Currency Correlation in Forex: How Pairs Move Together

Currency pairs do not move in isolation. Because many pairs share a currency, and because the same economic forces affect several economies at once, some pairs tend to rise and fall together while others move in opposite directions. Understanding correlation helps you avoid accidentally doubling your risk and helps you read the market more clearly.

What correlation measures

Correlation describes how closely two price series move together. It is expressed as a coefficient between −1 and +1:

Correlation describes direction, not size. Two pairs can be strongly correlated while one moves twice as far as the other in pips. It also does not prove that one pair causes the other to move; both are often responding to a shared driver.

Why currency pairs correlate

The most common reason is a shared currency. EUR/USD and GBP/USD both have the US dollar as the quote currency. When the dollar weakens broadly, both pairs tend to rise. EUR/USD and USD/CHF share the dollar too, but on opposite sides: the dollar is the quote currency in one and the base currency in the other. When the dollar strengthens, EUR/USD tends to fall and USD/CHF tends to rise, which historically has produced a negative correlation.

Other drivers include:

• Regional economic links. Economies that trade heavily with each other, such as the euro area and Switzerland, or Australia and New Zealand, often share cycles.
• Commodities. Currencies of commodity exporters can respond to similar price swings in raw materials.
• Risk sentiment. In risk-off periods, currencies seen as safe havens may rise together while higher-yielding currencies fall together.

Correlations change over time

Correlation is a statistical summary of the past, not a fixed property. Coefficients can shift when central banks move in different directions, when a country-specific shock hits one currency, or during market stress. A pair of currencies that moved closely for months can decouple quickly.

That is why analysts look at correlation over more than one window, for example 20 days and 120 days. If the short-window figure diverges sharply from the long-window figure, the relationship may be changing.

How correlation is calculated

The usual measure is the Pearson correlation coefficient. Calculate it on returns (such as daily percentage changes), not on raw price levels. Prices that both trend over time can appear highly correlated even if their day-to-day moves are unrelated. In a spreadsheet, put each pair's daily percentage changes in a column and use the correlation function across the same dates.

Worked example: hidden doubling of risk

Assume an illustrative account of $10,000 with a rule of risking 1% per trade, or $100.

A trader sees two setups and opens both:

• Long EUR/USD, 0.25 standard lots, stop-loss 40 pips. Pip value is $2.50 per pip, so risk = 40 × $2.50 = $100.
• Long GBP/USD, 0.20 standard lots, stop-loss 50 pips. Pip value is $2.00 per pip, so risk = 50 × $2.00 = $100.

Each trade respects the 1% rule. But suppose the 60-day correlation between the pairs is around +0.85. Both positions are effectively bets that the US dollar will weaken. If a strong US data release lifts the dollar, both stops could be hit together, for a combined loss of $200, or 2% of the account, from a single idea.

Now consider a trader who buys EUR/USD and also buys USD/CHF, believing the two trades diversify each other. If those pairs are strongly negatively correlated, the positions largely offset each other. The trader pays two spreads but has little net exposure. This is a form of hedging, whether intended or not.

Practical responses include treating highly correlated trades as one position and splitting the planned risk between them, or choosing only the clearest setup. The portfolio risk lesson builds on this idea.

Using correlation in analysis

Correlation can also help confirm or question a view. If EUR/USD breaks higher but other dollar pairs do not follow, the move may be euro-specific rather than broad dollar weakness. Watching the dollar index alongside individual pairs can clarify which currency is driving the move.

Common mistakes

• Assuming correlations are permanent. Recheck them regularly and on more than one timeframe.
• Calculating on prices instead of returns. This can exaggerate apparent relationships.
• Ignoring correlation when sizing. Multiple "1% trades" on the same theme can add up to a much larger risk.
• Using correlation as a timing signal. It describes relationships, not entry points.
• Confusing correlation with causation. A shared driver usually explains both moves.

Summary

Currency correlation measures how pairs move relative to each other on a scale from −1 to +1. Shared currencies, regional links, commodities and risk sentiment create these relationships, but they change over time and can break down under stress. Calculate correlation on returns, monitor it across different windows, and, most importantly, account for it in your risk management so that several trades do not quietly become one oversized bet.