How Do Correlation Coefficients Show Whether Stocks Move Together?

Correlation is the statistical measure that tells you whether two assets move in tandem or diverge when markets shift. Understanding correlation coefficient stocks is essential because it reveals whether adding a new investment to your portfolio truly reduces risk or merely adds the same exposure you already own. When two stocks are perfectly correlated, they rise and fall together; when they're uncorrelated, their movements are independent; when they're negatively correlated, one tends to rise as the other falls. This foundational concept separates investors who accidentally build concentrated portfolios from those who deliberately construct diversified ones.

What Is Correlation, and Why Should You Care?

Correlation measures the relationship between the price movements of two assets on a scale from -1 to +1. A correlation of +1 means perfect positive correlation: both assets move in lockstep, rising and falling together. A correlation of -1 means perfect negative correlation: when one asset rises, the other falls by a proportional amount. A correlation of 0 means no relationship: the movements of one asset tell you nothing about the movements of the other. In real portfolios, most asset pairs fall somewhere in the middle—perhaps a correlation of 0.6 with stocks and bonds, or 0.35 with technology and utilities.

Quick definition: Correlation coefficient is a statistical measure (ranging from -1 to +1) that quantifies how two assets' price movements relate to each other; +1 means they move identically, 0 means no relationship, and -1 means they move in opposite directions.

The reason correlation matters is simple: if every investment in your portfolio has high positive correlation, you're not diversifying at all. You're just making bigger bets on the same risk factor. If 80% of your holdings are tech stocks, utilities, and financials—all with correlations above 0.75—a sector downturn crushes you uniformly. Conversely, holding assets with low or negative correlations means when one part of your portfolio declines, other parts may hold steady or rise, dampening overall volatility.

Key Takeaways

• Correlation ranges from -1 to +1, quantifying how two asset returns move relative to each other, with +1 indicating perfect lockstep movement and -1 indicating opposite movement.
• Correlation of 0.5 to 0.7 is typical between stocks in the same market; correlations below 0.4 provide meaningful diversification benefits.
• Correlation is not static; it rises during crises when most assets sell off together, reducing the diversification benefit when you need it most.
• Negative or low-correlation assets (like government bonds during equity declines) are the most valuable for portfolio construction because they reduce total portfolio volatility.
• Correlation is not causation; two assets can move together without one causing the other, and correlation can change as market structure or economic conditions shift.

The Correlation Coefficient Formula: What It Actually Means

The correlation coefficient between two assets is calculated as the covariance between them divided by the product of their standard deviations:

Correlation = Covariance(Asset A, Asset B) / (Std Dev A × Std Dev B)

This formula normalizes the covariance (which measures how much two variables move together) by each asset's volatility. Think of it this way: if Stock A rises 5% whenever Stock B rises 5%, and falls 3% whenever Stock B falls 3%, they're highly correlated—maybe 0.85 or 0.90. But if Stock A sometimes rises when Stock B falls, and the pattern is inconsistent, the correlation coefficient is lower—perhaps 0.20 or 0.35.

The calculation uses historical price data. If you examine daily closing prices for Apple and Microsoft over the past two years, compute the daily returns for each, then apply the formula above, you'll get a single number representing their historical correlation. This is what investors mean when they cite "the correlation between tech stocks is 0.78"—they're usually referencing a defined historical period, often the past year or five years.

Why Correlation Changes Over Time

One of the most important insights in portfolio management is that correlation is not a constant. During normal market conditions, a stock and a bond might have a correlation of -0.15 or -0.20, meaning they often move in opposite directions. But during a financial crisis—2008, 2020, 2022—that correlation can swing to 0.05 or even positive territory. This happens because during crises, fear dominates market behavior, and investors sell across all asset classes simultaneously, searching for cash. The diversification benefit you counted on evaporates precisely when you need it.

For example, consider the correlation between the S&P 500 and U.S. Treasury bonds. In 2021, when inflation was muted and the Federal Reserve held rates near zero, stocks and bonds had a mild negative correlation of around -0.10 to -0.20. When markets were rising steadily, holding bonds didn't hurt returns much. But in 2022, as inflation surged and the Fed tightened policy, both stocks and bonds fell together, and their correlation spiked toward zero or even positive. Investors who thought they were "balanced" discovered they weren't diversified at all.

