Term Structure Intermediate IV across expiries Forward-looking

What is Term Structure? Term Structure

The one chart that tells you what India VIX cannot: not how frightened the market is, but for how long.

Quick answer: The volatility term structure is the curve you get when you plot at-the-money implied volatility against days to expiry — and its slope, not its level, tells you whether the market thinks the current volatility is here to stay or here to pass.

In simple words

Pick a single strike near the money — say the 24,000 NIFTY strike with spot at 24,000 — and look at its implied volatility in the nearest weekly expiry, then the next, then the monthly, then the quarter. You will get a different number each time: about 11.7% at 7 days, 12.9% at 30 days, 14.6% at 90 days, 15.4% at 180 days. Draw those points against the number of days to expiry and you have drawn the term structure. It is not one volatility. It is a curve, and the market is telling you a slightly different story at every horizon.

The headline India VIX number is a single point on a curve like this — a 30-day figure. When a screen says "VIX is 13", it has thrown away the shape and kept one dot. The shape is where the information is: a market at VIX 13 that expects to stay at 13 looks completely different, on this curve, from a market at VIX 13 that has just been frightened and expects to calm down. Both print 13 in the middle. They disagree everywhere else.

Not to be confused with: The volatility skew, which plots implied volatility across strikes at a single expiry. Term structure plots across expiries at a single (at-the-money) strike. Skew is the shape you read down an option chain; term structure is the shape you read across the row of expiry dates. Both are slices through the volatility surface — one holds time fixed, the other holds moneyness fixed.

At-the-money implied volatility across expiries

One strike, many horizons

At-the-money NIFTY implied volatility plotted against days to expiry, calm-market curve.

The curve rises steeply out of the front and then bends flat: the gap between 7 and 30 days is larger than the gap between 90 and 180 days. That bend is the fingerprint of volatility mean-reversion — the market will price a spike into next week but refuses to commit to six months of anything. A flat headline VIX hides this entirely.

Professional explanation

The slope is the signal; the level is the noise

Almost everyone who watches volatility watches the level — India VIX at 11, or 13, or 22. The level tells you how expensive 30-day optionality is right now, and very little else. The slope tells you what the market believes about the durability of that price. An upward slope says today's calm is expected to persist and the only reason far-dated options cost more is that more can happen over more time. A downward slope says today's fear is expected to fade. Two markets can print an identical India VIX and hold diametrically opposed views about the next two months, and the only place that disagreement is visible is in the slope of this curve. If you read one number off the volatility surface, read the level. If you read two, read the level and the slope, and the slope is the one that changes how you would actually position.

Why the curve flattens at the long end

Volatility mean-reverts. It clusters in the short run — a violent day tends to be followed by another violent day — but over months it is pulled back toward a long-run average that sits, for NIFTY, somewhere in the low-to-mid teens. The term structure is the market pricing that mean-reversion in real time. The front of the curve is free to move a long way, because a spike next week is entirely plausible and the market will charge for it. The back of the curve barely moves, because no credible story keeps volatility elevated — or suppressed — for six straight months. So the curve is steep where mean-reversion has not yet had time to act and flat where it has. The mathematical consequence is that a 7-day implied volatility can double while a 180-day implied volatility shifts by a point. The market will not commit to six months of anything, and the flat long end is that refusal, drawn.

The front is where the regimes disagree, and where you get paid or hurt

Look at the two-regime chart: the calm and stressed curves are far apart at 7 days and almost touching at 180. This is not a drawing convenience — it is structural. The long end is anchored to a mean that neither regime disputes, so both curves are dragged toward the same place. The front is unanchored, so it swings with the current mood. That has a hard practical consequence: nearly all the movement in a term-structure trade lives in the front two expiries. A position built on the shape of the curve is really a position on the front, and the front is precisely where implied volatility is most violent, where a scheduled event can add several points overnight, and where the number becomes unreliable in the final sessions before expiry. The information is concentrated at the front, and so is the risk. Those are the same sentence.

