What it means
Volatility measures how widely returns swing around their average. Asymmetric volatility describes a well-documented pattern: the swings themselves are not symmetric.
In equity markets, a sharp down day tends to be followed by more turbulence than a sharp up day of the same size. Two explanations dominate.
The leverage effect notes that when a company's share price falls, its debt becomes larger relative to its equity, so the stock mechanically becomes riskier and more volatile. The volatility-feedback effect works through investors: bad news makes people demand higher expected returns for holding risk, which pushes prices down further and amplifies the initial move.
The pattern is not just folklore. Academic work, including the NBER study of asymmetric volatility and risk in equity markets by Bekaert and Wu, documents the effect across markets and time.
Statistical models used by risk desks, such as GARCH variants, are built specifically to let negative shocks feed volatility more strongly than positive ones. For practitioners the asymmetry changes how risk is measured.
A model that assumes symmetric responses will understate the risk of holding equities through a sell-off, because each new loss raises tomorrow's expected volatility more than an equivalent gain lowers it. Options markets make this visible: the price of downside protection rises disproportionately, so out-of-the-money puts trade at richer implied volatilities than equivalent calls.
Managers outside trading meet the concept through pension and treasury policies. A cash reserve policy sized on average volatility may prove inadequate in a downturn, precisely when volatility clusters upward.
Because protection is priced off implied volatility that already reflects the pattern, buying puts after a sell-off begins means paying for yesterday's fear. The term should not be confused with skewness in the return distribution itself, which describes shape rather than how tomorrow's risk responds to today's return.
Researchers keep testing whether the effect can be traded profitably, and the honest answer is mixed, because transaction costs and crowded positioning eat much of the edge. Its safer use is defensive: better risk estimates and more honest buffers rather than speculative bets.
In practice
Real-world examples.
Example
A risk manager raises margin buffers after a loss-heavy week, knowing volatility tends to stay elevated after declines. The extra buffer is released only when realised volatility has settled for several weeks. This keeps the firm from being forced to sell positions into a falling market.
Example
An options desk notices put implied volatilities climbing faster than call volatilities as the index falls. The desk reads this as the market repricing downside protection and widens its quotes on puts. Clients who want protection that day pay noticeably more than they would have a week earlier.
Example
A pension committee reviews why its symmetric model understated risk during the last downturn before setting new reserves. The consultant shows that volatility after the first big loss ran well above the model's forecast for weeks. The committee adopts a model with a separate term for negative shocks and raises its liquidity buffer.
Formula
Calculation
A simple GARCH-style response: variance_t = base + a x (return_t-1)^2 + b x indicator(return_t-1 < 0) x (return_t-1)^2. The b term captures asymmetry: a positive b means negative returns add extra variance beyond what a symmetric model predicts.
Example with base = 0.00001, a = 0.05 and b = 0.10. After a +3% day, the squared return is 0.0009, so variance = 0.00001 + 0.05 x 0.0009 = 0.000055, and daily volatility is about 0.74%. After a -3% day the extra term adds 0.10 x 0.0009 = 0.00009, so variance = 0.000145 and daily volatility is about 1.20%. The same size move produces roughly 60% more expected volatility when it is a loss.Case study
Seen in the real world.
This is a fictional example. Aldergate Capital, an invented fund manager, runs a risk model that treats gains and losses symmetrically. During a two-week market slide, realised volatility runs far above the model's estimate, and the desk breaches its risk limits. Afterward the firm refits its model with an asymmetry term. Before the refit the model predicted daily volatility of about 1.0% during the slide while the market delivered nearer 2.0%; in the next sell-off the refitted estimates stay within predicted bounds, and the desk is no longer forced to cut positions at the worst moment.
Watch out
Common mistakes.
- Assuming volatility responds equally to up and down moves, which understates risk precisely when markets are falling. The two statistics answer different questions.
- Confusing asymmetric volatility with return skewness, which describes the shape of returns rather than the risk response to shocks. Clustering is the practical translation.
- Sizing reserves on long-run average volatility, ignoring that volatility clusters upward after losses and stays high for a while. Backtests should include the cost of the hedge.
Questions
People also ask.
Why does volatility rise more after losses?
Falling prices raise financial leverage mechanically, and investors demand higher returns for bearing risk, both of which amplify subsequent swings. Both mechanisms reinforce each other in practice.
Is asymmetric volatility only about stocks?
It is strongest in equities, but similar asymmetries appear in credit and other risky assets, driven by the same leverage and feedback logic. Safe-haven assets can show the reverse pattern.
How do practitioners model it?
With GARCH-family models that include a separate term for negative shocks, plus option-implied measures that capture the downside skew. The models name the asymmetry term explicitly.
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%Related