Understanding Asset Correlation

Correlation is the invisible force that makes or breaks diversification. Two stocks can move together so closely that owning both feels like owning one; conversely, two assets can move opposite enough that they act as hedges. Correlation quantifies this relationship on a scale from -1 (perfect negative correlation—one always rises when the other falls) to +1 (perfect positive correlation—they always move together) to 0 (no relationship). Most stocks correlate positively but incompletely; the goal of diversification is finding assets with low or negative correlation to reduce portfolio volatility. Mastering correlation is the key to building portfolios that actually work.

Quick definition: Correlation measures the degree to which two assets move together, ranging from -1 (perfect inverse movement) to 0 (independent) to +1 (perfectly synchronized movement).

Key takeaways

• Correlation ranges from -1 to +1; lower correlations mean better diversification.
• Most stocks correlate positively at 0.3–0.7; the key is finding assets with lower correlation.
• Negative correlation (one asset rises when another falls) is the holy grail; few true negative correlations exist.
• Correlation changes over time and spikes during market crises; historical correlation may not predict future behavior.
• Sector diversification works because different sectors have lower correlation; geographic diversification works for similar reasons.
• Bonds and stocks typically correlate at 0.2–0.5; gold and stocks near 0; real estate and stocks around 0.6.

The correlation spectrum

Correlation = +1.0 (Perfect positive): Asset A and Asset B move in lockstep. When A rises 10%, B rises 10%. Owning both is like owning one; no diversification benefit.

Correlation = +0.7 (High positive): A and B move in the same direction most of the time but with some divergence. Both falling 20% in a crash, but at slightly different times or magnitudes. Modest diversification benefit.

Correlation = +0.3 (Moderate positive): A and B move generally upward together but with meaningful independent variation. When A rises, B usually rises but sometimes doesn't. Good diversification.

Correlation = 0 (No correlation): A and B move independently. When A is up, B is equally likely to be up, down, or flat. Excellent diversification.

Correlation = -0.5 (Negative): A and B move in opposite directions much of the time. When A falls, B often rises. Strong diversification benefit.

Correlation = -1.0 (Perfect negative): A and B always move opposite. When A rises 10%, B falls 10%. Ideal for hedging but rare in practice.

Measuring correlation: the formula

Correlation is calculated as:

Correlation = Covariance(A, B) / (σ_A × σ_B)

Where σ_A and σ_B are the standard deviations (volatility) of assets A and B.

In plain terms:

1. Calculate how much A and B move together (covariance).
2. Normalize by each asset's individual volatility.
3. The result ranges from -1 to +1.

You don't need to calculate this manually; financial websites provide correlation matrices for free. But understanding the logic helps you interpret what correlation means.

Real-world correlation examples

Typical equity correlations:

• Microsoft vs. Apple: ~0.65 (both large-cap tech; move together often)
• Apple vs. Coca-Cola: ~0.55 (different sectors; lower correlation)
• Tech sector vs. Energy sector: ~0.45 (very different industries; meaningful diversification)
• Large-cap stocks vs. Small-cap stocks: ~0.80 (both equities; high correlation)

Cross-asset correlations:

• U.S. stocks vs. International stocks: ~0.85 (both equities; high correlation)
• U.S. stocks vs. Bonds: ~0.25 (different asset classes; low correlation)
• Stocks vs. Gold: ~0.05–0.20 (almost independent; excellent diversification)
• Stocks vs. Real Estate (REITs): ~0.60 (some overlap in economic sensitivity)
• Bonds vs. Commodities: ~-0.10 (weak negative correlation; useful hedge)

Visualizing correlation in motion

The chart shows the principle: lower correlation = better diversification = smoother combined returns.

Why bonds work as diversifiers

Stock-bond correlation typically ranges from 0.2 to 0.5. Why so low?

