What is Vega?

By EC Assets Research Team, Volatility & Derivatives Desk · Published 15 May 2026 · Updated 20 May 2026

Vega — The first-order sensitivity of an option's price to a one-percentage-point change in implied volatility. Vega peaks at the money and grows with the square root of time to expiration.

What Vega Actually Measures

Vega is the first-order sensitivity of an option's price to a one-percentage-point change in implied volatility. For a long option position, vega is positive: rising implied volatility lifts the option's mark, falling implied volatility weighs it down. For a short option position, vega is negative, and the relationship inverts. Vega applies equally to calls and puts under put-call parity.

Vega is one of the standard Black-Scholes Greeks, alongside delta, gamma, theta, and rho. Unlike the others, vega does not appear as a parameter in the Black-Scholes partial differential equation; it is a sensitivity to a model parameter ($\sigma$) that the model itself treats as constant. This inconsistency - that the same derivation producing vega also forbids volatility from changing - is well understood and accepted as a practical compromise. Markets quote vega despite the theoretical tension.

How It Works

The closed-form vega for a European call or put under Black-Scholes is:

$$\nu = S \cdot \phi(d_1) \cdot \sqrt{T - t}$$

where $S$ is spot, $T-t$ is time to expiry in years, $\phi$ is the standard normal probability density function, and:

$$d_1 = \frac{\ln(S/K) + (r - q + \sigma^2/2)(T-t)}{\sigma \sqrt{T-t}}$$

Quoted vega is typically scaled to express the change in option price for a one-vol-point move (for example, from 20% to 21%), which means dividing the raw derivative by 100.

Three structural facts follow from the formula:

1. Vega peaks at the money. The function $\phi(d_1)$ is largest when $d_1$ is near zero, which occurs near $S = K$ adjusted for interest, dividends, and time.
2. Vega scales with $\sqrt{T-t}$. Longer-dated options carry more vega per unit notional than short-dated options. A one-year at-the-money option holds roughly $\sqrt{12} \approx 3.46$ times the vega of a one-month at-the-money option of the same strike.
3. Vega is positive for both calls and puts. Put-call parity ensures that a call and a put with the same strike and expiry have identical vega.

The shape of vega across strikes (at fixed expiry) is bell-shaped, mirroring the normal density. The shape across time (at fixed strike) rises with $\sqrt{T-t}$.

Worked Example

Consider a European SPX call with the following inputs:

• Spot $S = 5000$
• Strike $K = 5000$
• Time to expiry $T - t = 30/365 \approx 0.0822$ years
• Risk-free rate $r = 5%$
• Dividend yield $q = 1.5%$
• Implied volatility $\sigma = 15%$

Compute $d_1$:

$$d_1 = \frac{\ln(1) + (0.05 - 0.015 + 0.5 \cdot 0.15^2)(0.0822)}{0.15 \cdot \sqrt{0.0822}}$$

The numerator equals $0.04625 \cdot 0.0822 \approx 0.0038$. The denominator equals $0.15 \cdot 0.2867 \approx 0.0430$. So $d_1 \approx 0.088$, and $\phi(0.088) \approx 0.397$.

Vega before scaling:

$$\nu = 5000 \cdot 0.397 \cdot 0.2867 \approx 569$$

Scaled per one-vol-point move: $\nu / 100 \approx 5.69$ per unit of underlying.

If the contract multiplier is 100 (SPX standard), one contract carries roughly USD 569 of vega exposure. A one-vol-point increase in implied volatility (from 15 to 16) raises the option's value by approximately USD 569.

Source: Black-Scholes computations at inputs shown; small rounding applies.

When It Applies (and Limitations)

Local versus global vega

Black-Scholes vega is computed at the current vol level. If implied volatility moves significantly, the vega itself changes (this second-order term is called volga, the vega of vega). For large vol shocks, a position's actual profit and loss deviates from $\nu \cdot \Delta\sigma$ by terms involving volga and vanna.

Surface vega

Real options books carry positions across many strikes and expiries. Aggregate vega assumes parallel shifts in the volatility surface, but the surface rarely shifts in parallel. Sophisticated risk systems decompose vega into ATM-vega, skew-vega (vanna), and curvature-vega (volga), each hedgeable separately.

Term structure of vega

A book may have a flat aggregate vega yet enormous exposure to twists in the term structure. Long short-dated vega combined with short long-dated vega is zero aggregate vega but carries significant P&L if the front and back of the curve move differently. Term-structure decomposition (vega-by-bucket) is standard practice on professional options desks.

Vega interacts with other Greeks

A delta-hedged option position has zero delta but nonzero vega, gamma, and theta. The relationship between these Greeks is structural: a position earning large theta typically pays for it with negative gamma or short vega. Vega cannot be analyzed in isolation; the trade-off across Greeks defines the strategy.

Non-Black-Scholes models

Local vol, stochastic vol (Heston, SABR), and jump-diffusion models produce different vegas. Black-Scholes vega is a market-standard quote, not a deep model statement. Practitioners reconcile by using Black-Scholes-implied vega for quoting and risk reporting, and using model-specific Greeks for hedging when the model better captures market dynamics.

Time decay of vega

As an option approaches expiration, its vega shrinks toward zero for both deep in-the-money and deep out-of-the-money strikes, while remaining elevated at the money. The fast decay of short-dated vega is a structural feature of any options book and forces continuous rebalancing on positions held through expiry.

Why It Matters for Institutional Investors

Risk reporting

Vega is the single most-watched options Greek after delta in most risk reports. Net vega expressed in dollars per vol point gives an immediate read of how an options book will move on a one-point implied vol change, independent of any spot move.

Volatility positioning

Funds that explicitly trade volatility (variance arbitrage, dispersion, relative-value vol) maintain target vega exposures by strike, expiry, and underlying. Vega budgets are the primary risk-allocation tool, often expressed as a vega limit per underlying or per maturity bucket.

Hedging structured products

Issuers of structured products (autocallables, accumulators, reverse convertibles, structured notes) carry embedded option positions whose vega evolves with spot, vol, and time. Dynamic hedging programs trade listed options to neutralize the vega of the structured book, generating much of the listed-options flow in equity markets.

Insurance and pension applications

Variable annuity providers and pension plans with embedded option-like guarantees carry implicit vega. Rising implied volatility increases the cost of the guarantee. Vega-hedging programs, often overlaying long-dated index options, are standard practice for managing this exposure on multi-billion-dollar liability portfolios.

Performance attribution

In any options strategy, decomposing profit and loss into delta, gamma, vega, theta, and residual is the diagnostic step. A surprise in P&L attributed to vega tells the manager that implied volatility moved unexpectedly. The same surprise attributed to gamma tells them spot moved more than implied predicted. The distinction drives subsequent positioning and is impossible without rigorous Greek attribution.

Capital allocation

Regulators and internal risk committees use vega exposure as one input to capital allocation for options-trading entities. The combination of vega, vanna, and volga limits forms the practical capital framework for any institution running a non-trivial options book.

References

1. Hull, J. C. (2022). Options, Futures, and Other Derivatives (11th ed.). Pearson.
2. Natenberg, S. (2015). Option Volatility and Pricing (2nd ed.). McGraw-Hill.
3. Chicago Board Options Exchange (CBOE). Options education materials. (https://www.cboe.com/education)