Circuit Trap Exit Modelling

First published 23 Aug 2026 · Last verified 29 Aug 2026

Lesson 58.1 — The Mechanics of the Circuit Trap

Every trading day on the Nepal Stock Exchange (NEPSE), every listed scrip is fenced in by a price band — a rule that says a stock cannot move more than a fixed percentage away from its previous closing price in a single session. As of the current rules (revised by the Securities Board of Nepal, SEBON, in April 2026), that band is 15% in either direction for most scrips during regular trading, up from the 10% band that applied for years before that. During the pre-open session — the short window before the main market opens, where the day's opening price is discovered — the band is tighter still, at 5%. If a stock closed yesterday at NPR 1,000, today it can trade anywhere between NPR 850 (the lower circuit) and NPR 1,150 (the upper circuit), but nowhere outside that range, no matter how many people want to buy or sell at a different price.

This is what the book has been calling, since Chapter 54, one of the defining structural features of NEPSE: a market where price discovery is deliberately slowed down by regulation, in exchange for reduced single-day volatility. The trade-off sounds reasonable in the abstract. A 15% band prevents a single panicked trading session from wiping out half a company's market value in an afternoon. But this chapter is about what happens on the other side of that trade-off: what a price band does to your ability to exit a position when the stock is falling, not rising.

Separately from the individual stock band, NEPSE also runs a market-wide circuit breaker pegged to the benchmark index rather than to any single stock. Under the rules revised in April 2026, this works on two tiers. If the NEPSE Index moves 5% from the previous close within the first two hours of trading, the entire market halts for 15 minutes, after which trading resumes. If the index moves 8% at any point during the session, the market shuts down completely for the rest of the day. This replaced an older three-tier system (4% / 5% / 6%, in place since 2007) that halted trading in smaller, more frequent steps. The market-wide breaker matters for portfolio-level risk, which earlier chapters in this Part have already covered. This chapter's concern is narrower and, for an individual investor, often more dangerous: what happens to one specific position you own when that one stock — not the whole market — is glued to its lower circuit.

To understand the circuit trap, you first need to understand what "hitting the circuit" actually means mechanically, because it is widely misunderstood. NEPSE, like most modern exchanges, matches buy and sell orders continuously through what is called an order book — a running list of everyone who wants to buy at a given price and everyone who wants to sell at a given price, sorted from best price to worst. When a stock's price reaches its lower circuit — say NPR 850 in the example above — the exchange does not simply "stop" the stock. Orders can still be entered at NPR 850. Sell orders can still be placed at NPR 850. The trading system continues to try to match them. What changes is that no order can be entered below NPR 850, because the system will not accept a price outside the band.

Here is the part that catches investors off guard: a trade only happens when a buyer and a seller agree on a price and quantity. If a hundred people want to sell at NPR 850 and only three people want to buy at NPR 850, then only the volume the three buyers are willing to absorb actually trades. The other ninety-seven sellers' orders simply sit in the queue, unfilled, until the market closes for the day — or until a buyer shows up. If zero buyers show up at NPR 850, then zero shares trade at NPR 850, even though the stock is technically "allowed" to trade there. The stock is not frozen by the exchange. It is frozen by the absence of a counterparty.

The clearest way to build intuition for this is a real-estate analogy, and it is worth sitting with for a moment because it maps almost exactly onto what is happening in the order book. Imagine you own a house you need to sell quickly — perhaps you have moved cities for a new job and need the cash. You list it at a fair price and nobody bids. You cut the price by 15%. Still nobody bids — in fact, word has gotten around the neighbourhood that something is wrong with the area (a factory closure, a flood risk, a change in zoning), and everyone who might have bought is waiting to see how much further prices will fall before they step in. You cut the price by another 15%. Still no buyers. You are not legally prevented from selling your house — there is no circuit breaker on real estate — but functionally, you are stuck, because a sale requires two willing parties and only one side of that transaction currently exists. A circuit-locked NEPSE stock is the same situation, except the price cuts are capped and rationed by the exchange at 15% per day rather than being something you can decide on your own, and the "listing" resets every morning at a new, lower, exchange-determined starting price.

