ETF correlation
Correlation measures how closely two ETFs' performances move together (from -1 to +1). Low correlation between portfolio assets improves real diversification.
Correlation between two ETFs is a statistical coefficient (ρ) ranging from −1 to +1, quantifying the linear relationship between their performance changes over time. A coefficient of +1 means they rise and fall exactly together. Zero means fully independent movements. −1 means perfectly opposite movements.
Correlation vs Overlap: two different concepts
Overlap measures portfolio overlap: which securities are shared and in what proportion. Correlation measures performance movement similarity over time. Two ETFs on the same index will have both high overlap AND correlation near 1. But two ETFs with zero overlap can still be highly correlated if their markets move together (e.g. a Europe ETF and a US ETF have 0% overlap but ~0.85 correlation).
Typical correlations observed
- MSCI World vs S&P 500: ~0.95-0.97 (near-perfect, high overlap)
- MSCI World vs MSCI Europe: ~0.88
- MSCI World vs MSCI Emerging Markets: ~0.75
- MSCI World vs Government Bonds (long-term): ~0.00 to −0.20
- MSCI World vs Gold: ~0.00 to 0.10
Correlation is not stable over time
Correlations often increase during crises: during severe market crashes (2008, March 2020), most risk assets fall simultaneously, even those usually uncorrelated. This 'correlation goes to 1 in stress' phenomenon means diversification across risky assets disappears precisely when you need it most.
Check your ETF overlap
Enter two tickers to instantly see their overlap and avoid duplicates in your portfolio.
Open simulator