Total variance can never decrease with time — the one hard rule

The term structure of implied volatility can slope any way it likes, but the term structure of total variance cannot. Total variance is σ²T — the implied volatility squared, multiplied by the time to expiry in years — and it must be non-decreasing as you go further out. The reason is a no-arbitrage argument: the variance realised between today and a far date is the variance to a near date plus the variance between the two, and that second piece cannot be negative. If total variance ever fell as you extended the horizon, you could construct a calendar position that locks in a risk-free gain, and the market would erase it. This is why a steeply backwardated curve — high near, low far — is not a contradiction even though near volatility exceeds far volatility: multiply each by its own T and the total still climbs. On the calm curve on this page the numbers run 0.00026 at 7 days, 0.00089 at 21, 0.00136 at 30, 0.00248 at 49, 0.00523 at 90, 0.01163 at 180 — strictly increasing, as it must be. A term-structure model that violates this is not aggressive; it is broken.

It is a forecast the market makes about itself, and it is systematically wrong at the front

The upward slope of the calm term structure is not a neutral fact about time — it is the market forecasting that volatility will rise from here, and it usually does not. In quiet regimes the front of the curve prints below the back, which is a bet that realised volatility will drift up toward the far implied. Most of the time it does not; the market stays quiet and the near options expire having overcharged, session after session, which is exactly the volatility risk premium showing up in the shape of the curve rather than in its level. The uncomfortable part is that the curve is least reliable at the front, which is where it moves most and where people most want to trade it. The steep, informative-looking front end is steep precisely because it is pricing events that mostly will not happen — and on the rare occasion one does, the front does not drift, it gaps, and it does so in the direction that hurts whoever was collecting the premium.

The same axis, two regimes

Calm (upward) and stressed (downward) term structures drawn on one pair of axes.

The two regimes agree least at the front and converge at the back. A calm curve rises from about 11.7% at 7 days; a stressed curve falls from about 26.1%. By 180 days they are within a couple of points of each other, because both are being dragged toward the same long-run average. Whatever the front is doing, six months out the market has stopped believing it will last.

Formula

The term structure and its no-arbitrage constraint

σ_ATM(T) plotted against T subject to σ²(T)·T non-decreasing in T

The term structure is simply the function that maps each time-to-expiry T to the at-the-money implied volatility for that expiry. It has no fixed shape, but it is not free: total variance, σ²(T)·T, must never fall as T rises. A curve that slopes downward in volatility (backwardation) is legal; a curve whose total variance slopes downward is an arbitrage.

σ_ATM(T)At-the-money annualised implied volatility for the expiry that is T years away, expressed as a decimal (0.129 = 12.9%). This is the y-axis of the curve.
TTime to expiry in years, measured as calendar days ÷ 365. This is the x-axis, usually drawn in days for readability.
σ²(T)·TTotal (undiscounted) variance to expiry T — the quantity that must be non-decreasing. It is the variance the market is pricing over the whole life of the option, not the annualised rate.

Forward variance — the increment between two expiries

σ_fwd(T₁,T₂) = √( (σ₂²T₂ − σ₁²T₁) / (T₂ − T₁) )

The volatility the market is implicitly pricing for the window between a near expiry T₁ and a far expiry T₂. The non-decreasing-variance rule is exactly the statement that the quantity inside this square root can never be negative. This forward volatility, not either spot implied volatility, is what a calendar spread actually trades.

How to read a volatility term structure

Fix the strike at the money — the strike nearest spot, 24,000 for NIFTY here — and hold it fixed across every expiry. Reading different strikes at different expiries mixes term structure with skew and tells you nothing clean.
Read the at-the-money implied volatility for each listed expiry: the nearest weekly, the following weeklies, the monthly, and any quarterly that trades. Write them next to their days-to-expiry.
Plot volatility on the vertical axis against days to expiry on the horizontal. Do not connect a weekly to a monthly with a straight line and trust the middle — the curve bends, so interpolate with the shape in mind.
Read the slope first. Upward (near cheaper than far) is contango, the default calm state. Downward (near richer than far) is backwardation, a stress signal.
Read the steepness of the front separately from the back. A steep front with a flat back is normal mean-reversion; a front that has lifted above the back is the market pricing imminent, temporary stress.
Sanity-check the no-arbitrage rule: multiply each implied volatility squared by its time in years and confirm the products never decrease as you go further out. If they do, one of your quotes is stale.
Only now form a view — and remember the shape is confirmed after the fact. The curve tells you the regime you are in, not the regime you are about to be in.

