What it means
Traditional risk measures such as standard deviation count every deviation from the average, whether it is a pleasant surprise or a painful one. Downside risk measures deliberately discard the good half of the distribution and focus on outcomes below a threshold.
That threshold might be zero, the rate of inflation or a target return the investor must achieve. The distinction matters most when returns are not symmetrical.
An investment that occasionally produces very large gains and rarely small losses can show a high standard deviation and look risky, when in fact its unusual results are almost all favourable. Downside measures separate the two cases cleanly.
Downside deviation is the most widely used calculation. It works exactly like standard deviation except that any period whose return sits above the threshold is treated as zero, so only shortfalls contribute.
That figure feeds the Sortino ratio, which divides excess return by downside deviation rather than by total volatility. Two other measures appear constantly in practice.
Value at risk states the loss that should not be exceeded over a set period at a given confidence level, such as a 5% chance of losing more than $2,000,000 in a month, while maximum drawdown records the worst peak-to-trough fall actually experienced. Downside thinking applies well beyond investment portfolios.
A finance team assessing a new product line, a lender sizing a loan or a treasurer choosing where to hold cash is generally asking how bad the bad case is, not how widely outcomes might vary in either direction.
In practice
Real-world examples.
Example
A pension trustee compares two funds with identical average returns and identical standard deviations. The downside deviation of the first is 3% and of the second is 6%, revealing that the second fund achieved its record through occasional deep losses offset by sharp rebounds.
Example
A bank's treasury reports a one-day value at risk of $1,200,000 at 99% confidence. The board reads this as an expectation that on roughly one trading day in a hundred, losses will exceed that figure, and sets position limits accordingly.
Example
A manufacturer appraising a new production line runs a downside case in which volumes come in 30% below forecast. Because the worst case still covers the loan repayments, the board approves the investment despite a wide range of possible outcomes.
Think of it
“Downside risk measures how bad things could get-the potential for losses, not total volatility.
Formula
Calculation
Downside deviation = the square root of the average of the squared shortfalls below the minimum acceptable return, where a period above the threshold contributes zero.
Take five monthly returns of +4%, -3%, +6%, -5% and +2%, with a minimum acceptable return of 0%. The shortfalls are 0, 3, 0, 5 and 0 percentage points respectively, because only the two negative months fall below the threshold.
Squaring the shortfalls gives 0, 9, 0, 25 and 0, which sum to 34. Dividing by the five periods gives 6.8, and the square root of 6.8 is approximately 2.61%. For comparison, the ordinary standard deviation of the same five returns is about 4.0%, so using total volatility would overstate the loss experience these numbers actually delivered.Case study
Seen in the real world.
Kestrel Harbour Capital is a fictional, illustrative investment manager created for this entry. In a hypothetical review of two of its funds, both had returned an average of 9% a year over six years with the same 12% standard deviation, and the sales team had been presenting them as equally risky.
Recalculating on a downside basis told a different story. The first fund's downside deviation was 5% and its worst peak-to-trough fall 14%, while the second showed a downside deviation of 9% and a maximum drawdown of 31%, since its record combined a handful of very sharp losses with unusually strong recoveries.
The illustrative firm changed how it reported. It added downside deviation, maximum drawdown and the Sortino ratio to every factsheet, and several clients moved between the two funds once they could see how differently the same average return had been earned.
Watch out
Common mistakes.
- Treating standard deviation and downside risk as the same measure. Standard deviation counts upside surprises as risk, which flatters investments with sharp losses and overstates the risk of those with sharp gains.
- Reading value at risk as the worst possible loss. It is a threshold at a stated confidence level, and losses beyond it can be far larger, which is exactly what happens in a crisis.
- Choosing a threshold without thinking. Downside deviation measured against 0% answers a different question from the same measure against a 6% target return, and the two are not comparable.
Questions
People also ask.
Why not just use standard deviation?
Because it assumes losses and gains matter equally, whereas most investors, boards and lenders care far more about shortfalls than about upside surprises.
What is the Sortino ratio?
It is a performance measure that divides return above a target by downside deviation, so an investment is rewarded for producing gains without being penalised for the size of those gains.
Is maximum drawdown a downside risk measure?
Yes, and it is the most intuitive one, because it states the largest fall from a peak that actually occurred and therefore what an investor would have had to sit through.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%