What are Volatility Skew and Smile?

By EC Assets Research Team · Published 15 May 2026 · Updated 20 May 2026

Volatility Skew and Smile — The empirical pattern by which implied volatility varies across strike prices for options on the same underlying and expiration. Equity indices typically exhibit reverse skew; FX and many commodity markets show symmetric smiles.

What Volatility Skew and Smile Actually Measure

Volatility skew and smile describe the pattern by which implied volatility varies across strike prices for options on the same underlying with the same expiration. Both terms map an observed truth: the constant-volatility assumption of the Black-Scholes model is violated by every traded options market, and the violation has a characteristic shape.

The terminology splits the visual geometry. A smile is the U-shape in which both deep out-of-the-money calls and deep out-of-the-money puts trade at higher implied volatility than at-the-money options. A skew, or reverse skew, is the monotonic shape in which downside puts trade rich and upside calls trade cheap. FX markets typically display smiles. Equity indices since 1987 display pronounced reverse skew. Commodity markets vary by product.

How It Works

Black-Scholes assigns a single σ to all options on a given underlying for a given expiry. The market disagrees, pricing each strike's option at a different premium. Inverting Black-Scholes back to volatility one strike at a time produces a strike-dependent implied volatility, the volatility surface.

Three competing rules describe how the surface evolves with the spot price:

Sticky strike: implied volatility for a given strike level remains constant as spot moves. The at-the-money implied volatility shifts because the at-the-money strike has changed.

Sticky delta (sticky moneyness): implied volatility for a given delta or moneyness remains constant as spot moves. The at-the-money implied volatility is unchanged, but each fixed strike's vol shifts.

Local volatility: the implied volatility surface is consistent with a single state-dependent volatility process for spot, calibrated via the Dupire formula.

Markets transition between these regimes. Equity indices behave roughly sticky-delta during calm periods and sticky-strike (or worse) during shocks, when downside puts repricing fast lifts the entire surface.

The standard quantitative model for skew and smile is the Stochastic Volatility Inspired (SVI) parameterization of total implied variance:

$$w(k) = a + b\left(\rho(k - m) + \sqrt{(k - m)^2 + \sigma^2}\right)$$

where $w = \sigma_{IV}^2 , T$, $k$ is log-moneyness, and the five parameters $a, b, \rho, m, \sigma$ control level, slope, asymmetry, location, and curvature. SVI replaced earlier polynomial fits because it admits closed-form no-arbitrage constraints across strikes.

A simpler practical summary used widely on trading desks is the 25-delta risk reversal:

$$\text{RR}{25\Delta} = \sigma{IV}(25\Delta,\text{call}) - \sigma_{IV}(25\Delta,\text{put})$$

A negative risk reversal indicates put skew (puts richer than calls); a positive risk reversal indicates call skew.

Worked Example

Consider SPX options expiring 30 days from quotation, with spot at 5000.

Indicative SPX 30-day implied volatility surface

Source: Indicative levels typical of SPX index options in a calm regime.

The 4500 put (downside) trades at 24.5 percent implied volatility; the 5500 call (upside) trades at 13.0 percent. The 11.5 vol-point differential is the skew. The surface is monotonically downward sloping in strike, characteristic of equity index options. Note that the smile element is muted; only the deep wings show curvature.

The 25-delta risk reversal, using the closest approximations from the table, is roughly $13.5 - 20.0 = -6.5$ vol points, a large negative number reflecting put richness.

The shape captures the price of crash insurance. Buying the 4500 put at 24.5 vol costs the seller materially more than the same put would in a flat-vol Black-Scholes world. The differential is the market's price for left-tail protection embedded in deep out-of-the-money options.

When and Why Skew Exists

The 1987 anchor

Before October 1987, equity option surfaces were roughly flat across strikes. The crash, which delivered a 22 percent single-day loss in the Dow Jones Industrial Average, revalued the cost of downside protection. By 1990, the equity index skew had crystallized into its modern shape and has remained there for more than three decades. The persistence of the shape, even outside crisis periods, is one of the most stable features of derivatives markets.