It is important to be precise about why this happens, because the cause matters for how you manage it. A stock does not get circuit-trapped because the exchange is malfunctioning. It gets circuit-trapped because the market has collectively decided — rightly or wrongly — that the fair value of the stock is meaningfully below where it is currently pinned, and nobody wants to be the buyer who catches a falling asset before the selling pressure exhausts itself. This is often triggered by a specific piece of information: a regulatory change affecting a sector (as happened with Nepal's microfinance institutions in 2023, discussed in Lesson 58.5), a disappointing earnings disclosure, a promoter share-pledge default, or simply a broad market-wide risk-off move that hits small, thinly held, high-beta scrips hardest. Once the selling starts and the stock touches its lower circuit with volume imbalanced toward sellers, a self-reinforcing dynamic takes hold: every investor watching the order book sees a wall of unmet sell orders and concludes, reasonably, that today is not the day to buy — better to wait and see if it locks again tomorrow at an even lower price. That expectation becomes self-fulfilling, and it is exactly this dynamic that produces the multi-day cascades covered next.

Lesson 58.2 — The Multi-Day Cascade: Why 15% Can Become 50% or More

A single day's 15% circuit is not, by itself, catastrophic for a well-sized position. What makes the circuit trap a genuine capital-destruction risk is that the lower circuit can repeat for multiple consecutive sessions, and each day's 15% compounds on top of the previous day's already-reduced base. This is arithmetic that a lot of investors underestimate, because they mentally treat "15% circuit" as a ceiling on how bad a single stock's loss can get. It is not a ceiling. It is a floor on how fast the loss can happen on any given day — but there is no rule limiting how many consecutive days that floor can be hit.

The math compounds multiplicatively, not additively, and the difference matters enormously at scale. If a stock falls 15% every day for six consecutive sessions, the investor has not lost 6 × 15% = 90%. The actual loss is 1 − (0.85)^6 ≈ 62.3%, because each day's 15% cut applies to an already-shrunken base. That is still a devastating loss — but the compounding math also means the position never technically reaches zero on paper (0.85 to any power stays positive), even while the investor's real capital is being ground down toward practical worthlessness, day after day, with no ability to sell.

The table below works through this cascade for a hypothetical NEPSE scrip that opens at NPR 1,000 and hits its lower circuit (down 15%) for eight consecutive sessions with no buyers stepping in — a realistic worst case during a sharp sector-specific selloff of the kind Nepal has seen in small-cap hydropower and microfinance names.

Two things about this table deserve emphasis, because they are exactly the features that make circuit-trap risk different from ordinary drawdown risk. First, notice the "Shares Traded" column. In a normal 72.8% drawdown spread over eight days, an investor with a modest position could likely have sold out somewhere in the first two or three days, taking a painful but survivable loss. In a genuine circuit-trap cascade, the investor cannot sell at all during the days when the loss is accelerating fastest (Days 1 through 3), precisely because that is when the seller-buyer imbalance is most extreme. Liquidity typically only starts coming back once the price has fallen far enough that bargain hunters — the same investors this book has discussed elsewhere as value-oriented, patient capital — decide the risk-reward has turned favourable. By the time an exit is actually possible, most of the damage has already been locked in on paper.

Second, notice that the table assumes the circuit eventually breaks on Day 8. It might not. There is no guarantee, and no exchange rule, that says a stock cannot lock lower-circuit for fifteen sessions, or twenty, if the underlying reason for the selling (a regulatory shock, a fraud disclosure, a sector-wide re-rating) is severe enough. The table is illustrative of the mechanics, not a promise about duration. Some of the worst-affected microfinance and finance-company scrips during Nepal's 2022–2023 correction went through cascades considerably longer than eight sessions before real two-sided trading resumed.

The deeper lesson here connects back to something Chapter 54 established at the start of this Part: liquidity is not a constant property of a stock, it is a state that can change discontinuously and without warning. A scrip that traded normally, with tight bid-ask spreads and healthy daily volume, for months can flip into a circuit-trapped state within a single session if the news flow turns bad enough. The circuit band does not prevent this — it only paces it, spreading what might otherwise be a one-day 70% crash into a multi-week grind that still ends up close to the same place, but with the added cruelty that the investor had to watch it happen in slow motion, powerless to act.

