Volatility Skew
The most important shape on an index option chain, and the one Black–Scholes most flatly denies.
Quick answer: The volatility skew is the downward-sloping implied-volatility curve that equity indices print, in which out-of-the-money puts are systematically dearer — in volatility terms — than out-of-the-money calls, so the market charges more to insure a fall than to bet on a rise.
In simple words
A skew is a smile that has been tilted. Instead of a symmetric U with both wings equally bid, the equity-index curve slopes downward from left to right: low strikes (out-of-the-money puts) print high implied volatility, and high strikes (out-of-the-money calls) print low implied volatility. On the site's reference NIFTY chain with spot at 24,000, the 25-delta put near 23,400 prints about 14.5% while the 25-delta call near 24,600 prints about 11.4% — a gap of roughly 3.1 volatility points. The at-the-money strike sits in between at about 12.8%. In plain terms: the market will charge you far more to insure NIFTY against a fall than it will charge you to bet on a rise of the same size. That asymmetry is the skew, and it is the single most important feature of index options.
The skew is why an equity index is not a currency. A currency is a ratio of two economies and can be shocked either way, so it prints a symmetric smile. An equity index has a built-in direction of danger: it falls faster than it rises, everybody who owns stocks wants the same downside protection at the same time, and nobody is naturally short that protection. All of that pushes the price of downside puts up and the price of upside calls down, and the curve tilts. If you only remember one shape from this whole site, remember this one.
The picture
The put wing towers over the call wing
Implied volatility of every strike in one NIFTY expiry, spot at 24,000.
Professional explanation
Three explanations, all partly true, and the weights are disputed
There are three genuinely distinct explanations for the equity-index skew, and honest practitioners disagree about how much weight each deserves. The first is that crashes are real and correlated: index selloffs are faster and more synchronised than rallies, because in a fall every constituent stock drops together as correlations spike toward one, whereas rallies are slower and more idiosyncratic. A fatter, more correlated left tail is worth more, so downside puts are dearer. The second is persistent hedging demand: almost everyone in the market is structurally long equities — pension funds, mutual funds, retail — and they all want the same downside protection at the same time, while almost nobody is a natural seller of it, so the price of that put is bid up and stays bid up regardless of any forecast. The third is the leverage effect: when a firm's equity falls, its debt stays roughly fixed, so its debt-to-equity ratio rises, which mechanically raises the volatility of the equity — falling prices genuinely cause higher volatility, which makes low strikes correctly command higher volatility. All three are partly true. What no one can tell you with confidence is the exact split, and anyone who claims a precise decomposition is overreaching.
How the skew is quoted: the 25-delta risk reversal
Traders do not quote the skew as a slope; they quote it as a risk reversal — the difference between the implied volatility of a fixed-delta out-of-the-money put and the implied volatility of the equal-and-opposite out-of-the-money call. The standard convention is 25-delta: take the put whose delta is −0.25 and the call whose delta is +0.25, and quote the put's volatility minus the call's. On the site's reference NIFTY chain that is roughly 14.5% minus 11.4%, a 25-delta risk reversal of about +3.1 volatility points. A positive number means the put is richer than the call, which is the normal state of an equity index. Delta is used rather than a fixed strike because it keeps the measure comparable as spot moves and as the smile itself changes shape — a 25-delta option is always "equally far out of the money" in a probabilistic sense, which a fixed strike is not.
Why the skew steepens in a selloff
The skew is not constant; it steepens dramatically when the market falls. Two things happen at once. First, demand for downside protection spikes exactly when the market is falling — the people who did not hedge now desperately want to, and they all reach for the same puts in the same expiry, bidding up the low strikes. Second, realised correlation across constituents jumps toward one in a selloff, so the index's left tail genuinely fattens in real time, justifying an even higher price for those puts. The combination means the put wing lifts far faster than the call wing during a decline, and the whole curve rotates steeper. This is why the skew is sometimes read as a fear gauge: a steepening 25-delta risk reversal is the market paying up for crash protection, and it often leads spot lower. It is also why short-skew positions — short the expensive puts, long the cheaper calls — are dangerous: they lose most exactly when the skew steepens, which is exactly when the market is falling and liquidity is leaving.
