%
| Item | Start | End |
|---|---|---|
| Won price index | 100 | 108 |
| USD/KRW | KRW 1,300 | KRW 1,430 |
| Won valuation | KRW 1,300,000 | KRW 1,404,000 |
| Dollar valuation | USD 1,000 | About USD 981.82 |
3. Three consecutive buying days can still leave a negative weekly total
Imagine five sessions with foreign net buying of −30, −20, +10, +10 and +20 billion won. The final three sessions are all net buying, but the five-session total is −10 billion won. 'The last three sessions shifted to net buying' and 'the full five-session window remained net selling' are both accurate.
Record the daily direction, consecutive-day count and cumulative amount separately. Five sessions are a convenient weekly observation window, not a validated trading threshold where a positive sum predicts a gain. If one large session dominates the total, recording its date conveys more than simply calling the flow 'improving.' The indicator's bn KRW unit means one billion won, equivalent to 10 Korean eok won; +20 bn KRW equals +200 eok won.
Net buying is the difference between a group's purchases and sales. Each matched trade has a counterparty, so foreign net buying is not total new money entering the market. It does not identify whether purchases were funded from an existing won balance, another asset sale or a transfer from abroad.
| Session | Daily net buying | Cumulative |
|---|---|---|
| Day 1 | −30 | −30 |
| Day 2 | −20 | −50 |
| Day 3 | +10 | −40 |
| Day 4 | +10 | −30 |
| Day 5 | +20 | −10 |
4. Check FX, then transactions, then the price response
Write down comparison dates and calculate the USD/KRW change from the corresponding observations. In KOSPI foreign net buying, the latest value and recent-observation table show daily amounts, while the chart and its tooltips show accumulation beginning with the first loaded observation. Do not assume that its first point is zero or that the accumulation automatically resets to your preferred start date.
For a chosen period, sum the daily original values. Alternatively, subtract the cumulative value on the observation immediately before the start date from the end-date cumulative value. If required dates are unavailable in the displayed table or chart, query the period in the KRX source below. Add KOSPI's price move to describe three observations without claiming a chain of cause and effect.
When the readings do not fit one simple story, the mismatch identifies the next question. Foreign net buying during won weakness shows that FX alone cannot explain the transactions. Net buying alongside an index decline calls for checking whether the purchased stocks were the same ones driving the index and whether trading was concentrated.
Finally, use Market breadth to see whether many stocks shared the price movement, then Korea Sector Flow to locate contributors in sectors you follow. It selects stocks with foreign and institutional activity in the same direction and estimates value by multiplying net share counts by closing prices. These estimates are neither actual execution totals nor total KOSPI foreign net buying. Use them to narrow the companies whose earnings and disclosures deserve a closer read.
5. Match observation times and market coverage
Do not assume that an equity close, an FX reference value and an investor-flow aggregate represent the same moment. Check the exchange-rate convention and stock-market trading date, then use common observation dates where possible. Holidays or missing values can mean that the latest five observations are not the same five sessions.
KOSPI foreign net buying does not cover KOSDAQ or equity-index futures. An investor might buy cash equities while adjusting risk through derivatives, but cash-equity net buying alone cannot establish whether this occurred. For a KOSDAQ holding, treat KOSPI flows as background and add information about the relevant market and company.
Avoid translating won weakness directly into higher profits for every exporter. Revenue currencies, input costs, foreign-currency debt, hedging and settlement timing can change the outcome. An FX chart can suggest questions for a company's annual report; it cannot by itself quantify that company's operating-profit change.
6. Keep observations separate from explanations in your weekly note
Write the first sentence using observations only: 'Period __ to __ / USD/KRW __% / KOSPI __% / cumulative foreign flow __ / recent daily pattern __.' Use the second sentence for a possible explanation and the third for the additional evidence needed to test it.
For example: 'Foreign investors were net buyers despite a weaker won. Activity in a particular sector may have contributed. Check stock-level net buying and company disclosures.' Keep the second sentence conditional until it is verified. Next week, check whether new evidence supports it instead of creating a fresh retrospective explanation each time.
The point is to make your own evidence visible, rather than to copy foreign investors' trades. Separate what you have actually established about currency, market transactions and company earnings. That distinction turns a familiar news headline into a useful research question.
Try it: KRW and USD returns
The starting values are fictional examples, not current market data. Use matching periods and coverage for your own inputs.
- Index return in KRW
- +8%
- (Ending index ÷ starting index − 1) × 100
- Change in KRW per USD
- +10%
- (Ending FX ÷ starting FX − 1) × 100
- Index return converted to USD
- -1.82%
- [(Ending index ÷ starting index) ÷ (ending FX ÷ starting FX) − 1] × 100
The USD result changes the reporting currency of the same index. A higher KRW-per-USD rate means more won are needed to buy one dollar. The exact calculation divides growth factors; it does not simply subtract returns. It excludes dividends, taxes, fees and hedging, and does not establish foreign investors’ realized returns or trading direction.
Inputs are calculated only on this page and are not sent or saved.
A record to keep when checking real data
Write a short note for each item. An unanswered field is a prompt to investigate before drawing a conclusion.
- 1
Quotation direction
Did you correctly interpret a USD/KRW increase as won weakness?
- 2
Return assumptions
Did you distinguish price from total return and specify hedging and cost assumptions?
- 3
Window and coverage
Did you separate consecutive days from cumulative amounts and check the KOSPI cash-equity scope?
- 4
Causation
Did you separate simultaneous observations from an explanation of why they occurred?
Sources and preparation
An educational explainer prepared with AI assistance using official definitions. Numerical examples are fictional demonstrations, not actual returns or validated trading rules. The original sources and calculations are provided for checking.
- Bank of Korea — Exchange-rate quotations and meaning ↗
Supports the currency quotation convention and meaning of an exchange rate. The dollar return is a fictional calculation derived from that definition.
- KRX Data System — Trading by investor ↗
Provides purchase, sale and net-purchase data by investor, market and period. This aggregate does not directly measure investors' currency hedges or cross-border funding.