Lesson 58.3 — Modelling Days-to-Exit: Building on the ADV Rule

Chapter 56 introduced the Position Size vs. ADV Rule — the governing discipline that an investor's position in any single scrip should be sized relative to that stock's average daily volume (ADV), so that exiting the position does not itself move the market against the investor. The core formula from that chapter, restated here because this chapter builds directly on it, was:

Days-to-exit (orderly market) = Position size (shares) ÷ (Participation rate × ADV)

Where the participation rate is the maximum fraction of a single day's volume an investor is willing to represent without materially distorting the price — commonly somewhere in the 10%–25% range for a disciplined institutional-style investor in a market as thin as NEPSE's small and mid-cap segment. If you hold 50,000 shares in a scrip with an ADV of 25,000 shares, and you cap yourself at 20% participation (5,000 shares a day), your orderly days-to-exit is 50,000 ÷ 5,000 = 10 trading days.

That formula is correct and useful — under normal, two-sided trading conditions. The entire point of this chapter is that a circuit trap invalidates the formula's core assumption. ADV is a historical average computed from days when the stock traded normally. On a day when the stock is locked at its lower circuit with no counterparty, effective tradeable volume for a seller trying to exit is not 20% of ADV, or even 5% of ADV — it can be zero. The days-to-exit formula needs a second, more pessimistic version that accounts for this.

To make this concrete, the table below models days-to-exit for a position under three liquidity regimes: a normal market (the Chapter 56 baseline), a stressed market (volume has thinned but the stock is still two-sided), and a circuit-trapped market (the stock is locked most days, with only intermittent thin trading). Position size is expressed as a multiple of the stock's normal 180-day ADV, which is the convention this Part has used since Chapter 56.

The pattern the table is meant to make visceral is this: the same position that would take five trading days to exit in a normal market — a manageable, almost routine liquidity event — can balloon to well over two calendar quarters of trading days to exit if the stock enters a genuine circuit-trap regime and the investor is unwilling to sell at whatever thin liquidity intermittently appears. And a position sized at 5x ADV, which some retail investors in NEPSE's small-cap segment do accumulate during a rally without ever stress-testing the exit side, could take over a year of trading days to unwind under stressed circuit conditions — by which point the fundamental thesis that justified owning the stock in the first place may be entirely irrelevant.

This modelling exercise is also where Chapter 57's free-float lens becomes directly relevant. A stock with a small free float — the portion of shares actually available for public trading, after excluding promoter and locked-in holdings — has, by definition, a shallower pool of potential counterparties on both sides of the market. Low free float does not cause a circuit trap by itself, but it makes the "circuit-trapped market" column of the table above the more realistic scenario, rather than the exception, for a larger share of NEPSE's listed universe than investors coming from deeper markets might expect. When you size a position using the ADV rule from Chapter 56 and the free-float ceiling from Chapter 57, you are implicitly also sizing your exposure to circuit-trap risk — the three chapters are not independent checks, they are three views of the same underlying constraint.

Lesson 58.4 — Queue Position: Who Gets to Sell First

When a stock is pinned at its lower circuit with a large imbalance of sell orders over buy orders, an important and often overlooked question is: among all the sellers stuck in that queue, who actually gets filled first when a buyer finally does show up? The answer matters directly to an investor trying to plan an exit, because it determines whether being an early or a late seller into the circuit actually helps.

NEPSE's order matching, like most modern exchange systems, follows price-time priority. At a given price level — here, the lower circuit price itself, since that is the only price sellers are permitted to quote — orders are filled in the order they were placed. A sell order entered at 10:15 AM will be matched before a sell order entered at 11:40 AM at the same price, assuming a buyer arrives willing to take that quantity. This means that on the first day a stock locks at its lower circuit, an investor who reacts quickly and places a sell order in the first minutes of trading has a materially better queue position than one who waits, hesitates, or is simply slower to react — even though both are nominally selling at the "same" circuit price.

The practical implication is significant and somewhat counter to how many retail investors instinctively behave. When a stock is falling toward its lower circuit, there is a natural temptation to wait — to hope for a bounce, to avoid "selling at the bottom," to see if the price stabilises before committing. But if the stock does lock limit-down, every investor who waited is now behind, in queue-time terms, every investor who placed their sell order earlier in the session or on a prior day. If any buying interest shows up the next morning, the orders that get filled first are the ones that have been sitting in the queue longest — not the most recently entered ones. An investor who is seriously considering an exit, and who sees a stock approaching its lower circuit with deteriorating order-book depth, is generally better served by entering a sell order promptly rather than waiting to "see how it plays out," because the cost of being wrong about a bounce is small relative to the cost of losing queue priority in an illiquid lock.

