Correlation risk
Family III · Risk
Not to be confused with forex risk, gap risk, overnight risk.
Correlation risk is the danger that two or more positions expected to move independently instead move together, reducing diversification and increasing combined loss potential. It arises when assets that normally have low or negative correlation become positively correlated, often during market stress. The result is that a portfolio behaves as if it holds a single, larger position.
How correlation risk arises
Correlation measures the degree to which two assets move in relation to each other, typically expressed as a coefficient between -1 and +1. A value near +1 means they tend to move in the same direction; near -1 means opposite; near 0 means little linear relationship.
Correlation risk is not the same as market risk. It is the risk that the assumed correlation between positions breaks down. For example, a trader may hold long EUR/USD and long AUD/USD, believing the pairs are only moderately correlated. If a broad US dollar rally occurs, both positions may fall together, producing a larger loss than expected.
Correlation is not stable. It can change with economic conditions, monetary policy, or shifts in risk sentiment. During periods of high volatility, correlations often increase across risk-sensitive assets, a phenomenon sometimes called correlation breakdown.
Worked example
Consider a portfolio with two long positions, each with a notional value of $10,000. Historically, their daily returns have a correlation of +0.2. The trader estimates the portfolio's daily volatility at 1.0% based on that correlation, implying a 1-day 95% Value at Risk (VaR) of approximately $165 (using a 1.645 multiplier).
During a stress event, the correlation rises to +0.9. The portfolio volatility increases to 1.4%, and the 1-day 95% VaR becomes approximately $230.
The VaR increases by about 39% solely because of the change in correlation, even though the individual position sizes and volatilities are unchanged.
Managing correlation risk
Traders and risk managers can reduce correlation risk by:
- Stress-testing portfolios under scenarios where correlations shift toward +1 or -1.
- Diversifying across asset classes, sectors, or strategies that have historically low or negative correlation.
- Monitoring rolling correlations and adjusting position sizes when correlations rise.
- Using hedging instruments that are expected to be negatively correlated, though their effectiveness can also vary.
No method eliminates correlation risk entirely, because historical relationships may not hold in future market conditions.
Often confused with
- forex risk
- Forex risk is the general exposure to adverse currency movements, while correlation risk specifically concerns the relationship between multiple positions; the visible sign is that forex risk exists even with a single currency position, whereas correlation risk requires at least two positions.
- gap risk
- Gap risk is the risk of a price jumping between trading sessions with no trading in between, while correlation risk is about the co-movement of assets; the visible sign is a price chart showing a blank space between the close and the next open, rather than a change in correlation coefficients.
- overnight risk
- Overnight risk is the exposure to adverse price changes while a market is closed, while correlation risk is about the interdependence of positions; the visible sign is that overnight risk applies even to a single position held through a market close, whereas correlation risk only arises with multiple positions.