Circuit Limits & Surveillance: The Mechanics That Trap Beginners

A price band — commonly called a circuit filter — is the maximum percentage a security's price is allowed to move away from a base price during a single trading session. The base price is normally the previous session's closing price, and the band is recalculated fresh every day.

The mechanism is completely mechanical. The exchange's system rejects any order priced outside the band. There is no committee deciding anything in the moment; an order at ₹106 on a stock whose upper limit for the day is ₹105 is simply not accepted.

Three reasons the rule exists. It gives time for information to reach everyone, rather than letting a rumour reprice a company in ninety seconds. It contains the damage from a fat-finger error (a typing slip, such as an extra zero) or a coordinated manipulation attempt. And it protects the settlement system, because the clearing corporation collects a margin (a deposit kept against each position) based on the price movement it expects, and an unlimited move would break those calculations.

One thing a circuit is not: it is not a judgement about the company, a suspension, or an allegation of anything. A perfectly ordinary business can hit an upper circuit because a large order arrived in a thin market. Beginners routinely read a circuit as a signal about the business. It is a signal about the order book.

Exchanges apply daily price bands of 2%, 5%, 10% or 20% to securities. Which band a security gets is decided by the exchange based on surveillance criteria and is revised from time to time — it is not something you can derive yourself, which is exactly why you have to look it up rather than assume.

Securities on which derivatives are available, and securities that form part of index derivatives, are handled differently. Instead of a fixed daily band they operate inside a dynamic price band — commonly 10% — which the exchange can widen in steps if the price keeps pressing against the limit. Per NSE's price-band flexing FAQ, the band starts at 10% and is relaxed by 5% at a time in the direction of the move, when the price keeps pressing the limit. (Needs verification: any cooling-off rule after a flex.) Newly listed securities also have special arrangements on their listing day.

Now the part that matters most and that almost nobody works out. The band resets every session on the new closing price. So a 5% band does not mean the stock can only move 5%. It means it can only move 5% today — and then 5% from the new base tomorrow.

Run the compounding. A ₹100 stock with a 5% band, hitting the upper limit for ten consecutive sessions, reaches ₹100 × 1.05^10 = ₹162.89. That is a 62.9% rise, every day of which was 'capped at 5%'. Downward, ten consecutive sessions at the lower limit takes ₹100 to ₹100 × 0.95^10 = ₹59.87 — a 40% fall, and on each of those ten days you may have been unable to sell.

This is the single sentence to carry away: the band limits a day, not a journey. A tight band is not protection. In a falling stock it is a slower, more prolonged trap.

Open the market depth window on a stock locked at its upper circuit and you see something lopsided. There is an enormous pending quantity on the buy side at the limit price, and effectively nothing on the sell side. The last traded price sits at the band. Nothing is trading, because everyone who owns shares wants a higher price and a higher price is not permitted today.

So if you are trying to buy, you cannot. Your order joins the back of a queue that is filled by price first and then by time, meaning the best price and the earliest order go first, and on most locked sessions that queue barely moves. Meanwhile the screen keeps showing a big green number, which makes the stock feel more desirable the longer you fail to get it.

If you already hold the stock, the position shows a large unrealised gain. Here is the honest way to read it: an unrealised gain in a thin counter is a number, not money. The same absence of two-sided trading that produced the upper circuit will produce lower circuits when the move reverses — and the exit door is exactly as narrow as the entrance door was.

Beginners routinely mix up the two circuits. At the upper circuit the missing side is the sell side, so the person who is stuck is the person trying to get in. At the lower circuit the missing side is the buy side, so the person who is stuck is the person trying to get out. Only one of those two costs you money.

There is also a psychological trap worth naming. Consecutive upper circuits are one of the most effective marketing devices in existence — they create urgency, apparent proof and scarcity all at once. That is precisely the structure a coordinated promotion campaign is trying to produce.

Now the version that costs real money. The stock opens at or falls to its lower band. The depth window shows a wall of pending sell orders and almost no buy quantity. You place a sell order. It joins a queue behind everyone who placed one earlier, sorted by price and then by time.

Put numbers on it. Suppose, for illustration, 45 lakh shares are pending on the sell side and 3,000 shares actually change hands all session because one or two small buyers appeared. Your order, somewhere in the middle of that queue, does not execute. It does not execute the next day either, if the same thing happens.

Then the band resets lower each session. Five consecutive lower circuits at a 5% band takes a ₹100 stock to ₹100 × 0.95^5 = ₹77.38, a 22.6% fall. On a ₹1,00,000 position that is ₹22,600 gone across five sessions in which you were placing sell orders every single morning and none of them filled.

This is the moment to understand what a stop-loss order actually is. A stop loss is an instruction that becomes an order when a price is reached. It is not a guarantee of execution. If there is no buyer, the triggered order simply sits there. In a circuit-locked stock, the stop loss you carefully placed does nothing at all.

Gap risk makes it worse. Overnight news can cause a stock to open directly at its lower circuit — below where your stop was. Your stop never had a chance to be tested at a reasonable price, because the price never traded there. The market skipped over it.

