FinUties 101 · Risk · Beginner
Volatility explained
Volatility measures how far returns typically stray from their average. Standard deviation is the usual yardstick, and scaling it from daily to annual shows why a quiet market can still be risky.
In this lesson
- Calculate a standard deviation for a short series of returns
- Convert daily volatility to an approximate annual figure
- Explain that volatility clusters and does not measure direction
Read first: What risk means in investing
Volatility is the most quoted number in risk. News reports say markets are "volatile" and funds publish a volatility figure. It is worth knowing exactly what it measures, and what it leaves out.
What volatility measures
Volatility is how much returns vary around their average. A market that moves 2% a day is more volatile than one that moves 1%. The direction does not matter: a large rise and a large fall both count. That is a limit. An investor does not mind large rises, so volatility captures a mix of good and bad surprises.
The usual yardstick is the standard deviation: a measure of the typical distance of results from their average.
A worked example
Five yearly returns for a fund: +10%, -5%, +15%, 0%, +5%.
- The average is (10 - 5 + 15 + 0 + 5) / 5 = 5%.
- The distances from the average are +5, -10, +10, -5, 0.
- Square each distance: 25, 100, 100, 25, 0. The sum is 250.
- Divide by one less than the number of results (4) to get 62.5. This is the variance.
- Take the square root: about 7.9%.
So the standard deviation is about 7.9 percentage points. Returns typically sat within roughly 8 points of the 5% average. With a bell-shaped distribution, about two thirds of results fall within one standard deviation of the average, and about 95% within two. Real markets are less tidy, as the tail risk lesson explains.
Scaling to a year
Volatility is quoted per period, and periods are compared by scaling. Under the common assumption that daily moves are independent, volatility grows with the square root of time. There are about 252 trading days in a year, and the square root of 252 is about 15.9. So:
- A daily volatility of 1% is roughly 16% a year.
- A daily volatility of 2% is roughly 32% a year.
The same logic runs the other way. A yearly figure of 20% means about 1.26% a day (20 divided by 15.9).
Volatility clusters
Volatility is not constant. Large moves tend to follow large moves, and quiet spells follow quiet spells. This is volatility clustering. It is why a calm market lulls people, and why risk measures that look only at the last few months underestimate what a sudden shock does. Black Monday in October 1987 produced a one-day fall of more than 20% in US stocks, in a market whose typical daily move was far smaller (see Black Monday, 1987).
Volatility also tends to rise when prices fall. Falling markets are usually more volatile than rising ones, which is one reason the two are not symmetrical in experience even if the formula treats them alike.
What volatility does not tell you
- It does not show how deep a fall can go. That is drawdown, covered next.
- It does not show how long a fall lasts.
- It assumes you hold through the swings. Someone forced to sell in a low faces more than volatility.
Common mistakes
- Reading low volatility as low risk. A fund can have calm returns for years and then lose heavily.
- Forgetting to scale. Comparing a daily figure with a yearly one is a unit error.
- Expecting volatility to stay at its recent level.
- Treating one standard deviation as a worst case. It describes a typical, not an extreme, deviation.
In practice
Look at volatility alongside the worst fall in the same period. The first tells you how bumpy the ride usually is, the second how bad it got. Analytics on FinUties shows the market state and alerts that include moves in volatility.
Check yourself
Fund A has an annual volatility of 8% and fund B of 20%. What can you say?
Fund B's yearly returns typically stray further from their average, so its results are more spread out in both directions. It does not say B will lose money, or that B is a worse choice.
A stock has a daily volatility of 1.5%. Roughly what is its annual volatility?
About 1.5% times the square root of 252 trading days, which is about 24%.
Calm months have passed. Does low recent volatility mean the next month is safe?
No. Volatility tends to cluster, but it can jump when news arrives. A calm period is a description of the past, not a limit on the future.
Key terms
Sources
- The (Mis)behavior of Markets: A Fractal View of Financial Turbulence. Basic Books (Benoit Mandelbrot, Richard L. Hudson), 2004