When the economy enters recession and stocks crash, investors flee to safety. Demand for bonds surges, prices rise, and yields fall. This is a negative correlation event: stocks down, bonds up. Conversely, during high inflation when the Fed hikes rates, bonds fall (higher rates = lower bond prices). But stocks may also fall, increasing positive correlation.

Over long periods, the stock-bond correlation averages positive (both are economic assets), but the periods when they diverge—crises, most importantly—are exactly when you need diversification. Bonds' low correlation during crashes is why a 60/40 stock-bond portfolio is so durable: the 40% bonds cushion the blow when stocks fall.

Geographic diversification and correlation

U.S. and international stocks typically correlate around 0.80–0.85. Why not 1.0? Because:

1. Currency movements: When the U.S. dollar strengthens, U.S.-denominated returns are boosted while international returns (in local currencies) are dampened.

2. Cyclical divergence: The U.S. economy and European economies operate on different cycles. When the U.S. is in recession, Europe might be recovering, or vice versa.

3. Regulatory and political risk: Each country faces unique policy risks. Brexit affected U.K. stocks but not U.S. stocks.

4. Industry exposure: Different countries emphasize different industries. Japan heavy in autos and electronics; Norway heavy in energy. The diversification emerges naturally.

For a U.S. investor, holding 20% international stocks reduces correlation drag and provides hedging against dollar weakness.

The correlation problem during crises

Here's where correlation becomes dangerous: during market crashes, correlation spikes toward 1.0. The 2008 financial crisis offers a stark example:

Normal times (2005–2007):

• Stocks vs. Bonds correlation: ~0.20
• A diversified portfolio with 60% stocks and 40% bonds would experience smooth returns.

Financial crisis (September 2008):

• Stocks vs. Bonds correlation: ~0.80
• Bonds and stocks both fell together. The diversification benefit evaporated.
• A 60/40 portfolio fell ~20%; a 100% stock portfolio fell ~50%. Diversification still helped, but less than expected.

This teaches a critical lesson: diversification works best in normal times and least when you need it most. This is why some investors add alternative hedges (gold, options, real assets) that have more reliably negative correlation during crises.

Sector diversification and correlation

Different sectors have different economic sensitivities:

• Consumer Discretionary: Rises in strong economies, crashes in recessions. High beta. (Corr among sector stocks: ~0.60)
• Consumer Staples: Stable in good and bad times. Low beta. (Corr: ~0.50)
• Energy: Driven by commodity prices and oil geopolitics. (Corr: ~0.40)
• Healthcare: Driven by demographics and drug pipelines, less economic sensitivity. (Corr: ~0.45)
• Technology: High-growth, high-volatility; driven by innovation cycles. (Corr: ~0.70)

A portfolio split evenly across these sectors has lower correlation than a 100% technology portfolio. Not every sector falls when the economy contracts; some (healthcare, staples) hold up.

Negative correlation: the unicorn

True negative correlation is rare in equities. Two stocks almost never reliably move opposite. But across asset classes, negative correlation exists:

1. Bonds vs. Stocks (during crises): Often slightly negative (investors fleeing equities buy bonds).
2. Stocks vs. Gold: Near zero; sometimes slightly negative as gold is viewed as inflation/crisis insurance.
3. Dollar vs. Commodities: Negative (strong dollar weakens commodity prices).
4. Long-dated bonds vs. Inflation: Negative (inflation erodes bond values).

For practical diversification, near-zero correlation (independence) is often sufficient. True negative correlation is nice but not required.

Real-world example: the 2022 crash

In 2022, bonds and stocks both fell significantly:

• S&P 500: -18%
• 60/40 stock-bond portfolio: -16%

The correlation spiked because:

1. The Fed raised rates, hurting both stocks (lower growth valuations) and bonds (lower prices).
2. Inflation hurt growth stocks disproportionately, and growth stocks are more duration-sensitive (similar to bonds).

A portfolio with 60% stocks, 30% bonds, and 10% gold would have fared better because gold held relatively flat. Geographic diversification (20% international instead of 0%) would have also helped slightly. The lesson: diversification across multiple low-correlation assets beats binary stock-bond diversification.