The skew breaks Black–Scholes delta, and the fix has a name
This is the practical payoff of the whole page, and it is uncomfortable enough that a marketing team would cut it: if the skew exists, a delta computed from a flat-volatility Black–Scholes model is simply wrong, for every strike that is not at the money. The reason is that Black–Scholes delta answers "how does the option's price change when spot moves, holding volatility fixed?" — but when spot moves, the option's own implied volatility moves too, because it slides along the skew. The true hedge ratio must account for that second effect. The corrected delta is called the skew delta, the minimum-variance delta, or the smile-adjusted delta, and it equals the Black–Scholes delta plus vega times the rate at which the option's implied volatility changes as spot moves. For an equity index, where implied volatility tends to rise as spot falls, this correction typically makes the true delta of a call SMALLER than its Black–Scholes delta. A trader who hedges an index book on flat-vol deltas is systematically mis-hedged, and the error is largest precisely on the downside puts that matter most in a crash.
The skew is the front-to-back tilt of the whole surface
The same downward tilt repeated across expiries builds the volatility surface.
Formula
The skew as a curve with a linear tilt
σ(K) = σ_ATM − β · ln(K / S) + c · [ln(K / S)]²
The equity-index skew is a smile (the quadratic c term) plus a downward tilt (the linear −β term). The linear term is what makes the curve lopsided: it lifts low strikes and lowers high strikes, so the put wing towers over the call wing. On the site's reference NIFTY chart the fitted values are σ_ATM ≈ 0.128, β ≈ 0.62 and c ≈ 2.1, which reproduce the 14.5% put, 12.8% at-the-money and 11.4% call quoted on the chart. A pure smile has β = 0; a steepening skew is a rising β.
- σ(K)Implied volatility of the option struck at K, as an annualised decimal (0.145 = 14.5%).
- σ_ATMAt-the-money implied volatility — about 0.128 (12.8%) on the reference NIFTY chain.
- βSkew slope — the strength of the downward tilt. Larger β means a steeper skew and a richer put wing. It rises in selloffs.
- cCurvature (convexity) of the underlying smile, about 2.1 on the reference chart. It controls how fast both wings turn up far from the money.
- KStrike price of the option contract.
- SSpot price of the underlying — 24,000 for NIFTY here.
- lnNatural logarithm; ln(K/S) is log-moneyness, negative for puts below spot and positive for calls above it.
How the skew is quoted, and the delta it breaks
RR₂₅ = σ(25Δ put) − σ(25Δ call); Δ_skew = Δ_BS + ν · (∂σ/∂S)
The 25-delta risk reversal RR₂₅ is the market's standard scalar summary of the skew — about +3.1 volatility points on the reference NIFTY chain. The second identity is the correction the skew forces on your hedge: the true (skew) delta is the Black–Scholes delta plus vega ν times the rate at which the option's implied volatility moves with spot, ∂σ/∂S. Where skew exists, that second term is not zero, so a flat-vol delta is wrong.
How to read and quote a volatility skew
- Take one expiry and plot each strike's midpoint implied volatility against strike or log-moneyness, marking spot. On an equity index the curve should slope downward from left to right.
- Identify the 25-delta put and the 25-delta call — the strikes whose deltas are about −0.25 and +0.25. On the reference NIFTY chain these are near 23,400 and 24,600.
- Compute the 25-delta risk reversal: the put's implied volatility minus the call's. On the reference chain that is about 14.5% minus 11.4%, or +3.1 volatility points. A positive number is the normal equity-index state.
- Compare that risk reversal against its own recent range to judge whether the skew is steep or flat for this underlying — a level is only meaningful relative to its own history.
- Watch the risk reversal for steepening: a widening put-over-call gap means the market is paying up for downside protection, which often accompanies or precedes a fall.
- Before hedging, do not use the flat-vol Black–Scholes delta. Estimate ∂σ/∂S — how each option's implied volatility moves as spot moves — and apply the skew-delta correction, because on the wings the flat-vol delta is materially wrong.
- Re-measure after any large move: the skew steepens in selloffs and flattens in rallies, so a risk reversal is a live number, not a constant.
Practical example
NIFTY worked example
NIFTY is at 24,000. Read three points off the reference chain. The 25-delta put near the 23,400 strike prints an implied volatility of about 14.5%. The at-the-money 24,000 strike prints about 12.8%. The 25-delta call near the 24,600 strike prints about 11.4%. The 25-delta risk reversal is therefore 14.5% − 11.4% = 3.1 volatility points, positive, meaning the put is richer than the call. Now interpret it. In premium terms, the extra 3.1 points of volatility on the put is what you pay for the market's structural fear of a fall. Convert it to intuition: at a 30-day horizon, roughly three volatility points of extra implied volatility on a downside put is a meaningful surcharge — you are being charged noticeably more to insure a 2.5% fall than to bet on a 2.5% rise. And that surcharge is not a forecast that NIFTY will fall; it is the price of insurance against a fall that everyone wants and nobody wants to sell. The skew is a supply-and-demand fact wearing a probability costume.
