[next →] [← prev] [prev-tail] [tail] [up]

Volatility Drag

Volatility drag is the gap between the average return a portfolio advertises and the compound return it delivers. To illustrate this, imagine two cars racing on a track. One car speeds up and slows down erratically, while the other maintains a steady pace. Despite both cars having the same average speed, the one with the erratic pace finishes the race later. This is akin to how volatility drag affects your investment returns.

What is Volatility Drag? Volatility drag refers to the reduction in compound returns caused by the volatility of investment returns. When an investment experiences fluctuating returns, the geometric mean (or compound return) tends to be lower than the arithmetic mean (or average return). This discrepancy is due to the mathematical nature of compounding, where losses have a disproportionately larger impact than gains.

Be careful with the word “drag,” because it implies you lost something you could have had. You did not: the arithmetic mean was never an achievable outcome. It is the average of the individual period returns, and no investor earns it — what you earn is the geometric mean, always. The gap is not a leak to be plugged; it is the difference between two ways of averaging, one of which describes reality and one of which does not.

That matters for what you do about it. Reducing volatility improves compound growth only when you can do it without giving up arithmetic return — through diversification across imperfectly correlated assets, or through rebalancing between assets of similar expected return (section “Rebalancing”). Shifting from stocks to bonds also reduces drag, and reduces compound growth along with it, because it cuts the arithmetic mean by more than it saves in variance. Any advice to “minimize volatility” that does not say what you are giving up in return is selling you the wrong half of the trade.

How to Calculate Volatility Drag To calculate volatility drag, you need to understand the difference between the arithmetic mean and the geometric mean of your investment returns. Here’s a step-by-step approach:

Calculate the Arithmetic Mean (AM)

This is simply the average of your returns. If your investment returns are $r_{1}, r_{2}, \dots, r_{n}$ , then:

AM = \frac{r_{1} + r_{2} + \dots + r_{n}}{n}

Calculate the Geometric Mean (GM)

This represents the compound return (also Compound Annual Growth Rate (CAGR)) and is calculated as:

GM = {(\prod_{i = 1}^{n} (1 + r_{i}))}^{\frac{1}{n}} - 1

Determine Volatility Drag

The volatility drag is the difference between the arithmetic mean and the geometric mean:

Volatility Drag = AM - GM

Suppose you have an investment with annual returns of 10%, -10%, and 15%. Let’s calculate the volatility drag:

AM = \frac{10 - 10 + 15}{3} = \frac{15}{3} = 5 %

GM = {((1.10) \times (0.90) \times (1.15))}^{1 ∕ 3} - 1 \approx 4.42 %

Volatility Drag = 5 % - 4.42 % = 0.58 %

This example shows that despite an average return of 5%, the actual compound return is only 4.42% due to volatility drag.

To estimate the compound (geometric) return rate, use the formula:

CAGR \approx AM - \frac{{SD}^{2}}{2}

where AM is the arithmetic mean of returns, and SD is the standard deviation, representing the volatility or risk of these returns. The adjustment term, $\frac{{SD}^{2}}{2}$ , accounts for the volatility drag, which reflects how volatility reduces the geometric mean compared to the arithmetic mean. This formula is derived from the log-normal distribution assumption of returns, where the geometric mean is lower than the arithmetic mean due to the compounding effect of volatility. This approximation is particularly useful for estimating long-term growth rates in volatile markets, highlighting the impact of risk on investment performance.

The operative lesson is to quote and plan on geometric returns. Any projection built on an arithmetic average overstates what you will actually accumulate, by roughly ${SD}^{2} ∕ 2$ per year — which at 20% equity volatility is a full two percentage points annually, compounding.

Two things genuinely reduce the gap, and one widely repeated suggestion does not:

Diversification: Combining imperfectly correlated assets lowers portfolio variance while the portfolio’s arithmetic mean stays the weighted average of the components’. That is the one free lunch in the subject, and it is why it gets called that.
Rebalancing between similar-return assets: The same mechanism seen from another angle; section “Rebalancing” works through why, showing that the bonus is small and conditional in real portfolios.
Not a longer horizon: You will read that a long investment period “mitigates volatility drag as markets stabilize over time.” It does not. The drag is a property of the per-period return distribution, and lengthening the horizon compounds it instead of diluting it — thirty years of a 2% annual gap is a much larger shortfall than five years of it. What genuinely falls with horizon is the standard deviation of the annualized average return, which is a different quantity, and confusing the two is the time diversification fallacy (section “Time Diversification”). A long horizon is an excellent reason to hold equities. It is not a reason to expect the arithmetic mean.

[next →] [← prev] [prev-tail] [front] [up]