Measuring correlation yourself

Financial websites offer correlation matrices:

1. Yahoo Finance: Search two stocks, click "Compare."
2. Google Sheets: Use CORREL(range1, range2) to calculate historical correlation.
3. Portfolio trackers: Morningstar, Vanguard, and others provide correlation breakdowns.

Look for:

• Stocks with 0.4–0.7 correlation: Good diversification.
• Stocks with >0.8 correlation: Poor diversification; likely in same sector.
• Assets with <0.3 correlation: Excellent diversification.
• Assets with negative correlation: Holy grail; use liberally.

Common mistakes

1. Confusing correlation with causation: Low correlation doesn't mean one asset causes another; it means they move independently.

2. Assuming historical correlation predicts future correlation: Correlation changes over time. A stock pair with 0.3 correlation in 2020 might have 0.7 in 2024. Use longer historical periods when possible.

3. Ignoring correlation during backtesting: When testing a portfolio strategy, use realistic crisis correlations. Normal-time correlations are optimistic.

4. Believing diversification prevents all losses: Even with low correlation, a portfolio can fall if all assets decline. Diversification reduces volatility; it doesn't prevent drawdowns.

5. Over-relying on negative correlation hedges: Gold and long-term bonds provide negative correlation, but they also drag returns during bull markets. Use them sparingly (5–10% of portfolio).

FAQ

Q: What's an ideal correlation for diversification? A: Less than 0.5 is good; less than 0.3 is excellent. Zero is perfect. Negative is rare but ideal.

Q: Do I need to calculate correlation myself? A: No. Most financial websites provide correlation matrices. Use those tools to check if holdings are truly uncorrelated.

Q: Does low correlation guarantee low portfolio volatility? A: Not if you weight positions equally. A 50/50 portfolio of a high-volatility stock and low-volatility bond, with 0.3 correlation, is less volatile than 100% of the high-volatility stock but still fairly volatile. Position sizing matters.

Q: Can I rely on correlation during a crash? A: No. Correlation spikes during crises. Plan for that by holding some uncorrelated assets (bonds, gold, cash) specifically for crash scenarios.

Q: Why do all stocks seem to fall together in a crash? A: Because systematic risk dominates. During a crisis, the market is pricing in macro dangers; individual company factors matter less. Diversification fails when systematic risk overwhelms idiosyncratic factors.

Q: Should I buy negatively correlated assets even if they drag returns? A: Yes, in small amounts. Gold correlates near zero with stocks and sometimes negatively during crises. A 5% allocation to gold is cheap insurance; the 1–2% return drag is worth the crash protection.

Related concepts

• Covariance: The degree to which two assets move together; a precursor to correlation.
• Beta: A stock's systematic risk; correlation is one component of beta.
• Diversification benefit: The reduction in volatility achieved by combining uncorrelated assets.
• Hedging: Using negatively correlated assets (e.g., gold, puts) to offset portfolio risk.
• Rebalancing: Periodically adjusting positions to maintain target correlations and allocations.

Summary

Correlation is the foundation of intelligent diversification. Assets with low (near 0) or negative correlation provide genuine diversification benefit. Stocks and bonds correlate at 0.2–0.5, making them excellent partners. Stocks and gold correlate near 0, making gold useful as crash insurance. Most individual stocks correlate at 0.4–0.7, so diversifying across sectors and geographies reduces correlation. But remember: correlation spikes during crises, the exact moment you need diversification most. This is why a simple 60/40 stock-bond portfolio is robust—it combines two asset classes with low normal-time correlation and meaningful negative correlation during crises. For sophisticated investors, layering in low-correlation alternatives (real assets, commodities, options strategies) adds protection. For most investors, stocks and bonds suffice.

Next

Explore the dangerous illusion: many investors believe they're diversified when they're not, holding seemingly different assets that move together.