BANKNIFTY worked example
BANKNIFTY skews harder than NIFTY, and the reason teaches a different lesson. With BANKNIFTY at 52,000, the 25-delta put near 50,700 might print about 17.5% while the 25-delta call near 53,300 prints about 13.8%, a 25-delta risk reversal of roughly 3.7 volatility points — steeper than NIFTY's 3.1. The lesson is about correlation. BANKNIFTY is a concentrated basket of a dozen lenders whose fortunes are tightly linked; in a selloff they fall together, correlations snap to one, and the index's left tail fattens more violently than a broad, diversified index like NIFTY. A steeper skew on BANKNIFTY is therefore not a mispricing to be arbitraged against NIFTY — it is a correct price for a basket that genuinely crashes harder. Comparing the two skews tells you about the difference in tail risk between a concentrated sector and a diversified market, not about a trade.
Lot sizes used above (NIFTY 75, BANKNIFTY 30) are those in force at the time of writing; NSE revises them periodically. Figures exclude brokerage, STT, exchange charges, stamp duty and GST. Examples are teaching scenarios built on round numbers — they are not historical quotes, not backtests and not trade calls.
Advantages & limitations
What it is good for
- It is the market's most direct price of downside fear. The 25-delta risk reversal turns "how worried is the market about a crash?" into a single quotable number in volatility points.
- It is a leading-ish signal. Because the skew steepens as hedgers rush for protection, a widening put-over-call gap often coincides with or slightly leads a fall, giving a read on positioning that raw spot does not.
- It corrects your hedge. Knowing the skew exists is what lets you replace the wrong flat-vol delta with the skew delta, so an index book hedged with skew awareness carries less unintended directional risk.
- It is comparable across underlyings in a meaningful way. A steeper skew on BANKNIFTY than NIFTY is a legible statement about relative tail and correlation risk, not just a difference in level.
- It prices the exact insurance most portfolios need. Since almost every equity portfolio is structurally long and fears a fall, the skew is the price of the one hedge that actually matches the risk investors carry.
Where it breaks down
- Its cause is genuinely disputed. Crashes, hedging demand and the leverage effect all contribute, and because no one can pin down the weights, you cannot cleanly say a given skew is "too steep" without smuggling in an assumption about which explanation dominates.
- It is not constant, so any level you measure is already changing. The skew steepens in selloffs and flattens in rallies, so a risk reversal measured this morning may not describe this afternoon's curve.
- The delta correction it forces depends on an unobservable. The skew delta needs ∂σ/∂S, the rate at which each option's volatility moves with spot, and that rate depends on which regime — sticky strike or sticky delta — the market is in, which you cannot observe directly.
- The wings that define it are illiquid. The 25-delta strikes are reasonably liquid, but the deep skew is quoted on far strikes with wide spreads and stale prints, so the extreme steepness is the least reliable part of the curve.
- It says nothing about timing. A steep skew tells you the market is paying for downside protection, not when — or whether — the fall it is insuring against will actually arrive.
Common mistakes
- Hedging an index option book with flat-volatility Black–Scholes deltas. When skew exists, the true delta of every non-at-the-money strike differs from its flat-vol delta, and the error is largest on the downside puts, so a flat-vol-hedged book carries hidden directional risk exactly where it hurts most.
- Reading a steep skew as a guaranteed crash forecast. The skew is a price for downside insurance driven partly by structural hedging demand; it can stay steep for months while nothing happens, and it is not a timing signal.
- Selling the rich put wing and calling it an edge because the puts are "overpriced". They are richer on average because they insure a real, correlated tail, and the position loses most precisely when the skew steepens into a falling market.
- Quoting the skew as a raw volatility difference between two fixed strikes rather than two fixed deltas. Fixed strikes drift out of the money as spot moves, so a fixed-strike skew is not comparable over time — the 25-delta risk-reversal convention exists to fix exactly this.
- Comparing BANKNIFTY's steeper skew to NIFTY's and treating the gap as an arbitrage. BANKNIFTY is a more concentrated basket that genuinely crashes harder, so its steeper skew is a correct price for higher correlated tail risk, not a dislocation.