Practical example

NIFTY worked example

Take the calm NIFTY curve with spot at 24,000. The 7-day at-the-money option implies about 11.7%, the 30-day about 12.9%, the 90-day about 14.6% and the 180-day about 15.4%. The slope is upward — contango — so the market is calm and expects to stay that way. Now convert each to a total variance to check the hard rule: 0.117² × 7/365 = 0.00026, 0.129² × 30/365 = 0.00136, 0.146² × 90/365 = 0.00523, 0.154² × 180/365 = 0.01163. Strictly increasing, as required. Notice the shape of the increases: volatility rose by 1.2 points from 7 to 30 days but only 0.8 points from 90 to 180 days, even though the second gap covers three times as many days. That flattening is the whole point. The market is charging hard for the possibility of a spike inside the next month and almost refusing to raise its price for the six-month horizon, because it does not believe any condition lasts that long. The curve is not predicting calm forever — it is predicting reversion to the mean, and drawing exactly how fast it expects that reversion to bite.

BANKNIFTY worked example

BANKNIFTY runs the same shape at a higher level and a steeper front, and the reason is instructive. BANKNIFTY is a concentrated basket of lenders, so a single RBI policy decision or a credit shock moves the whole index at once — which means its near-dated implied volatility is both higher and jumpier than NIFTY's. On a calm day you might see a BANKNIFTY term structure of roughly 14% at 7 days rising to 17% at 90 days, a steeper climb than NIFTY's. But watch what a scheduled RBI Monetary Policy Committee meeting does: it lifts only the expiries that contain the meeting date, so the weekly straddling the announcement pokes above its neighbours and the curve develops a local bump rather than a smooth slope. That bump is not a regime change — it is a single event priced into a single expiry, and it collapses the moment the decision is public. Reading it as backwardation would be a mistake. The lesson NIFTY's smooth curve does not teach: term structure is not always monotone, and event calendars are why.

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.

Risk note. An upward-sloping term structure is frequently sold as a reason to be short front-month volatility — collect the high near-dated premium, let it decay, repeat. The slope does make that premium real, but the slope is compensation for the risk that the front gaps up, not a free harvest. The front of the curve is exactly where volatility moves fastest and least predictably, and a term-structure trade concentrates its exposure there. The shape tells you the regime you are in; it does not tell you the regime is about to persist, and it inverts fastest precisely when a short-front position is largest.

Advantages & limitations

What it is good for

It separates the level of volatility from the market's view on its durability — information that a single India VIX print physically cannot contain, because VIX is one point on this curve.
Its slope is a cleaner regime indicator than its level. A market can be at an unremarkable VIX and still, through an inverted curve, be signalling acute short-term stress.
It makes the no-arbitrage structure of volatility legible. Total variance must rise with time, so the curve constrains what any two expiries can jointly imply and flags stale or mispriced quotes.
It is the natural frame for any multi-expiry position — calendars, diagonals, and volatility-futures rolls all trade the shape of this curve rather than the level of any single point on it.
It exposes mean-reversion directly. The rate at which the curve flattens is the market's own estimate of how quickly volatility returns to average, readable at a glance.

Where it breaks down

The curve is confirmed only after the fact. It tells you which regime you are in today; it has no reliable power to tell you which regime you will be in tomorrow, and it can invert or re-steepen without warning.
The front end, which carries most of the information and most of the movement, is also where implied volatility is least stable and becomes meaningless in the final sessions before expiry as the at-the-money premium decays toward zero.
It is distorted by scheduled events. A single expiry containing an RBI meeting or the Union Budget will bump above its neighbours, breaking the smooth shape without any change in the underlying volatility regime.
It assumes you can read a clean at-the-money implied volatility at every expiry. Far-dated index options in India are often thinly traded, so the long end of the curve can be built from stale or wide quotes that imply a smoothness that is not really there.
It is a single strike's story. The at-the-money term structure says nothing about how the skew changes with expiry, and two markets with identical at-the-money curves can carry very different tail pricing.