Separate from individual stock bands, there is a nationwide mechanism that halts trading across the entire equity and equity-derivatives market. It triggers when either the Sensex or the Nifty 50 moves by 10%, 15% or 20% — whichever index breaches the level first.

The percentages are converted into absolute index point levels in advance, calculated on a quarterly basis using the index closing value on the last trading day of the previous quarter, and rounded. So the exact point levels for the current quarter are published and knowable ahead of time, not decided in the moment.

How long trading stops depends on both the size of the move and the time of day, on the logic that a shock late in the session leaves less time for recovery and orderly reopening. After every halt, trading restarts with a 15-minute pre-open call auction session rather than jumping straight back into continuous trading.

For you as a participant, the important consequence is simple and uncomfortable: during a halt you cannot exit anything. Any leverage you were carrying is carried straight through the halt, and you find out where you stand only when the auction produces a reopening price.

It is also worth knowing that a market-wide halt is a cooling mechanism, not a catastrophe in itself. It is designed to interrupt a feedback loop. Panicking about the halt rather than about your own position size is the wrong reaction to have.

ASM stands for Additional Surveillance Measure. It is a framework run jointly by SEBI and the exchanges that places securities showing unusual trading behaviour under tighter conditions. It comes in two flavours — a Short-Term framework and a Long-Term framework — each with escalating stages.

Selection is based on objective, published criteria applied mechanically. The parameters have included things like high-low price variation, close-to-close price variation, volume variation, the number of unique client codes (separate trading accounts) in the scrip, how concentrated the trading is among a few clients, delivery percentage, market capitalisation and the price-to-earnings ratio. NSE's ASM FAQ lists these criteria, with minimum market-cap conditions for each. No human is forming an opinion about the company.

This is the most important sentence in the section, and exchanges state it explicitly themselves: inclusion in ASM does not mean the company has done anything wrong. It is not an allegation, an investigation or a verdict. It is a measure aimed at a trading pattern.

What it actually does to you is about money. The applicable margin is raised. Margin is the part of the trade value you must keep blocked with your broker, and normally it is only a fraction of the value. Under ASM it escalates at higher stages up to 100% of the trade value, so the whole value of the position must be blocked and leverage in that scrip effectively ceases to exist. Per NSE's ASM FAQ, Short-term ASM starts at 50% margin (or your existing margin if higher) at Stage I and goes to 100% at Stage II, while Long-term ASM stages apply 100% margin. At higher stages the price band may be narrowed and the security may be moved into trade-for-trade settlement.

The practical danger is timing. If you are holding a leveraged position in a scrip when it enters or escalates within ASM, the margin requirement can jump overnight. You either add funds immediately or your position is squared off (closed by the broker) — often at exactly the price you would least like to accept. Brokers may also apply their own restrictions above the exchange minimum, or block the scrip entirely, and that is their risk policy rather than a regulatory rule.

GSM stands for Graded Surveillance Measure. Where ASM is largely about trading patterns, GSM targets securities whose price behaviour is not supported by their financials — the criteria have drawn on things like net worth, net fixed assets, market capitalisation and the price-to-earnings ratio alongside the price movement itself.

GSM escalates through stages, and each stage removes something you were taking for granted. First, settlement moves to trade-for-trade, meaning every trade must result in delivery. Then the price band tightens. Then an Additional Surveillance Deposit (ASD) is collected from the buyer. This is a percentage of the trade value that the exchange holds for a set period. Then, at the higher stages, continuous trading may be replaced by a periodic call auction. Needs verification: the stage-wise band, deposit percentage, holding period and auction frequency, because these have been revised more than once.

A periodic call auction is worth understanding properly because it is completely unlike normal trading. Instead of orders being matched continuously through the session, orders are collected over a defined window and then matched at a single price at the end of that window. At the higher GSM stages, that window has at times occurred only once a week, or even only once a month (Needs verification against the current framework). There is no 'sell it now'. There is only 'place an order into the next window and find out what price the auction produces'.

The deposit is a real cash cost that beginners never anticipate. Suppose a stage carries a 100% deposit. Buying ₹50,000 of the stock then means committing ₹50,000 + ₹50,000 = ₹1,00,000, and the deposit half earns nothing and cannot be used for the deposit period. Your capital is locked twice over: once in the position and once in the deposit.

Put the two together and the picture is stark. At a high GSM stage you may own a position in a stock that trades only in an occasional auction, with a tight band, an extra deposit blocked and delivery compulsory. Every exit route you assumed existed has been removed, one stage at a time.

Trade-to-trade — often written as T2T — means exactly what it says: every single trade must be settled by delivery. On NSE these securities carry the series code 'BE' instead of the normal 'EQ'. On BSE they sit in the 'T' group rather than 'A' or 'B'.

The consequences are strict and catch people out constantly. You cannot square off a T2T position within the same session. If you buy in the morning and the price collapses by afternoon, you cannot sell — you must take delivery and pay the full value. You can only sell after the shares have actually settled into your demat account, which under the present T+1 cycle is normally the next trading day.

