On Wednesday 10 Jun 2026 Nifty closed at 23,214.95. The June future settled at 23,240.10 and the July future at 23,344.90. One index, three prices, each higher than the last. That staircase is easy to read as a forecast. It is mostly the price of time. The June future carried 25.15 points of premium over the index with 20 days to run, and interest alone should have added about 65.0. I read a premium as a schedule before I read it as a signal. This lesson takes it apart: what the basis is, what the cost of carry adds, and why the premium has to melt by expiry. Every figure comes from NSE end of day files, dated and not live.
3By the end you can
- Say what the basis is and work out the interest part of a premium from the spot, the rate and the days.
- Explain why a premium is not a forecast and why it reaches zero at expiry.
- Name what pushes a premium away from interest: dividends, arbitrage costs and demand.

The idea, plainly
A future and the index it follows meet on one day, the expiry. Until then they can differ, and the difference has a name. The basis is the future's price less the index. When the future is higher, the basis is a premium.
Future = spot
+ interest on the spot
less dividends
Interest =
spot x rate x days / 365
One index, three prices: each later date adds more carry
Read 5 more paragraphs, 278 words
Why would anyone pay more for the future? Think of rent on money. To own the index today you pay for it today. A future lets you pay at expiry. The seller, who can hold the index meanwhile, charges you for the money tied up. Against that, whoever holds the shares collects the dividends paid before expiry. The future's owner does not, so the price is lowered by them.
That is the cost of carry. Interest is the easy part.
I use a short term government bill yield as the rate, a proxy and not anyone's real funding cost. Dividends are the part you can only estimate.
What does a premium cost the buyer of a future? If the index finishes where it started, a long future loses the whole premium by expiry, and that looks like a fee. Set it against the money you did not spend. For the same exposure you left your rupees in a bank or a bill, and they earned the interest that the premium is made of. Before dividends and costs the two roughly cancel. The premium is the price of holding the index on your behalf, not a toll.
Two things follow. First, the premium is a schedule. At expiry the exchange sets the future's final price to the index close, for expiries before 7 Dec 2026, so the premium has to be zero on that day whatever the index does. Second, the premium is not a forecast. Interest, dividends and the cost of arbitrage set it in advance, and none of them knows where the index goes next. The short answer is on the cost of carry question page.
Who keeps the premium honest?
Traders who can hold the whole index basket keep the future close to its fair value. When the future is dear against the shares, they sell it and buy the shares. When it is cheap, they do the reverse. They earn the gap, less their costs, so a gap smaller than their costs can stay. For most retail traders the costs eat it. The trade is the subject of module I29, and the fair value arithmetic with dividends is in module I01.
Arbitrage keeps the premium near fair value, down to the cost of doing it
Worked example: the premium on 10 Jun and 29 Jun 2026
Here are the June and July futures on 10 Jun 2026, with the interest part worked out at the bill yield for each horizon, 5.11% for June and 5.15% for July. Interest is worked out with the formula above.
| 10 Jun 2026 | June future | July future |
|---|---|---|
| Days left | 20 | 48 |
| Future settles at | 23,240.10 | 23,344.90 |
| Premium over the index, points | 25.15 | 129.95 |
| Interest on the index, points | 65.0 | 157.2 |
| Interest less premium, points | 39.9 | 27.3 |
Annualised, the next month premium ran highest in late 2025 and lowest by mid 2026
The June future's premium trended down in jumps, not in a straight line
Read 4 more paragraphs, 279 words
On both contracts the premium sat below the interest. Dividends can explain part of that gap, because the holder of the shares collects whatever the companies pay before expiry. They cannot explain all of it. The July gap is smaller than June's, yet a longer window holds more dividends, not fewer. So something else is at work: the cost of arbitrage, which lets a gap stay open, and plain supply and demand for the future itself. An end of day file cannot split them.
To compare contracts with different days left, annualise the premium: premium over spot, scaled to a year. The June premium came to 1.98% a year and the July premium to 4.26%, against bill yields of 5.11% and 5.15%. Annualising turns points into a rate, so a premium of a few points over a few days and one of many points over many days can sit on one scale.
The premium also drifts away from interest as expiry nears. On Monday 29 Jun, with 1 day left, interest on the index came to only 3.3 points, yet the premium was 26.65, which is 23.4 above it. On Tuesday 30 Jun the final settlement price was the index close, so whatever premium was left had to go. Had the index closed flat on 30 Jun, a long future bought at the 29 Jun settlement of 23,972.90 would have lost the whole premium, Rs 1,732.25 a lot.
The payoff builder below opens on the Long future of 10 Jun 2026, June expiry. Switch the expiry to July as well. Take the index close off each future and check the premium rows of the table above.
The common mistake
The belief: The future trades above the index, so the market expects Nifty to rise.
The formula already explains a premium with no view on direction: spot, interest, dividends. Its size did not line up with what came next either. On 10 Jun the June premium of 25.15 points sat well under the 65.0 points of interest, and the index rose 774.20 points in the next four sessions. On 29 Jun the premium of 26.65 points sat far over the 3.3 points of interest for one day, and the index closed lower on expiry day. A premium is set in advance by interest, dividends and the cost of arbitrage, and it melts on a schedule that the index does not control.
Key takeaways
- The basis is the future less the index. A positive basis is a premium.
- Future is roughly spot, plus interest on the spot, less dividends.
- At expiry the final price is the index close, so the premium is zero.
- A premium is set in advance. It is not a forecast of direction.
- Dividends pull a future below its interest carry. Arbitrage costs let gaps stay open.
- Annualise a premium to compare contracts with different days left.
2 terms this section turns on
Before you trade. Derivatives trading carries a high risk of loss. SEBI's study "Profitability of Individual Traders in the Equity Derivatives Segment (FY25 to FY26)" (August 2026) found that 87.7% of individual traders in equity F&O made net losses in FY26, with aggregate net losses of Rs 91,685 crore. Read the SEBI study