Common mistakes

Reading only the level and ignoring the slope. Two markets at India VIX 13 can hold opposite views on the next two months, and the entire disagreement lives in a slope that a level-only reading throws away.
Treating an inverted curve as a trade signal rather than a regime description. Backwardation confirms stress that is already here; it does not tell you the stress will deepen, persist, or reverse, and it frequently does the last of those just as positioning gets crowded.
Comparing implied volatilities at different expiries without checking total variance. A far volatility below a near one is not automatically an arbitrage — you must multiply each by its own time before the comparison means anything.
Interpolating the curve with straight lines. The term structure bends sharply out of the front, so drawing a line from a weekly to a monthly and trusting the midpoint systematically misprices the expiries in between.
Mistaking an event bump for backwardation. A single expiry lifted by a scheduled RBI or Budget date is a local artefact, not a downward-sloping regime, and selling it as if the whole curve had inverted is a common and expensive error.
Assuming the front of the curve predicts realised volatility well because it looks the most informative. It is the steepest and the noisiest part of the curve, and in calm regimes it systematically overprices the movement that subsequently arrives.

Professional usage

Volatility desks quote and hedge off the whole term structure, not a single expiry. A market maker fits a smooth arbitrage-free curve through the at-the-money implied volatilities of every listed expiry and uses it to price the gaps the exchange does not quote — the weeks between a weekly and a monthly — while continuously checking that total variance never decreases across the curve, because a violation is either an arbitrage for a competitor to seize or a data error in their own feed. Relative-value volatility traders express views on the shape rather than the level: long the flat, cheap belly against the rich front, or the reverse, structured so the position is neutral to a parallel shift in volatility and exposed only to a change in slope.

Risk managers watch the front-to-back spread as a stress gauge that leads the level. An inverting term structure often shows up before the headline volatility number spikes, because the near expiries are the first to reprice when hedging demand arrives, so the slope is an early-warning instrument even when the level still looks calm. On the sell side, the term structure is decomposed into forward volatilities — the implied volatility of each future window between expiries — and it is those forward figures, not the spot implieds, that are marked, risk-managed, and traded, because they are what a calendar or a diagonal is actually long or short.

Key takeaways

The term structure is at-the-money implied volatility plotted against days to expiry. Its slope carries more information than its level, and the slope is invisible in a single India VIX print.
Upward slope (contango) is the calm default; downward slope (backwardation) says the market expects current stress to fade. The curve flattens at the long end because volatility mean-reverts and the market will not commit to six months of anything.
Total variance, σ²T, must be non-decreasing with expiry — a hard no-arbitrage rule. A downward-sloping volatility curve is legal; a downward-sloping variance curve is an arbitrage.
The front is where the two regimes disagree most and where nearly all the movement — and risk — of a term-structure position lives.
The curve describes the regime you are in, not the one you are heading into. It is confirmed after the fact and can invert without warning.

Stop reading volatility as a single number and start reading it as a curve, and a whole layer of the market becomes visible that India VIX alone conceals. The level tells you the price of 30-day uncertainty; the slope tells you whether the market believes that price will hold. The front tells you the mood and the back tells you the mean it expects to return to. None of it forecasts tomorrow — the curve is a snapshot of belief, not a schedule of events — but a trader who reads the shape at least knows which weather they are standing in, which is more than the headline number will ever tell them.

Frequently asked questions

What is the volatility term structure?

The volatility term structure is the curve you get by plotting at-the-money implied volatility against days to expiry, holding the strike near the money. On a calm NIFTY it might run from about 11.7% at 7 days to 15.4% at 180 days. It is not a single volatility — it is a different number at every horizon, and the shape is the information.

Why does the term structure slope upward most of the time?

