Skip to main content

Why Average Returns Don't Tell the Full Compounding Story?

Investment marketing is filled with numbers: "Our fund has delivered 9.5% average returns over the past ten years." Investors see this figure and assume that $100 invested a decade ago would have grown to approximately $244 after ten years of such returns. But this intuition is often wrong, sometimes dramatically so. The disconnect reveals a critical distinction in how returns are measured—the difference between arithmetic mean and geometric mean, two fundamentally different ways of averaging that produce wildly different results when compounding is involved.

The problem is not that investment firms are being dishonest. They're often calculating the arithmetic mean correctly. The issue is that arithmetic mean, the average most people learned in school, is mathematically inappropriate for describing compound returns. When money compounds, geometric mean—the return rate that actually governs your wealth's growth—tells the truth. Arithmetic mean tells a story that can be misleading by a factor of several percentage points annually, which compounds into enormous differences over decades.

Quick Definition​

The arithmetic mean is the simple average: add all values and divide by the count. The geometric mean is the rate at which wealth actually grows: the return rate that, when compounded, produces the actual final wealth starting from the initial wealth. For volatile investments, geometric mean is always lower than arithmetic mean, sometimes substantially so.

Key Takeaways​

  • Arithmetic mean and geometric mean are different; geometric mean governs actual wealth growth through compounding
  • Volatility (variation in returns) creates a "drag" that makes geometric mean lower than arithmetic mean
  • The drag increases with volatility; a volatile investment with high average returns might underperform a steady investment with lower average returns
  • Financial literature often reports arithmetic mean; you should calculate or find geometric mean to understand true compounding
  • A portfolio's "average return" cited in marketing materials may not match the rate at which your actual money compounded

Arithmetic Mean: The Basic Average​

The arithmetic mean is what most people call "average." If a stock returned 50% in year 1 and lost 30% in year 2, the arithmetic mean is (50 + (-30)) / 2 = 10% per year.

This figure sounds reasonable. But it's mathematically misleading for returns. Here's why: if you actually invested $100 in this stock, what happened?

Year 1: $100 × 1.50 = $150 Year 2: $150 × 0.70 = $105

You earned a 10% arithmetic mean return but only a 5% geometric mean return. Your actual wealth grew from $100 to $105 over two years. The true annualized growth rate is approximately 2.47%, not 10%.

The gap is staggering. The arithmetic average says 10%; the actual compounded return is 2.47%. This is not a rounding error or a minor distinction. It's the difference between describing an investment accurately and misleading people about its performance.

Geometric Mean: The Actual Compounding Rate​

The geometric mean, also called the compound annual growth rate (CAGR) when applied to annual returns, is the constant return rate that produces the actual final wealth from the initial wealth over the period studied.

For the example above:

Initial wealth: $100 Final wealth: $105 Number of years: 2

Geometric mean = (105/100)^(1/2) - 1 = 1.0247 - 1 = 0.0247 = 2.47%

This 2.47% rate, if compounded annually, produces exactly the actual outcome: $100 × 1.0247 × 1.0247 = $105.

This is the return that matters to you as an investor. It's the rate at which your actual money grew. No marketing department should ever cite 10% arithmetic mean when 2.47% geometric mean is what actually happened.

Why Volatility Creates the Gap​

The mathematical difference between arithmetic and geometric mean emerges directly from volatility. Here's the key insight: percentage losses hurt more than percentage gains help, when both are the same magnitude.

A 50% gain on $100 produces $150. A 50% loss on $150 produces $75.

Notice that the gains and losses are both 50%, but the dollar impact is asymmetrical. The 50% gain adds $50; the 50% loss removes $75. Starting from $100, you end at $75—a 25% loss overall from a 0% arithmetic average return.

This asymmetry is called "volatility drag" or "rebalancing drag" in technical literature. High volatility, even if the arithmetic mean return is positive, creates a mathematical headwind that reduces the geometric mean.

The relationship between arithmetic mean, geometric mean, and volatility can be approximated by:

Geometric Mean ≈ Arithmetic Mean - (Variance / 2)

(More precisely, Geometric Mean = Arithmetic Mean - (Volatility^2 / 2), though the exact relationship is slightly more complex.)