Demand-supply imbalance

Institutional portfolios hold systematic long equity exposure. They demand puts as protection. There is no symmetric natural buyer of upside calls; corporate share-issuance hedging and call overwriting both reduce the price of upside vol. The flow imbalance bids puts up and calls down, sustaining the skew shape.

Leverage effect

When equity prices fall, debt-to-equity ratios rise mechanically. Higher leverage implies higher realized volatility going forward. The market prices this anticipated correlation between spot and vol, producing a negative spot-vol correlation that shows up empirically as steeper put skew.

Stochastic volatility models

Models like Heston, SABR, and the Bates jump-diffusion family generate skew and smile structurally. They parameterize a vol-of-vol process and a spot-vol correlation. Negative correlation reproduces equity skew; symmetric correlation reproduces the FX smile. Jump components add curvature to the wings.

FX symmetry

FX markets show smiles rather than skews because the underlying is symmetric: a fall in EUR/USD is a rise in USD/EUR, and both sides of the market have natural hedgers. Wing demand exists for both upside and downside scenarios, lifting both ends of the surface relative to the at-the-money.

Limitations and Pathologies

The surface is constrained by no-arbitrage conditions that any practical model must respect.

Butterfly arbitrage forbids negative implied risk-neutral densities. The second derivative of the call price with respect to strike must be non-negative everywhere. Some naive surface parameterizations violate this and produce arbitrage signals where none exist economically.

Calendar arbitrage forbids decreasing total implied variance across expiries. A two-month option must have at least as much total variance as a one-month option of the same strike.

Wing illiquidity affects deep out-of-the-money strikes, which trade thinly. Quoted implied volatility in the wings can reflect indicative pricing rather than transactable levels. Surfaces calibrated using illiquid wing quotes can produce misleading risk numbers.

Term-structure interaction matters. Skew shape varies across expiries. Short-dated options exhibit sharper skew; long-dated options flatten. A risk-reversal trade across expiries carries exposure to both skew level and skew term-structure changes.

Discrete strikes mean that real surfaces are sampled at limited price points. Interpolating between them requires care to avoid spurious smile or skew structure that reflects fitting artifacts rather than market information.

Why It Matters for Institutional Investors

Risk-reversal positioning

A long-call-short-put structure is short skew. The cost of the put relative to the call moves directly with the skew level. Trading risk reversals expresses a view on skew compression or steepening, separate from spot vol level.

Hedging cost

The price of a protective put depends on its implied volatility, not the at-the-money level. A pension fund hedging a billion-dollar equity book against a 10 percent drawdown is buying 90-percent puts; the relevant cost is the wing vol, not the at-the-money. Skew steepening makes downside protection more expensive even when at-the-money vol is unchanged.

Variance swap replication

The variance swap strike is a weighted integral of option prices across all strikes, with weights of $1/K^2$. Wing options receive disproportionate weight at low strikes. A steep put skew lifts the variance swap strike above the at-the-money implied volatility. The difference, sometimes called the convexity premium, is itself traded.

Stress signaling

Skew shape carries information about market expectations of tail events. A steepening 25-delta put skew signals rising fear without necessarily lifting the spot VIX. Allocators track risk reversals as a forward-looking stress indicator that often moves before more visible volatility benchmarks do.

Greek hedging beyond delta

A book hedged for delta and gamma at the at-the-money strike will still bleed P&L if the skew shape moves. Vanna, the sensitivity of delta to vol, and volga, the sensitivity of vega to vol, capture skew exposure and require monitoring on any complex options book. Sophisticated risk systems decompose vega into ATM-vega, vanna, and volga buckets, each hedgeable separately.

Structured product pricing

Autocallables, accumulators, reverse convertibles, and similar structured notes carry embedded option positions deep in the surface. Their pricing depends on the full skew shape, not a single ATM vol. Issuers running large structured product books are net short skew through the embedded puts and require continuous skew hedging.

References

1. Sinclair, E. (2013). Volatility Trading (2nd ed.). Wiley.
2. Natenberg, S. (2015). Option Volatility and Pricing (2nd ed.). McGraw-Hill.
3. Cboe. VIX White Paper: The Cboe Volatility Index. (https://www.cboe.com/vix)