This also has an important implication for how an institutional-scale or large individual investor should think about a position that is starting to look shaky, well before it actually locks limit-down. If you are managing a position sized at, say, 2x the stock's ADV — a size that Chapter 56's framework would already flag as requiring a multi-day orderly exit — and you see early warning signs of a coming selloff (a negative sector news item, a broken technical support level, deteriorating order-book depth on the buy side), the queue-position mechanics argue strongly for beginning to scale out immediately rather than waiting for confirmation that the thesis has fully broken. Every day you delay, if the stock does end up circuit-locking, is a day you fall further back in a queue that may only clear a small fraction of its backlog before the next lower-circuit session resets it entirely.

There is a further wrinkle worth flagging for larger investors: because queue position is time-based and not size-based, a large seller cannot "jump the queue" by placing a bigger order — a large sell order simply waits, in its entirety or in the unfilled remainder, behind smaller orders placed earlier at the same price. This means position size compounds the queue-position problem rather than mitigating it. A large holder who is late to start selling is not just behind in time, they also have more total quantity that needs to work through a queue that is already backed up. This is one more reason the position-sizing discipline from Chapters 56 and 57 is not optional risk management — it is the primary lever an investor actually controls, because queue position, once a stock is genuinely circuit-trapped, is largely outside anyone's control.

Lesson 58.5 — Historical Case Studies: Circuit Cascades in NEPSE

The mechanics described so far are not theoretical. NEPSE has produced repeated real-world instances of the circuit trap, concentrated in two identifiable kinds of episodes: broad market-wide corrections that hit thinly-traded small caps hardest, and sector-specific regulatory shocks that trigger sharp, concentrated selling in a single industry group.

The clearest broad example is Nepal's 2021–2022 market cycle. The NEPSE Index ran from roughly the 1,200s in mid-2020 to an all-time high above 3,200 in mid-2021, driven substantially by retail participation, margin lending, and pandemic-era liquidity — a rally this book has referenced in earlier chapters as a case study in unsustainable market-wide leverage. When the correction came through late 2021 and into 2022, driven by monetary tightening, margin calls, and a sharp reduction in market liquidity, the index fell back into the 1,800s–2,000s range over the following months. That decline was not smooth. On the worst days, market-wide circuit breakers triggered at the index level (the mechanism covered briefly in Lesson 58.1), and beneath the index-level halts, a large number of individual small-cap and mid-cap scrips — particularly in hydropower, finance, and microfinance, the sectors with the thinnest free floats per Chapter 57 — spent repeated sessions locked at their lower circuits, with order books showing overwhelming seller imbalance.

A second, more sector-concentrated example came with Nepal's microfinance sector in 2022–2023. A combination of Nepal Rastra Bank's tightened interest-rate spread caps, stricter loan-loss provisioning requirements, and rising over-indebtedness concerns among microfinance borrowers triggered a sharp re-rating of microfinance equities across the board — not isolated to one or two companies, but affecting the sector as a cohort. Because microfinance institutions on NEPSE are disproportionately small-cap and often have modest free floats, the selling pressure that followed the regulatory news overwhelmed available buy-side interest in scrip after scrip. Multiple microfinance names spent several consecutive sessions locked at their lower circuits, and the sector as a whole underperformed the broader index by a wide margin over the following months.

This second case study is instructive for a different reason than the first: it shows that circuit-trap risk is not purely a function of an individual company's specific fundamentals. An investor can do thorough, careful bottom-up research on a single microfinance company, conclude correctly that its loan book and management quality are sound, and still find that scrip circuit-locked for days at a time — not because the company itself did anything wrong, but because it was correlated, through sector membership, with other companies that triggered the initial selling wave. This is a form of correlation risk that pure fundamental analysis does not capture, and it is precisely why the liquidity-engineering discipline built across this entire Part (Chapters 54 through 58) has to sit alongside fundamental analysis rather than being treated as a secondary concern.

