🔍 Correlation Finder
Find ETFs that don't move together. Lower correlation = better diversification.
What Is ETF Correlation?
Correlation measures how closely two ETFs move together, expressed as a number between -1 and +1. A value of +1 means the two funds rise and fall in perfect lockstep. A value of -1 means they move in exactly opposite directions. A value of 0 means they move completely independently.
Most equity ETFs have correlation above +0.7 simply because they all contain similar large-cap US stocks. The goal of correlation analysis is to find the combinations that genuinely reduce portfolio volatility — typically by mixing asset classes (stocks + bonds), geographies (US + international), or strategies (growth + value).
EigenDex calculates correlation from 3 years of daily price returns by default, which captures recent market conditions while smoothing out short-term noise.
Correlation vs. Overlap: What Is the Difference?
ETF overlap measures shared holdings (do both ETFs own the same stocks?). ETF correlation measures shared price movement (do both ETFs go up and down together?).
High overlap usually causes high correlation, but not always. Two international ETFs can have low overlap in individual stocks yet still have high correlation (0.90+) because they both react to the same global macro forces — interest rate changes, risk-off sentiment, dollar strength.
For building resilient portfolios, correlation is the more important metric. It tells you how much portfolio volatility actually falls when you add a new position.
Common ETF Correlation Examples
| ETF Pair | Correlation | Interpretation |
|---|---|---|
| SPY vs VOO | +0.99 | Identical — both S&P 500 |
| QQQ vs XLK | +0.92 | Very high — both tech-heavy |
| SPY vs EEM | +0.65 | Moderate — US vs. emerging markets |
| VOO vs BND | +0.10 | Low — stocks vs. bonds ✓ |
| SPY vs TLT | −0.20 | Negative — long bonds hedge equities ✓✓ |
How to Use Correlation to Build a Diversified Portfolio
The mathematical benefit of diversification is maximized when you combine assets with low or negative correlation. Harry Markowitz won the Nobel Prize for proving that portfolios of uncorrelated assets deliver better risk-adjusted returns than any individual holding.
In practice, this means:
- Adding bonds (BND, AGG, TLT) to a stock portfolio significantly reduces volatility without sacrificing much return.
- Adding international ETFs (VXUS, EFA, EEM) to a US-only portfolio reduces home-country concentration risk.
- Adding real assets (commodity ETFs, REITs) provides inflation protection that stocks and bonds rarely offer together.
- Blending growth (QQQ) with value (VTV) or dividend (SCHD) ETFs reduces drawdowns during tech selloffs.
A target average correlation below 0.70 across all your ETF pairs is a reasonable goal for a well-diversified portfolio.
Frequently Asked Questions About ETF Correlation
Does low correlation guarantee lower risk?
Lower correlation reduces one type of risk — the risk that all positions fall together. But it does not eliminate individual asset risk. A portfolio of gold, crypto, and emerging market bonds has low correlation between positions but each position carries significant standalone volatility. True risk management requires both low correlation AND reasonable individual position quality.
Does ETF correlation change over time?
Yes — significantly. During market crises, correlations between assets tend to spike toward +1.0 because investors sell everything simultaneously. The 2020 COVID crash briefly correlated stocks, bonds, gold, and REITs. This is called "correlation breakdown" and is why diversification feels less effective exactly when you need it most. Long-run correlation averages are still useful for portfolio construction even knowing this limitation.
Which ETF pairs provide the best diversification?
For US investors, the most reliable diversifiers are long-duration Treasuries (TLT) with correlation around −0.2 to −0.3 against S&P 500 ETFs. International developed-market ETFs (EFA, VEA) add geographic diversification at +0.75 to +0.85 correlation — meaningful but not dramatic. Commodities (DJP, GSG) average about +0.25 to +0.40 versus equities over full cycles, providing the most reliable non-correlated exposure among public market assets.