Impact Cost on NSE: How the Exchange Measures Liquidity and Why It Decides Index Eligibility

Two stocks can show the same last traded price and the same tiny bid-ask spread on your screen, and still cost you very different amounts to buy in size. That gap is what impact cost on NSE tries to measure. The exchange uses it as its standard yardstick of liquidity, and it is one of the filters that decides whether a company can even be considered for the Nifty 50.

Most retail traders never see the number. It isn’t printed next to the quote, and it doesn’t show up on the contract note. But once you understand how it’s calculated, you start reading an order book differently, and you understand why some perfectly large companies sit outside the index for years.

Impact cost on NSE: the definition

NSE’s own impact cost explainer defines it as the cost of executing a transaction in a given stock, for a specific predefined order size, at any given point of time. In plain terms: how much worse than the “fair” price do you end up paying (or receiving) because you had to eat through several layers of the order book?

The fair price here has a specific name. NSE calls it the ideal price, and it is simply the average of the best buy and best sell price:

• Ideal price = (Best bid + Best offer) ÷ 2
• Impact cost (buy) = (Average buy price − Ideal price) ÷ Ideal price × 100
• Impact cost (sell) = (Ideal price − Average sell price) ÷ Ideal price × 100

The answer is a percentage. NSE describes it as the percentage mark-up observed while buying or selling the desired quantity of a stock with reference to that ideal price. The Nifty Indices methodology document (September 2026 edition) carries the same definition and adds an important line: the cost being measured is not brokerage, transaction charges or depository fees. It is the cost that comes purely from a lack of liquidity.

NSE lists four caveats that are easy to skim past but matter a lot:

• Impact cost is computed separately for buying and for selling.
• It can differ for different transaction sizes.
• It is dynamic and depends on the orders sitting in the book at that moment.
• Where a stock is not liquid enough, a penal impact cost is applied.

Why the bid-ask spread isn’t enough

The spread tells you what a round trip costs for a small order. If the best bid is ₹249.90 and the best offer is ₹250.10, buying a handful of shares and selling them straight back loses you 20 paise a share. Fine.

The trouble starts when your order is bigger than the quantity sitting at the best price. NSE’s methodology makes exactly this point with its own example: the spread is valid only as long as the order fits inside the top level of the book. Once you need more shares than are on offer at the best price, the rest of your order fills at the next level, and the next. Your average price drifts away from the quote you saw.

That’s why the exchange treats impact cost as a more realistic measure of liquidity. It is, in NSE’s words, closer to the true cost of execution faced by a trader than the bid-ask spread.

The spread tells you what a small trade costs. Impact cost tells you what your trade costs, at your size, in the book as it stands right now.

A worked example with a hypothetical order book

Let’s build one. The stock and every number below are made up for illustration; they are not live market data.

Step 1: the ideal price. (₹249.90 + ₹250.10) ÷ 2 = ₹250.00.

Step 2: a small buy of 800 shares. The whole order fills at the best offer of ₹250.10. Impact cost = (250.10 − 250.00) ÷ 250.00 × 100 = 0.04%. For a small order, impact cost is just half the spread expressed as a percentage.

Step 3: a bigger buy of 4,000 shares. Now the order walks up the book:

• 800 shares at ₹250.10 = ₹2,00,080
• 1,500 shares at ₹250.30 = ₹3,75,450
• 1,700 shares at ₹250.60 = ₹4,26,020

Total outlay is ₹10,01,550, so the average price is ₹250.3875. Impact cost = (250.3875 − 250.00) ÷ 250.00 × 100 = 0.155%. The quote said ₹250.10. We actually paid almost 39 paise a share above the ideal price.

Step 4: a sell of 4,000 shares. The bid side is thicker near the top, so the picture changes:

• 1,000 shares at ₹249.90 = ₹2,49,900
• 2,000 shares at ₹249.80 = ₹4,99,600
• 1,000 shares at ₹249.60 = ₹2,49,600

That’s ₹9,99,100 received, an average of ₹249.775. Impact cost on the sell side = (250.00 − 249.775) ÷ 250.00 × 100 = 0.09%. Same stock, same moment, same quantity, and selling is cheaper than buying. This is exactly why NSE computes the two sides separately.

Step 5: double the size to 8,000 shares. Run the same arithmetic and the buy-side impact cost rises to about 0.26% (average price roughly ₹250.65), while the sell side rises to about 0.17% (average roughly ₹249.59). Bigger order, bigger percentage. Again, this is the second of NSE’s caveats in action.

The same exercise on a thin stock

Now take a second hypothetical stock with an ideal price of ₹100 and a much shallower book: offers of 150 shares at ₹100.50, 250 at ₹101.20 and 600 at ₹102.50. Buying 1,000 shares costs ₹15,075 + ₹25,300 + ₹61,500 = ₹1,01,875, an average of ₹101.875. Impact cost: 1.875%.

On the bid side, with 200 shares at ₹99.50, 300 at ₹99.00 and 500 at ₹98.00, selling 1,000 shares fetches ₹98,600, an average of ₹98.60. Impact cost: 1.40%.