Positive, Negative, and Zero Correlation: Real Examples

Positive correlation occurs when two assets tend to rise and fall together. Technology stocks and the Nasdaq 100 index have high positive correlation (often 0.75+) because the index is dominated by tech holdings. Copper prices and industrials stocks have moderate positive correlation (0.50-0.65) because both respond to economic growth. When economic data is strong, both tend to rise; when it weakens, both tend to fall.

Negative correlation occurs when one asset tends to rise as the other falls. U.S. Treasury bonds and equities often have negative correlation during calm markets because investors flee stocks for safety when uncertainty rises, driving bond prices up. Gold and the U.S. dollar sometimes have negative correlation because a stronger dollar makes gold more expensive for foreign buyers, depressing gold prices, while investors may sell stocks and buy dollars for safety. In a portfolio, negative correlation is the holy grail because it means one position hedges another automatically.

Zero correlation occurs when two assets' movements are completely unrelated. The historical correlation between wheat prices and technology stock returns is near zero; knowing that wheat is up tells you nothing about whether Nvidia will rise or fall. Small real-estate investment trusts (REITs) in different geographic regions often have near-zero correlation with foreign currencies, because REIT returns depend on local rental rates and property values, not exchange rates.

How to Interpret Correlation in Practice

When reading a correlation coefficient, use these rough benchmarks:

• 0.8 to 1.0: Very high correlation; assets move almost in lockstep; little diversification benefit.
• 0.6 to 0.8: High correlation; assets move together but not identically; modest diversification benefit.
• 0.4 to 0.6: Moderate correlation; meaningful diversification benefit; assets decouple in many scenarios.
• 0.2 to 0.4: Low correlation; strong diversification benefit; assets frequently move independently.
• 0 to 0.2: Very low correlation; excellent diversification; movements are mostly unrelated.
• -0.2 to 0: Near-zero or slightly negative; strong diversification; occasional opposite movement.
• Less than -0.2: Negative correlation; powerful diversification; tends to rise when portfolio falls.

A classic example: the correlation between U.S. equity index funds (like VOO or IVV) and intermediate-term bond funds (like BND or LQD) typically ranges from -0.15 to 0.10 depending on the period. This means that in roughly half of quarterly periods, when stocks are down, bonds are up. That's why a 60/40 stock-bond portfolio is more stable than 100% stocks—the bond portion provides a cushion during equity downturns.

Correlation vs. Volatility: Don't Confuse Them

A common mistake is conflating high correlation with high volatility. Two assets can have low correlation but high volatility individually, or high correlation but low volatility. A speculative penny stock and a stable dividend-paying utility might both have high volatility relative to the market, but if one is a distressed technology play and the other is a regulated monopoly, they could have low correlation.

Conversely, two low-volatility assets can have very high correlation. For example, two money-market funds tracking the same benchmark might have near-zero individual volatility but perfect correlation (both yield essentially 5%, whether rates go up or down, at slightly different spreads).

The Diversification Benefit of Low Correlation

When you hold two uncorrelated or negatively correlated assets, your portfolio's total volatility is lower than the average of the two assets' individual volatilities. This is the mathematics of diversification. If Asset A has 15% volatility and Asset B has 20% volatility, and they're uncorrelated, a 50/50 portfolio of both will have roughly 12% volatility—lower than either component. This effect is powerful.

As correlation increases, this benefit shrinks. If the same two assets become 0.8 correlated, the 50/50 portfolio might have 17% volatility—closer to the average. And if correlation reaches 1.0, portfolio volatility is simply the average, negating any diversification benefit.

Real-World Examples

Consider a hypothetical two-asset portfolio: Case 1 holds 50% SPY (S&P 500 ETF) and 50% AGG (Aggregate Bond ETF). Historically, their correlation is around 0.05 to 0.10. SPY's annual volatility is roughly 15-18%; AGG's is 3-5%. The blended portfolio volatility is often in the range of 7-9%, well below the simple average. This is diversification working.

Case 2 holds 50% QQQ (Nasdaq 100 ETF, dominated by mega-cap tech) and 50% VGT (Vanguard Technology ETF). Both track the same sector with slightly different composition. Their correlation is 0.92. Both have volatility near 22-25%. The blended portfolio volatility is roughly 23%, nearly the average—almost no benefit from the split. This is false diversification.

Case 3 holds 50% VTI (total U.S. stock market) and 50% VXUS (total international stocks). Their correlation is typically 0.75-0.80. VTI volatility is 15%; VXUS volatility is 17%. The blended portfolio has roughly 13-14% volatility, modestly lower than the average—real but not dramatic diversification benefit.

