How Correlated Positions Increase Portfolio Risk

How correlated positions increase risk: why three trades can be one bet, the currency and index overlaps that hide it, and how to size the group rather than the trade.

Correlation risk is the risk that positions you treat as separate move together, so that a single market event hits all of them at once. Three trades each sized to lose $50 are not $150 of risk if they are the same trade in three symbols.

Key Takeaways

• Positions in different instruments can be one exposure. Long EURUSD, long GBPUSD and short USDCHF are three ways of being short the dollar.
• Correlation is not fixed. It tightens under stress, which is exactly when it matters.
• Size the group: the combined risk of correlated positions, not each one alone.
• Diversification that is only nominal offers no protection.

Where Correlation Hides

Currencies. The US dollar is on one side of most pairs. A dollar rally lowers EURUSD, GBPUSD and AUDUSD and raises USDCHF and USDJPY together. A trader long the first three and short the last two has five positions and one view. Commodity currencies — AUD, CAD, NZD — move together on global growth. JPY and CHF move together on risk sentiment.

Indices. US500, US100 and US30 overlap heavily in constituents and move together on almost every day that matters. DE40 and FRA40 track each other and the US close. "Three index positions" is usually one.

Commodities and currencies. WTI and USDCAD are linked through Canadian exports. Gold and the dollar tend to move inversely. Long oil and short USDCAD are the same bet twice.

Shares and their sector. Four large-cap US technology CFDs are one sector position expressed four times, and that sector is most of US100.

Crypto. BTCUSD, ETHUSD, SOLUSD and XRPUSD are highly correlated with one another and, in recent years, with US100.

Worked Example: Three Trades, One Move

$5,000 account. Three positions, each sized by position sizing to risk $50 at a 30-pip stop (0.17 lot each, rounded).

The dollar rallies broadly on a hawkish central bank statement. All three move against the position at once.

The arithmetic is unchanged; what changed is the understanding. The trader believed three independent $50 risks were open. One $150 risk was. Over three positions the difference is tolerable. Over ten it is a drawdown.

(Illustrative. Stops assumed to fill at level.)

Correlation Tightens Under Stress

In calm markets, correlations are loose and nominal diversification works reasonably. In a sell-off, correlations across risk assets move toward one: equities, commodity currencies, crypto and most emerging-market exposures fall together while the dollar, yen and franc rise. The protection diversification seemed to offer disappears at the moment it is needed. See risk-on vs risk-off.

Sizing the Group

The practical response is to size correlated positions as one:

1. Identify the shared driver — the dollar, the sector, risk sentiment.
2. Decide the total risk you are prepared to take on that driver.
3. Divide it across the positions, rather than assigning each its own full allocation.

If the total dollar risk you accept is $100 and you want three dollar-related positions, each gets roughly $33, not $50. Fewer positions on the same theme with the same total risk is usually simpler.

Risks Related to Correlation

• Counting positions instead of exposures.
• Trusting historical correlation as if it were fixed.
• Adding a "hedge" that is correlated with the position it hedges.
• Confusing instruments with diversification. Different symbols, same driver.

Which currency pairs are most correlated?

Dollar pairs with one another, in the direction the dollar sits. EURUSD and GBPUSD tend to move together; EURUSD and USDCHF tend to move inversely. Commodity currencies AUD, NZD and CAD tend to move together.

Is holding EURUSD and GBPUSD diversification?

Only partially. Both are short-dollar positions and share most of their movement. The difference between them is the EUR–GBP relationship, which is a much smaller driver.

Does correlation risk apply to CFDs on shares?

Yes. Shares in the same sector, and most large shares against their index, move together.

How can I check correlation?

By observing how instruments moved together over a period, or by identifying the shared driver logically. The second method is more reliable under stress, because past correlations change.

Can correlation be used deliberately?

Yes — a short in a correlated instrument can offset a long. The point of this guide is to make sure it is deliberate.

Related Terms

Position sizing · Drawdown · Risk management in trading · How leverage increases trading risk · Risk-on vs risk-off