This formula reveals the cruel mathematics of volatility. If your investment has an arithmetic mean of 10% and a standard deviation of 20%, then:

Geometric Mean ≈ 10% - (0.20^2 / 2) = 10% - 0.02 = 9.8%

A 2% volatility drag might seem minor. But compounded over 30 years, this 0.2% annual difference reshapes wealth:

At 10% geometric mean compounding for 30 years: $100 becomes $1,744 At 9.8% geometric mean compounding for 30 years: $100 becomes $1,583

That 0.2% annual difference creates $161 of difference on a $100 starting investment over three decades.

Real Volatility Produces Dramatic Gaps​

Let's move to a realistic example. Suppose an investor sees this track record:

Year 1: +80% (portfolio grows from $100 to $180) Year 2: -60% (portfolio shrinks from $180 to $72) Year 3: +60% (portfolio grows from $72 to $115.20)

The arithmetic mean return is (80 + (-60) + 60) / 3 = 26.7%.

But the actual portfolio grew from $100 to $115.20 over three years. The geometric mean is:

(115.20 / 100)^(1/3) - 1 = 1.0476 - 1 = 0.0476 = 4.76%

The arithmetic mean claimed 26.7% while the actual compounding rate was 4.76%. A person who read "26.7% average returns" would have massively overestimated the investment's true performance. They would expect $100 to grow to roughly $227 over three years (at 26.7% annually), when it actually grew to only $115.20.

This is not an extreme or unusual pattern. Stocks with significant downside volatility produce this gap routinely. A "strong" average return that comes with wild swings tells a different story when converted to geometric mean.

The S&P 500: A Real-World Case​

Consider the S&P 500 during the period 1980-2025, a span of 45 years. The index delivered highly variable annual returns:

  • A 58% gain in 1995 (the dot-com boom accelerating)
  • A 1% gain in 2005 (relatively flat)
  • A -37% loss in 2008 (financial crisis)
  • A 30% gain in 2013 (post-crisis recovery)
  • Many years of mid-teen percentage returns in between

If you calculated the arithmetic mean of all 45 annual returns, you'd get approximately 11.3%.

But if you calculated what an investor actually earned—the geometric mean—you'd get approximately 10.2%.

That 1.1 percentage point annual difference seems small. Over 45 years, an investment growing at 11.3% annually would become approximately 12x larger. The same investment growing at 10.2% would become approximately 5.5x larger. The volatility of the market created roughly $6.50 of difference per dollar invested, purely through the gap between arithmetic and geometric mean.

Why Financial Marketing Often Cites Arithmetic Mean​

The gap between arithmetic and geometric mean exists because of mathematical reality, not because of fraud. However, it's worth understanding why the gap receives so little attention in marketing.

Arithmetic mean is always equal to or higher than geometric mean (they're only equal when there's no volatility). A mutual fund advertisement saying "10.5% average returns" sounds better than "9.8% geometric mean returns," even if both are calculating the same historical period.

This is not inherently deceptive—the arithmetic mean is a true statement about the average of all yearly returns. But it's incomplete. Professional investors know to look for geometric mean (often labeled as CAGR or "compound annual growth rate") because that's what governs actual wealth accumulation.

An investor shopping for funds would be wise to:

  1. Ignore the "average returns" figure unless it's explicitly labeled CAGR or geometric mean
  2. Look for both the average return and the standard deviation or volatility
  3. Calculate the gap: a 2% difference between arithmetic and geometric mean suggests significant volatility

The Formula for Geometric Mean from Annual Returns​

If you have a series of annual returns r₁, r₂, r₃, ... rₙ, the geometric mean is:

Geometric Mean = ((1 + r₁) × (1 + r₂) × (1 + r₃) × ... × (1 + rₙ))^(1/n) - 1

This formula embeds the compounding directly. You multiply all the growth factors together (which is what compounding does), then take the nth root (which back out the number of periods), then subtract 1 to convert back to a percentage return.

For the earlier example with +50%, -30%:

Geometric Mean = ((1.50) × (0.70))^(1/2) - 1 = (1.05)^0.5 - 1 = 1.0247 - 1 = 0.0247 = 2.47%

This formula is the truth. It's what your actual money did.