How Correlation Breaks Down in Crises

The most treacherous aspect of correlation is that it often rises dramatically during market stress. In March 2020, when Covid-19 triggered the fastest bear market in history, many assets that had shown negative or low correlation for years suddenly moved in tandem. Stock-bond correlation flipped from negative to near-zero as both fell. Commodity correlations converged. The diversification that worked in 2019 failed in 2020.

This phenomenon is called "correlation breakdown" or "crisis correlation," and it's why relying solely on historical correlation for portfolio construction is inadequate. Assets that hedge you well in normal times may fail to hedge in crises. This is covered in depth in the chapter on Hidden Correlations That Appear in Crashes.

Key Drivers of Correlation: What Changes It

Correlation between two assets depends on several factors:

• Shared economic sensitivity: Assets affected by the same economic driver (like GDP growth or interest rates) tend to have higher correlation.
• Sector concentration: Stocks in the same sector naturally correlate more highly than stocks in different sectors.
• Market regime: In bull markets, correlations tend to be high (rising tide lifts all boats). In bear markets, correlations rise further (panic sells across all assets).
• Fed policy and interest rates: When the Fed signals tightening, asset correlations often increase as investors recalibrate portfolios uniformly.
• Liquidity conditions: During liquidity crises, correlations spike because forced selling hits everything simultaneously.

Common Mistakes

1. Assuming correlation is permanent: Calculating correlation over the past three years and then never updating it, even though correlation shifts with market regime and economic conditions.

2. Confusing correlation with causation: Believing that because two assets are correlated, one causes movements in the other (they may both respond independently to the same third factor).

3. Using only average correlation in crisis scenarios: Designing a portfolio based on average historical correlation, then being shocked when crisis correlation is much higher.

4. Ignoring correlation tail risk: Focusing on average correlation and ignoring that during extreme market moves, correlations often approach 1.0 even when historical average is lower.

5. Treating all diversifying assets equally: Adding a 0.3 correlation asset to your portfolio provides much more benefit than adding a 0.6 correlation asset, but some investors treat them the same.

FAQ

What's the difference between correlation and covariance?

Covariance measures how two variables move together in absolute terms; correlation normalizes covariance by volatility, making it scale-independent and ranging from -1 to +1. Correlation is easier to interpret and compare across asset pairs with different volatilities.

Can correlation be negative?

Yes. When one asset tends to rise as the other falls, correlation is negative. Government bonds and equities often have negative correlation during calm markets; gold and the U.S. dollar sometimes do as well. Negative correlations are valuable for portfolio hedging.

Why does correlation matter more than individual volatility?

Because portfolio volatility depends on both individual volatility and correlation. Two high-volatility assets with negative correlation can create a low-volatility portfolio, while two low-volatility assets with perfect correlation create a low-volatility portfolio with no diversification benefit.

How often should I recalculate correlation?

Quarterly is reasonable for tactical decisions; annually is standard for portfolio rebalancing. During market regime changes (recession onset, Fed policy shifts), checking correlation more frequently is prudent.

Does high correlation mean I should sell one of the assets?

Not necessarily. If both assets are otherwise sound, high correlation might mean they serve the same portfolio role. Consider whether you need both, or whether one could be replaced by a lower-correlation alternative that offers the same fundamental exposure.

What correlation should I target for diversification?

Aim for an average portfolio correlation below 0.5 among major asset classes. This typically happens naturally when you hold stocks, bonds, and alternatives; correlations are often lower than investors expect if properly diversified across geographies and sectors.

Is correlation the only measure of diversification?

No. Correlation is essential but incomplete. You should also consider sector overlap, geographic overlap, and drawdown correlation (how assets move together during downturns). Reading a Correlation Matrix provides deeper structure.

Related Concepts

• Reading a Correlation Matrix
• Hidden Correlations That Appear in Crashes
• Single-Stock Concentration Risk
• Sector Concentration: Too Much of One Bet
• What Portfolio Beta Really Measures

Summary

Correlation coefficient is the quantitative foundation of diversification, measuring on a -1 to +1 scale how two assets' price movements relate. High positive correlation indicates little diversification benefit; low or negative correlation indicates strong benefit. However, correlation is not constant—it tends to rise during crises precisely when diversification is most needed. Understanding correlation prevents the common portfolio construction error of believing you're diversified when all holdings are actually highly correlated to a single risk factor. The next step is learning to read a full correlation matrix, which shows how entire portfolios of assets interrelate.

Next

→ Reading a Correlation Matrix