Look at the money involved. A trade of roughly ₹1 lakh moved the price against us by close to 2% on the buy side. For a trader aiming to make 2-3% on a swing, that cost has eaten most of the target before the position has even moved. And it doesn’t appear on any bill.

How Nifty uses impact cost to pick index stocks

This is where the number stops being academic. The methodology document says that for selection into indices such as the Nifty 50 and Nifty Next 50, liquidity is screened on the basis of impact cost.

For the Nifty 50, the rules in the methodology document are specific:

• Universe: constituents of the Nifty 100 that are available for trading in NSE’s Futures & Options segment.
• Trading frequency: 100% in the last six months.
• Liquidity: an average impact cost of 0.50% or less during the last six months, for 90% of the observations, for a portfolio of ₹10 crore.
• Size: average free-float market capitalisation of at least 1.5 times that of the smallest constituent in the index.

The “portfolio of ₹10 crore” part trips people up. The document describes impact cost here as the cost of executing a transaction in a security in proportion to its index weight. So the test isn’t “can you buy ₹10 crore of this one stock cheaply?” It is closer to “if someone bought a ₹10 crore basket of the index, could they buy this stock’s slice of it without moving the price too much?” As a rough illustration, a stock carrying a 2% weight would be tested on a slice of about ₹20 lakh, while a 0.5% weight would mean about ₹5 lakh. Those weights are hypothetical; the proportional principle is what the methodology describes.

The “90% of the observations” part matters too. A stock does not fail just because its book went thin on one bad afternoon. It has to stay under the 0.50% bar consistently over six months.

Broader indices use a looser bar

Impact cost also shows up further down the market-cap ladder. To enter the eligible universe for the Nifty 500, the methodology says companies must have an average impact cost not greater than 1% over the previous six months, and must have traded on at least 90% of days in that period. The Nifty 50 ceiling of 0.50% is half that, and it is tested on a specific ₹10 crore basket.

Our thin hypothetical stock above, at 1.4% to 1.9% on a ₹1 lakh trade, would be in trouble even against the 1% bar if that pattern held across six months of observations. A single snapshot proves nothing; the averages over six months are what count.

Size alone doesn’t get a company into the Nifty 50. It has to be large and cheap to trade, consistently, for six months.

When the index reviews happen

According to the methodology document, the Nifty 50 is reviewed semi-annually using data for the six months ending January and July. Replacements, if any, take effect from the last trading day of March and September, and the market gets four weeks’ prior notice.

There’s a cap on churn: as part of the semi-annual reviews, a maximum of 10% of the index, that is five companies, may be added in a calendar year. But the document is clear that this limit does not apply when a company is excluded for failing eligibility, and it lists higher impact cost among those reasons, alongside things like losing F&O availability or lower trading frequency. In other words, a constituent whose liquidity dries up can be removed regardless of how many changes have already happened that year.

What this means for ordinary traders

You don’t need to calculate impact cost on every trade. But the idea is worth carrying around.

1. Check depth, not just the quote. Before placing a market order in a mid or small cap, look at the top five bids and offers. If your quantity is bigger than what’s sitting at the best price, you will pay more than the screen shows. Our guide to reading market depth before you place an order walks through the five-level book.
2. Use limit orders in thin counters. A limit order caps the price you’ll pay. You may get a partial fill, but you won’t sweep the book by accident.
3. Split large orders. If the book is shallow, breaking an order into smaller pieces over time gives fresh orders a chance to arrive. The trade-off is that the price can move while you wait.
4. Remember the exit. Impact cost hits you twice: once going in and once coming out. In our thin-stock example, a round trip of about ₹1 lakh would have cost a little over 3% before charges. Position sizing that ignores this is optimistic; our risk management in trading guide covers how to size trades around real costs.
5. Very illiquid scrips trade differently. Some stocks are moved out of continuous trading altogether. Our explainer on call auctions for illiquid securities on NSE covers how that works.

Where the measure has limits

Impact cost is a snapshot of visible orders. It can’t see hidden intentions: a large seller waiting off-screen, or buyers who would step in if the price dipped. A book can look deep at 10:30 AM and thin at 3:20 PM.

It also says nothing about direction. A stock with a 0.05% impact cost can still fall 5% in a session. Liquidity makes a stock cheaper to trade. It doesn’t make it safer to own.

And the index test is about averages. A Nifty 50 constituent passed the 0.50% bar on 90% of observations over a review period; that doesn’t guarantee every order you place in it will be cheap, especially around results, block deals or a sharp market fall.

Quick recap

Impact cost is the percentage gap between the price you actually get and the ideal price, the midpoint of the best bid and offer, for a given order size. NSE calculates it separately for buys and sells, it grows with order size, and it changes minute to minute. Nifty Indices uses it as a hard liquidity screen: 0.50% or less on 90% of observations over six months, for a ₹10 crore basket, for the Nifty 50, and an average of no more than 1% for the Nifty 500 universe.

For the rest of us, it’s a reminder that the price on the screen is only the price for the first few shares.

Disclaimer: This article is for information and education only and is not investment advice. The order books, stocks and index weights in the worked examples are hypothetical. Index rules can change; check the latest Nifty Indices methodology document. Please consult a SEBI-registered investment adviser before making any investment or trading decision.