Because in calm markets the front is cheaper than the back — more can happen over more time, and the market is not pricing any imminent shock. This upward slope is called contango and it is the default state of index option volatility. It is also, in part, the volatility risk premium visible as a shape: the front usually overcharges for movement that does not arrive.

What does the slope of the term structure tell me that the level does not?

The slope tells you whether the market thinks the current volatility will last. Two markets at an identical India VIX can slope in opposite directions and therefore hold opposite views on the next two months. The level prices 30-day uncertainty; the slope prices its durability, and only the slope changes how you would position across expiries.

Why does the term structure flatten at the long end?

Because volatility mean-reverts. The front can move a long way, since a spike next week is plausible, but no credible story keeps volatility elevated or suppressed for six straight months, so the back barely moves. The flat long end is the market refusing to commit to any condition lasting half a year.

Is the term structure the same as the volatility skew?

No. The skew plots implied volatility across strikes at one expiry; the term structure plots across expiries at one at-the-money strike. Skew is read down the option chain, term structure across the row of expiry dates. Both are slices of the same volatility surface, one holding time fixed and the other holding moneyness fixed.

What is the no-arbitrage constraint on the term structure?

Total variance, which is implied volatility squared times time to expiry, must never decrease as you go further out. A downward-sloping volatility curve is perfectly legal, but if the total variance ever fell with time you could build a calendar position that locks in a risk-free gain, so the market prevents it.

Can a far-dated implied volatility be lower than a near-dated one?

Yes — that is exactly what backwardation is, and it is not an arbitrage. Volatility can slope downward as long as total variance still rises, because the far expiry multiplies a lower volatility by a larger time. You must compare σ²T, not σ, before deciding whether two expiries are jointly consistent.

How is India VIX related to the term structure?

India VIX is a single point on this curve — a model-free 30-day implied volatility for NIFTY. It captures the level at one horizon and discards the shape. Two markets with the same India VIX can have completely different term structures, which is why the curve tells you things the headline number cannot.

Does the term structure predict future volatility?

Not reliably. It describes the regime you are in today, and it is confirmed only after the fact. The front, which looks the most informative, systematically overprices subsequent movement in calm markets, and the curve can invert or re-steepen without warning. It is a snapshot of belief, not a schedule of events.

What causes the term structure to invert?

A sharp selloff or a liquidity shock, when hedging demand hits the near expiries first and drives front-dated implied volatility above the back. The market is saying the current stress is real but temporary — high now, expected to fade. Inversion is a regime description, not a forecast that the stress will deepen.

Why is the front of the curve the most important part?

Because that is where the two regimes disagree most and where nearly all the movement of a term-structure position lives. The back is anchored to a long-run mean both regimes accept; the front is unanchored and swings with the current mood, which makes it the most informative and the riskiest part of the curve at once.

How do I read the term structure off an option chain?

Fix the at-the-money strike, read its implied volatility in each expiry, and plot those against days to expiry. Hold the strike constant — reading different strikes at different expiries mixes term structure with skew. Then read the slope first and the front-versus-back steepness second.

What is a normal shape for the NIFTY term structure?

In calm conditions, gently upward-sloping and concave: a steep rise out of the front that flattens toward the long end, roughly 11.7% at 7 days easing up to 15.4% at 180 days. The concavity is mean-reversion. A downward or humped shape signals stress or a scheduled event rather than a normal regime.

Can the term structure have a bump instead of a smooth slope?

Yes, and scheduled events are usually why. An RBI Monetary Policy Committee meeting or the Union Budget lifts only the expiries that contain the event date, so a single weekly can poke above its neighbours. That bump is a local artefact of one event in one expiry, not a change in the overall regime.

Why does the term structure matter for calendar spreads?

Because a calendar trades the spread between two points on this curve and the way that spread moves, not the level of volatility. The relevant quantity is the forward volatility between the two expiries, which is derived from the curve. If you do not understand the term structure, you do not understand what your calendar is actually long or short.

What is forward volatility on the term structure?