Volatility Clustering and Mean Reversion​

One subtlety worth noting: the gap between arithmetic and geometric mean assumes independent returns. In reality, investment returns are not random. They cluster and mean-revert. A bad year is sometimes followed by another bad year (volatility clustering). Or a bad year is sometimes followed by a recovery (mean reversion).

These patterns can change the relationship between arithmetic and geometric mean in specific cases. However, the general principle holds: high volatility creates a gap between arithmetic and geometric mean, and that gap compounds into meaningful differences in final wealth.

the Gap​

When Arithmetic Mean Equals Geometric Mean​

Arithmetic and geometric mean are identical only when there is zero volatility—when returns are constant across all periods. An investment returning exactly 8% every year has an arithmetic mean of 8% and a geometric mean of 8%.

In the real world, this never happens. Markets have volatility. Even stable bond portfolios have small variations in annual returns due to interest rate changes and credit spread movements. Whenever there's variation, arithmetic mean exceeds geometric mean.

The larger the variation (the higher the volatility), the larger the gap.

Common Mistakes in Comparing Geometric and Arithmetic Mean​

Assuming they're interchangeable: They're not. Geometric mean is always correct for describing wealth compounding. Arithmetic mean is appropriate only for describing the average of disconnected events, not for describing sequential investment returns.

Ignoring the volatility component: Two funds might have identical 10% arithmetic mean returns but different volatility. The fund with lower volatility will have a higher geometric mean and will actually produce more wealth.

Confusing mean reversion with volatility drag: Volatility drag (the gap between arithmetic and geometric mean) is a mathematical fact. Mean reversion (the tendency of returns to balance out over long periods) is a market behavior. Both affect geometric returns, but they're distinct phenomena.

Using arithmetic mean to project future wealth: Never project future wealth using arithmetic mean. Always use geometric mean or CAGR. An investment with "9% average returns and 25% volatility" will compound at roughly 8.8%, not 9%.

Overlooking volatility in stable-looking returns: A bond fund showing consistent 4% annual returns might have slightly lower volatility than a stock fund, producing a smaller gap between arithmetic and geometric mean. But even small volatility creates gaps. Always account for it.

FAQ​

What's the practical difference between 9.8% and 10% geometric mean? Over 30 years, the difference compounds into roughly 10% different final wealth ($1,583 vs. $1,744 on a $100 start). It seems small annually but becomes material over time.

Can geometric mean ever exceed arithmetic mean? No. Mathematically, geometric mean is always less than or equal to arithmetic mean. They're equal only when all values (returns) are identical.

Should I always choose lower volatility if arithmetic mean is the same? Often, yes. Lower volatility produces higher geometric mean, meaning more actual wealth. However, risk tolerance and personal circumstances vary. But from a pure compounding perspective, lower volatility at the same arithmetic mean is preferable.

How do I find the geometric mean return for an investment I own? Check your account statement or investment prospectus. Look for "CAGR" (compound annual growth rate) or "geometric mean return." If not listed, use the formula: ending value / beginning value raised to the power of (1 / years) minus 1.

Does the gap between arithmetic and geometric mean matter for bonds? Yes, though bond volatility is typically lower than stock volatility, so the gap is smaller. A stable bond fund with 4% arithmetic mean might have 3.97% geometric mean. Over 30 years, this creates noticeable differences.

What if returns are negative? Geometric mean still applies. A sequence of negative returns compounds negatively. The geometric mean of a losing investment is the constant rate of loss that produces the actual final value.

Summary​

The difference between arithmetic mean and geometric mean returns is not a technical footnote—it's the difference between reported performance and actual wealth compounding. Arithmetic mean, the simple average of all returns, can dramatically overstate investment performance. Geometric mean, the actual rate at which wealth compounds, tells the truth.

Volatility creates the gap. Higher volatility means a larger spread between arithmetic and geometric mean, even if the arithmetic average return is impressive. This gap is not a flaw in the investment; it's a mathematical reality that must be accounted for when projecting wealth or comparing portfolios.

Professional investors always use geometric mean (or CAGR) when discussing compounding returns. When you see "average returns" in marketing materials, be skeptical. Find the geometric mean. Find the volatility. Calculate the actual gap. Then you'll know how fast your money actually compounded and how much the volatility cost you relative to a smooth return path.

Next​

Effective vs Nominal Interest Rate →