Second-Order Greeks - Vanna, Volga, and Charm
By EC Assets Research Team, Derivatives Research · Published · Updated
Second-Order Greeks: Second-order Greeks measure how the first-order Greeks themselves change. Vanna is delta's sensitivity to volatility (and vega's to spot), Volga is vega's sensitivity to volatility, and Charm is delta's decay with time. They are essential risk for large, long-dated, or skew-heavy options books, where a first-order hedge quietly un-hedges itself.
What Second-Order Greeks Measure
The first-order Greeks, delta, vega, theta and rho, each measure how an option's price responds to one input. They are not constant. Delta changes when the underlying moves, which is gamma, but it also changes when volatility moves and simply as time passes. Second-order Greeks measure those changes: how fast a hedge stops being a hedge.
For one vanilla option held for a day they are a refinement. For a large book, long-dated positions, skew trades or anything path-dependent they are core risk. Three matter most on a volatility desk: vanna, volga and charm.
How It Works
| Greek | Definition | What it answers |
|---|---|---|
| Vanna | ∂Δ/∂σ = ∂ν/∂S | How much does my delta change if implied volatility moves? |
| Volga | ∂ν/∂σ | How much does my vega change if implied volatility moves? |
| Charm | ∂Δ/∂t | How much does my delta change by tomorrow if nothing moves? |
All three are close to zero at the money and largest for options some distance from the strike.
Under Black-Scholes without dividends:
$$\text{Vanna} = -\frac{\varphi(d_1),d_2}{\sigma}, \qquad \text{Volga} = \nu,\frac{d_1 d_2}{\sigma}$$
Vanna matters because spot and volatility move together. Equity indices tend to fall when implied volatility rises, so a delta hedge that ignores vanna drifts exactly when markets are stressed. Volga is to volatility what gamma is to spot: it makes out-of-the-money options gain more than vega predicts in a volatility spike. Charm is the slow drift of delta towards 0 or 1 as expiry approaches. Its sign depends on where the strike sits: out-of-the-money calls lose delta with time while in-the-money calls gain it, so a book's net charm depends on how its strikes are distributed around the current price.
Worked Example
A dealer is long 1,000 contracts of a 30-day call struck at 110 on a stock at 100, with implied volatility at 20 percent and a contract size of 100. The call's delta is 0.0512, so on Friday evening the dealer is short 5,117 shares and delta-flat.
| Monday scenario, spot unchanged at 100 | Call delta | Change in book delta, shares |
|---|---|---|
| Three days pass, volatility unchanged (charm) | 0.0423 | -891 |
| No time passes, volatility rises to 25% (vanna) | 0.0978 | +4,667 |
| Three days pass and volatility rises to 25% | 0.0857 | +3,452 |
The dealer who did nothing over the weekend is no longer hedged: charm alone moved the book by a sixth of its hedge, and a volatility jump moves it by almost the whole hedge again.
Volga shows up in the same scenario. The book's vega is 3,012 dollars per volatility point at 20 percent and 4,952 at 25 percent, 64 percent more: the next volatility point is worth considerably more than the last. For comparison, an at-the-money 30-day call has a vega of 0.1143 at both volatility levels and a delta that moves only from 0.5114 to 0.5143. The second-order risk sits in the wings.
When It Applies (and Limitations)
They are model-dependent. The numbers above assume Black-Scholes with a single volatility. On a real surface, implied volatility at each strike moves differently as spot moves, and the effective vanna depends on whether volatility is assumed to be sticky by strike or by moneyness. Two systems can report different vannas for the same book.
They interact. A weekend combines charm and theta; a sell-off combines gamma, vanna and volga. Scenario grids that revalue the book across joint moves in spot, volatility and time are more reliable than adding Greeks one at a time.
Higher orders exist. Speed (the change in gamma with spot), colour (the change in gamma with time) and zomma (the change in gamma with volatility) matter for large books near expiry. Most institutional reporting stops at the second order and covers the rest through scenarios.
Hedging them costs money. Neutralising vanna or volga requires options at other strikes, which carry their own bid-ask spreads and their own second-order exposures. Desks usually manage them within limits rather than to zero.
Why It Matters for Institutional Investors
They decide whether a hedge holds. Market makers and volatility funds re-examine hedges as volatility and time move, not only as spot moves. A risk report without vanna and charm on a large book describes the position as of the close and little after.
They carry the shape of the surface. Vanna and volga are how exposure to skew and smile, and to the volatility of volatility, appears in the Greeks. The vanna-volga method, widely used in currency options, prices exotic options from exactly these two quantities.
They move markets around expiry. Aggregate dealer vanna and charm positions are watched ahead of large index expiries, because the hedging they force as volatility falls and expiry nears can add buying or selling pressure in the underlying.
They matter for structured products. Autocallables and barrier notes leave issuers with large, changing exposures to the wings of the surface. Second-order Greeks are how that exposure is measured before it is hedged in listed markets.
References
- Taleb, N. N. (1997). Dynamic Hedging: Managing Vanilla and Exotic Options. Wiley.
- Castagna, A., & Mercurio, F. (2007). The Vanna-Volga Method for Implied Volatilities. Risk, 20(1), 106-111.
- Bossu, S. (2014). Advanced Equity Derivatives: Volatility and Correlation. Wiley.
- Hull, J. C. (2022). Options, Futures, and Other Derivatives (11th ed.). Pearson.
Frequently asked questions
What is vanna?
The sensitivity of delta to implied volatility, or equivalently of vega to the underlying price. Because equity prices and volatility tend to move in opposite directions, vanna determines how much a delta hedge drifts when volatility jumps in a sell-off.
What is volga?
The sensitivity of vega to implied volatility, also called vomma: the convexity of an option's value in volatility. Out-of-the-money options carry most of it, which is why they gain more than vega predicts in a volatility spike and why they are exposed to the volatility of volatility.
What is charm?
The change in delta as time passes with nothing else moving, also called delta decay. Out-of-the-money options lose delta and in-the-money options gain it as expiry approaches, so a hedge set today is wrong tomorrow even if the market is unchanged. It matters most over weekends and in expiry week.
Why do they matter if delta and vega are already hedged?
Because the hedge is only correct at today's spot, volatility and date. Vanna, volga and charm measure how fast it stops being correct. A book hedged only to first order is exposed to every combined move in price, volatility and time.
How do dealer Greeks affect the market?
When dealers hold large option positions, the hedges forced by vanna and charm translate into buying or selling of the underlying as volatility falls or expiry approaches. Around large index expiries these flows are widely monitored because they can support or pressure prices.
Stay informed
Market commentary, firm news and research from EC Assets - direct to your inbox.