- Assuming the skew is symmetric enough to ignore. It is the single most asymmetric feature of an index chain, and ignoring it means systematically undercharging for downside — the opposite of the mistake you can afford to make.
Professional usage
Managing the skew is a large part of what an index-options market-making desk actually does. Customer flow is lopsided — clients overwhelmingly buy downside puts and sell upside calls to fund them — so the desk accumulates a short-skew position almost by default and must manage it consciously. The desk quotes and hedges in terms of a fitted skew, marks its 25-delta risk reversal as a live risk factor alongside vega and gamma, and hedges spot exposure using skew deltas rather than flat-vol deltas, because a flat-vol hedge would leave a systematic directional error that grows as the market falls. A dispersion desk goes further, trading the index skew against the skews of its constituents, because the index skew is steeper than a diversified basket of single-name skews would imply — the difference being the market's price of correlation.
Risk managers read the skew, and specifically the steepness of the 25-delta risk reversal, as a positioning and stress indicator: a rapidly steepening skew signals that large hedgers are paying up for protection, which feeds directly into stress scenarios and margin models. On the sell side, the skew is watched as an early-warning fingerprint, because heavy institutional put buying shows up as a steepening skew in option prices before it shows up as a move in spot. And structured-product desks live and die by the skew, because a capital-protected note or an autocallable is, in risk terms, a package of skew exposure that must be hedged strike by strike along the whole downward-sloping curve.
Key takeaways
- The volatility skew is the downward-sloping implied-volatility curve of equity indices, where out-of-the-money puts print systematically higher volatility than out-of-the-money calls.
- On the reference NIFTY chain the 25-delta put prints about 14.5% and the 25-delta call about 11.4%, a 25-delta risk reversal of roughly 3.1 volatility points — the market's price for fearing a fall more than a rise.
- Three explanations contribute — correlated crashes, structural hedging demand, and the leverage effect — and practitioners genuinely disagree about the weights, so no clean decomposition should be trusted.
- The skew is quoted as a 25-delta risk reversal, uses delta rather than fixed strikes so it stays comparable as spot moves, and steepens sharply in selloffs.
- Where skew exists, a flat-volatility Black–Scholes delta is wrong: the true hedge ratio is the skew delta, Δ_BS + ν·(∂σ/∂S), and ignoring the correction mis-hedges an index book most on the downside strikes that matter.
If you take one shape from this whole site, take the skew. It is the daily, quotable statement that an equity index fears a fall more than a rise — not as a forecast, but as a price, driven by real correlated crashes, by everyone wanting the same put, and by the mechanical leverage effect, in proportions nobody can pin down. It steepens when you are most frightened and it quietly breaks the delta you thought you knew. Learn to read the 25-delta risk reversal as the price of the one hedge your portfolio actually needs, and remember that the skew is expensive for a reason: the tail it insures is real, and it is the tail that arrives when everything else is going wrong at once.
Frequently asked questions
What is the volatility skew in simple terms?
Why do index puts have higher implied volatility than calls?
What is the difference between a skew and a smile?
How is the volatility skew measured?
Why is delta used to measure skew instead of fixed strikes?
What are the three explanations for the equity-index skew?
What is the leverage effect?
Why does the volatility skew steepen in a selloff?
Does a steep skew predict a crash?
What is a 25-delta risk reversal?
Why is a flat-volatility delta wrong when skew exists?
What is the skew delta or minimum-variance delta?
Is the skew the same across all expiries?
Why is BANKNIFTY's skew steeper than NIFTY's?
Can the volatility skew be traded directly?
Does the skew tell you anything about direction?
How does the skew relate to India VIX?
Why does hedging demand keep the put wing permanently bid?
What does a positive versus negative risk reversal mean?
Does the skew flatten when the market rallies?
Voice search & related questions
Natural-language questions people ask about volatility skew.
Why are puts more expensive than calls on NIFTY?
Is the skew a good crash indicator?
How do traders quote the skew?
Why does my delta hedge keep drifting on index options?
Does every market have a skew like NIFTY?
Can I earn the skew premium safely?
Why is BANKNIFTY's skew different from NIFTY's?
Sources & references
- Emanuel Derman — Regimes of Volatility (Goldman Sachs, 1999)
- John Hull & Alan White — Optimal Delta Hedging for Options (2017)
- Fischer Black — Studies of Stock Price Volatility Changes (1976, leverage effect)
- NSE — India VIX methodology
Last reviewed 10 July 2026. Educational content only — not investment advice.