Forward volatility is the implied volatility for the window between two expiries — for example, the volatility priced from day 21 to day 49, distinct from either spot implied. It is computed as the square root of the difference in total variances divided by the time gap, and it is what multi-expiry positions actually trade.

Is an upward-sloping term structure a reason to sell front-month options?

Not on its own. The slope makes the near premium genuinely richer, but that richness is compensation for the risk that the front gaps up, not a free harvest. Selling the front concentrates your exposure in the most violent, least predictable part of the curve, and the slope tells you nothing about whether the calm will last.

How does the term structure differ between NIFTY and BANKNIFTY?

BANKNIFTY, a concentrated basket of lenders, runs a higher and steeper front because a single credit or policy shock moves the whole index at once. Its curve is also more prone to event bumps around RBI meetings. NIFTY, broader and more diversified, tends to show a smoother, flatter term structure at a lower level.

Why can the long end of the curve be unreliable in India?

Because far-dated index options are often thinly traded here, so the quotes are wide or stale. A smooth-looking long end can be an artefact of interpolation rather than genuine two-sided markets, which means the flatness you see may be partly the absence of trading rather than a real market view.

Does the term structure tell me anything about direction?

No. Like all implied volatility, it is sign-agnostic — it prices the magnitude of expected movement at each horizon, not its direction. An inverted curve says the market expects large near-term movement, not whether that movement is up or down, and reading it as directional is a common error.

What is the difference between the volatility term structure and the VIX futures curve?

The volatility term structure is built from option implied volatilities across expiries on the same underlying. The VIX futures curve plots the prices of futures on the volatility index itself across their delivery months. They usually slope the same way and for related reasons, but they are different instruments — one is options, the other is futures on an index.

Why does total variance have to increase but volatility does not?

Because variance is additive over non-overlapping time windows and cannot be negative, so cumulative variance can only grow. Volatility is variance per unit time, an average rate, and an average can fall even as the cumulative total rises — which is exactly how a backwardated curve stays arbitrage-free while sloping downward in volatility terms.

Voice search & related questions

Natural-language questions people ask about what is term structure?.

What does the term structure of volatility actually show me?

It shows you the market's price for uncertainty at every horizon at once, instead of just the 30-day snapshot that India VIX gives. You see whether near-term volatility is cheaper or richer than long-term, which tells you what regime you are in far better than a single number ever could.

Why is the curve steep at the front and flat at the back?

Because volatility comes back to average over time. The market will happily price a spike into next week, so the front swings around, but it won't believe volatility stays high or low for six months, so the back sits still. The flattening is mean-reversion drawn as a curve.

Is a downward-sloping curve a warning sign?

It's a description of stress that is already present, not a warning about stress to come. A downward slope — backwardation — says the market thinks current fear is real but temporary. It confirms the regime you're in; it does not forecast that things will get worse or better from here.

Can I trust the shape of the curve to time a trade?

Be careful. The shape is confirmed after the fact, and it can flip without warning. It is genuinely useful for knowing which regime you are standing in and for pricing multi-expiry positions, but treating an inverted or steep curve as a timing signal is how people get caught, because it inverts fastest when positioning is most crowded.

Why do people say the slope matters more than the level?

Because the level just tells you how expensive 30-day options are, while the slope tells you whether the market thinks that price will hold. Two markets can be at the same VIX and disagree completely about the next two months, and that disagreement only shows up in the slope. It is the part that changes how you'd position.

How is this different from the smile everyone talks about?

The smile, or skew, is across strikes at one expiry — how volatility changes as you move away from the money. The term structure is across expiries at one strike — how volatility changes as you move further out in time. Same surface, different slice. The smile is a column of the chain; the term structure is a row.

If the near volatility is higher than the far, isn't that an easy arbitrage?

No, and that instinct is exactly the trap. A higher near volatility is backwardation, and it is arbitrage-free as long as total variance still rises with time. To check whether two expiries are actually inconsistent you have to compare variance times time, not the volatilities themselves — and when you do, the apparent inconsistency almost always vanishes.

Sources & references

Last reviewed 10 July 2026. Educational content only — not investment advice.