Volatility

Before you rely on it

What to check

Read it alongside maximum drawdown.

The common error

Where people go wrong with volatility

Calling a low-volatility investment safe. Strategies with steady returns and rare catastrophic losses score beautifully on volatility right up to the point they do not.

Worked through

Volatility in plain numbers

A fund's monthly returns have a standard deviation of 4%. Annualize by multiplying by the square root of 12, about 3.46, and you get 13.9% a year. If its average return is 8%, then roughly two years out of three you would expect a result between −5.9% and +21.9%, and about one year in twenty outside twice that range, assuming returns behave like a bell curve. A bond fund with 4% annual volatility would rarely surprise you by more than a few points. Volatility is a measure of how widely results scatter, up and down, and it does not say which way the next move will go.

Easily confused

Volatility and its neighbours

Volatility is not risk in the full sense, and it is not the same as beta or drawdown; it measures the dispersion of returns, usually as a standard deviation, and counts upside swings as well as downside ones. Beta measures sensitivity to a market. Drawdown measures the decline from a peak. Implied volatility, taken from option prices, is the market's forecast, while historical volatility is measured from past returns; the VIX index tracks the implied volatility of a broad stock index. Each says something different about uncertainty.

In real decisions

Where volatility turns up

Investors use volatility to compare the bumpiness of funds and to set expectations. Portfolio models use it to build allocations and to compute risk-adjusted measures like the Sharpe ratio. Traders price options from it; advisors use it to check that an allocation matches a client's comfort with swings. Risk teams use it to set position sizes and limits. A calculator turns a series of returns into annualized volatility.

Know the edges

Where volatility stops being reliable

Volatility changes over time and clusters, calm periods followed by turbulent ones. Real returns have fat tails, so extreme moves happen more often than a bell curve suggests; it treats gains and losses alike, while investors mostly dislike losses. Different data frequencies give different estimates. Use it with drawdown and stress tests, and do not treat it as a limit on possible losses.

Quick answers

What people ask about volatility

How do I annualize volatility? Multiply monthly volatility by the square root of 12, or daily volatility by the square root of about 252. This assumes returns are independent.

Is high volatility bad? It means larger swings, and whether that is bad depends on your goals and horizon. Long-term investors can accept more if they can hold through declines.

What does implied volatility mean? It is the volatility the market's option prices imply for the future, as opposed to the historical figure measured from past returns.