There is no margin either. You need 100% of the trade value up front, both as a buyer and as a seller, because the seller must genuinely hold the shares in demat before selling. Intraday product types offered by your broker are unavailable in these securities, and orders using them are rejected.

Securities are shifted into trade-to-trade based on criteria reviewed periodically by SEBI and the exchanges, drawing on parameters such as the price-to-earnings ratio, price variation and market capitalisation. A security can move in and out of T2T between one review and the next.

One related label to know: BSE's 'Z' group flags companies that have not complied with listing requirements or have unresolved investor complaints. It is a different mechanism from trade-to-trade, but the practical effect of running into it is the same — heavy restrictions and very little liquidity.

Entry is mechanical. On the review date, the exchange applies the published criteria to the whole universe. Anything that meets them is added, with effect from a stated date, and the addition is published in the exchange's daily surveillance files. Reviews run on defined cycles that differ between frameworks (for ASM, NSE's FAQ says the stage-wise review is weekly, and Long-term ASM stocks stay at least 90 calendar days; Needs verification: the GSM review frequency), and the current lists are published daily.

Exit is deliberately slower than entry. A security generally has to remain outside the criteria for a defined minimum period before it can come out, and escalated stages usually de-escalate one step at a time rather than releasing the stock straight back to normal conditions. Getting in is quick; getting out takes time.

The consequence for you is a timing risk that nobody warns beginners about. You can research a stock on a Sunday, find it perfectly normal, and place an order on Wednesday into a scrip that has since entered a framework — with a different band, a different margin and possibly a different settlement type from the one you studied.

A security can also migrate between frameworks or escalate stages with no company news whatsoever. Nothing needs to be announced. The trading pattern alone can move it, which is why treating 'no news' as 'no change' is unsafe here.

The discipline this implies is simple and takes about thirty seconds: verify the status on the day you place the order, every time, not on the day you did the research.

Everything in this lesson is publicly available and free. The exchanges publish price bands, settlement series, surveillance lists and the daily bhavcopy on their own websites, and your broker's terminal surfaces most of it on the order entry window.

The security's page on the exchange website is the single most useful destination. It shows the current price band, the settlement series or group, the traded quantity and value, the delivery percentage, and any surveillance indicator applicable to the scrip.

Your broker also plays a part. Most terminals display a surveillance tag next to the symbol and throw an acknowledgement pop-up before accepting an order in a flagged scrip. That pop-up is not a formality to be dismissed — it is often the only warning you will get, and reading it takes five seconds.

Below is the full check. Run it on the day you place the order. If two or more items come back unfavourable, the honest response is usually to skip the trade rather than to reduce your size and proceed.

None of the mechanics in this lesson are dangerous on their own. They become dangerous in a specific order, and it is worth seeing the sequence written out, because once you recognise it you can stop at step one.

It begins with a small counter that is up sharply and is being talked about. It is up sharply because it keeps hitting the upper circuit — and it keeps hitting the upper circuit because very little stock is available to trade, not because a large number of informed buyers agreed on something. The apparent strength and the underlying thinness are the same fact.

For several sessions the beginner cannot buy at all, which increases the desire rather than reducing it. Then one session the lock opens, orders fill, and they are in — near the top, with a larger position than planned, because the wait built up conviction.

The move reverses. The same thinness now produces lower circuits. The stop-loss order triggers and cannot execute because there is no bid. Every morning a sell order is placed and every evening it is still pending, while the band resets lower.

Meanwhile the exchange notices exactly what everyone else noticed. The scrip enters a surveillance framework. Margin goes to 100%, so any leveraged position must be funded immediately or squared off. The band narrows. Settlement becomes trade-for-trade. If escalation continues, continuous trading is replaced by a weekly or monthly call auction with an additional deposit attached.

At no point in that sequence did anyone need to defraud this person. The mechanics did all the work. That is precisely why understanding the mechanics is the defence — and why the defence has to be applied at step one, when the decision to buy a thin, circuit-locked counter is still available to make differently.

Everything above compresses into a short list you can run before any order in any security you do not already know well. It is not a filter that finds good trades. It is a filter that removes trades whose exit route may not exist.

Ask: what is today's price band? What is the settlement series or group? Is it on any ASM or GSM list, and at what stage? What margin is required? What does the market depth actually look like right now — is there real quantity on both sides? And finally, the question that ties it together: if I could not sell this position for ten consecutive sessions, would that be survivable at this size?

That last question is the useful one, because it converts all the mechanics into a single sizing decision. If the answer is no, the size is wrong regardless of how good the chart looks.

One further point on judgement. None of this tells you whether a company is good or bad. A large, liquid company can have a terrible year and a small illiquid one can be a fine business. What these mechanics tell you is what happens to your ability to act — and your ability to act is the thing that risk management depends on entirely.

A closing note that is not a formality. Markets carry genuine risk, and understanding circuit limits and surveillance frameworks reduces avoidable damage without removing the possibility of loss. Everything in this lesson is educational content, not investment advice, and nothing here is a recommendation to buy, sell or hold any security. The percentages, halt durations and stage descriptions here reflect frameworks that SEBI and the exchanges revise from time to time — verify the current rules on the exchange and SEBI websites before you act on any of it.