Both case studies point to the same underlying asymmetry that this chapter's brief specifically asked to be addressed, and it is worth stating plainly because it is easy to underweight emotionally while a rally is underway: entering a position in NEPSE, especially when buying into strength during a rising or momentum-driven market, is almost always fast. Orders execute near-instantly because sellers are plentiful — everyone wants to sell into a rally at a good price, so a buyer's order finds a counterparty within seconds. Exiting a position, especially when the stock is falling and thinly held, can be dramatically slower — not because the investor is indecisive, but because the other side of the trade, the buyer, may simply not exist in adequate size at the price the seller needs. This asymmetry is structural, not a matter of skill or timing, and it is the single most important intuition this chapter is trying to build: the ease of getting into a position tells you nothing about the ease of getting out of it, and the two should never be assumed to be mirror images of each other.

Lesson 58.6 — Mitigation: Sizing, Staged Exits, and Selling Into Bounces

Everything in this chapter so far has been diagnostic — understanding how the circuit trap works mechanically, how it cascades, how to model its effect on exit timelines, and how it has actually played out in NEPSE's history. This final lesson turns to what an investor can actually do about it, and the honest starting point is that there is no mitigation technique that eliminates circuit-trap risk entirely. Once a stock is genuinely locked with no buy-side interest, there is very little any individual holder can do in that moment. The real mitigation has to happen earlier — in position sizing before the fact, and in exit discipline during the early stages of a decline, before the trap fully closes.

The first and most important mitigation is the one this Part has already built across two prior chapters: position sizing discipline relative to ADV (Chapter 56) and relative to free float (Chapter 57). Everything modelled in Lesson 58.3 confirms why this matters specifically for circuit-trap risk — a position sized at 0.25x–0.5x ADV has a circuit-trapped days-to-exit measured in weeks, which is painful but survivable; a position sized at 5x ADV has a circuit-trapped days-to-exit measured in well over a year, which for most investors is not survivable, either financially or psychologically. The single highest-leverage decision an investor makes about circuit-trap risk happens before the position is even fully built, at the moment they decide how large a stake to accumulate relative to the stock's demonstrated trading liquidity.

The second mitigation is staged or partial exits, initiated when a thesis first shows cracks rather than only after it has fully broken. Many investors, reasonably enough, want confirmation before selling — they wait for a clear signal that the investment case has failed before acting, to avoid selling a good position on a false alarm. The circuit-trap dynamic punishes this instinct specifically, because by the time the thesis has "fully broken" in a way that is unambiguous, the stock is often already several sessions into a lower-circuit cascade, and the investor's remaining position is the hardest part to exit — the tail end of a queue that has been building for days. A more robust approach is to treat deterioration in a thesis as a dial, not a switch: trim 20–30% of a position on the first serious warning sign (a broken support level, a negative sector development, a disappointing quarterly disclosure), trim further if conditions continue to worsen, and reserve full-position conviction only for theses that are still fully intact. This does mean occasionally selling a position that goes on to recover — that is the acknowledged cost of the discipline, and it is a cost worth paying against the alternative of being fully trapped in a cascading decline.

The third mitigation, and the one most specific to NEPSE's circuit-band structure, is pre-committing to sell into any bounce or upper-circuit day that occurs during an established downtrend, rather than holding out for a full trend reversal. This runs against a natural instinct — an investor watching a position recover 10–15% in a single session understandably wants to believe the worst is over, and selling into that relief rally can feel like giving up gains just as the stock turns around. But in a market with a hard daily price band, a single strong up-day inside a broader downtrend is often exactly the moment when buy-side liquidity is temporarily healthiest — which is to say, precisely the moment when an exit is actually executable, as opposed to the days when the stock is grinding toward its lower circuit and liquidity is thinnest. An investor who has already decided, in Lesson 58.5's terms, that a position's risk-reward has deteriorated should treat a bounce day not as a signal to hold on for more upside, but as a rare liquidity window to use.

None of these three mitigations — sizing discipline, staged exits, and selling into bounces — requires predicting when a circuit trap will occur. That is the point. An investor cannot reliably forecast which stock will lock lower-circuit next, or for how many sessions, any more than they can forecast the next sector-wide regulatory shock. What an investor can control is how much capital is at stake relative to demonstrated liquidity, how early they begin reducing exposure once a thesis starts to wobble, and whether they use the liquidity windows the market actually offers rather than the ones they wish it would offer. Circuit-trap risk cannot be eliminated in a market structured the way NEPSE is structured. It can be sized down to a survivable level, and that is the realistic and sufficient goal.