A small hydropower stock hits its upper circuit three sessions in a row. The trading groups light up. Someone posts a floorsheet screenshot, someone else names a broker, and by evening the verdict is in: operators are pumping it. Maybe they are. But there is a duller explanation that fits the same facts, and most retail investors never check it. The stock simply has very few shares available to trade.
That is what free float on NEPSE measures: the slice of a company’s shares that can actually change hands in the market on a normal day. Not the paid-up capital. Not the market cap printed on the portal. The tradable slice. And on NEPSE that slice is often a small fraction of the whole company.
Here is this article’s position. Low free float explains a large share of the volatility that Nepali retail investors blame on manipulation. Manipulation is real, and thin float is exactly what makes it cheap. But if you only ever ask “who is pumping this?” and never ask “how many shares are actually out there?”, you will misread price moves, overpay for scarcity, and get blindsided when the float suddenly expands.
What free float on NEPSE actually means
Every listed company has a total number of shares. On NEPSE, that total is split into groups that behave very differently.
Promoter shares belong to the founders and sponsoring institutions. They are locked in for a period after the IPO and, in many companies, trade separately from ordinary shares even after that. Government holdings, where they exist, rarely move. Shares allotted to employees and to residents of a hydropower project’s affected area are typically locked in alongside the promoters. What is left, the ordinary shares held by the general public, is the pool the market actually trades.
Free float is the size of that tradable pool, usually expressed as a percentage of total listed shares:
Free float (%) = (Total listed shares − promoter, government and locked-in shares) ÷ Total listed shares
The concept is not exotic. Index providers worldwide adjust for it. NEPSE itself publishes a Float Index and a Sensitive Float Index alongside the headline NEPSE Index. As ShareSansar has explained in its guide to the NEPSE index family, the float indices exclude shares held by promoters, the government and board members, but NEPSE does not have a method for measuring how much of the public’s holding actually trades, so the float indices still count all ordinary shares. Keep that limitation in mind. It means even the official “float” measure overstates how much stock is really available. More on that below.
If you are still mixing up how many shares exist with what a company is worth, start with our explainer on market cap versus paid-up capital. Free float is the next layer down: of all those shares, how many can you actually buy?
Why NEPSE floats are so thin
Three features of Nepal’s market keep the tradable slice small.
The first is the promoter-public structure. Nepali companies are built with a large promoter block by design. Banks and microfinance institutions have been moving toward a 51:49 promoter-to-public ratio, with each conversion approved by Nepal Rastra Bank, as a long series of company notices reported by ShareSansar shows. That still leaves roughly half the company in promoter hands. In hydropower, the public portion at IPO is often smaller still.
The second is state ownership. Nepal Telecom is the clearest case. According to Kantipur’s reporting on the FY 2083/84 budget, the government holds 91.49 percent of the company, the Citizen Investment Trust 0.03 percent, and the general public and employees 8.48 percent. One of the largest companies on the exchange by market value trades on a public slice of under a tenth. The budget announced in May 2026 proposes cutting the government’s stake to 66 percent through a sale to the public. At the time of drafting, available reporting showed the sale still at the planning stage, with no public offering open yet.
The third is the lock-in. Promoter, employee and project-affected local shares cannot be sold for a set period after the IPO. Recent company notices on ShareSansar, citing SEBON guidelines, typically describe that period as three years from IPO allotment, though at least one hydropower notice, from Mid Solu Hydropower, described a two-year lock-in. During that window, a large part of the company simply does not exist for trading purposes.
A worked example: one hydropower company, two very different floats
Abstract percentages hide how extreme this gets. Take a real case.
Three Star Hydropower (TSHL) published a notice, reported by ShareSansar on June 1, 2026, that the lock-in on its promoter, project-affected local and employee shares would end on Ashadh 19, 2083 BS (early July 2026 AD). The notice gave the breakdown. TSHL had 4,925,000 shares listed on NEPSE. Promoters held 3,693,750 of them. Project-affected residents held 492,500. Employees held 14,775.
Do the subtraction. Before the lock-in ended, the shares held by the general public came to 723,975. That is about 14.7 percent of the company. At the closing price ShareSansar reported that day, NPR 602.50, the entire publicly tradable slice was worth roughly NPR 43.6 crore, while the company’s full listed value was close to NPR 297 crore.
Now think about what that means for price discovery. For three years, the market valued the whole company at NPR 297 crore based on trading in shares worth about NPR 44 crore. Every rupee of buying and selling was competing for one share in seven.
Then the lock-in ends. In principle, the shares available to trade can jump from about 7.2 lakh to well over 40 lakh, close to six times the old pool. The notice itself added an important limit: shares held by the board of directors and senior management remain non-tradable after the lock-in. So the real increase is smaller than the headline. But the direction is unmistakable. The stock that traded like a scarce asset is about to stop being scarce.
This is not a prediction about TSHL’s price. Promoters may not sell. But anyone who assumed the old, thin float was permanent bought a different stock from the one that exists after unlock.
How a small float turns ordinary demand into big price moves
The mechanics are simple once you see them.
Imagine two NEPSE companies, each with one crore shares listed and each trading at NPR 400. Company A has a free float of 50 percent, so 50 lakh shares trade. Company B has a free float of 7 percent, so 7 lakh shares trade. These are illustrative numbers, not real companies.
A wave of new interest brings NPR 2 crore of fresh buying into each stock. At NPR 400 a share, that money wants about 50,000 shares.
In Company A, 50,000 shares is 1 percent of the tradable pool. Sellers can absorb that without much drama. The price moves a little.
In Company B, the same 50,000 shares is about 7 percent of everything that can trade. Most holders are not sellers at today’s price, so buyers have to climb the order book to find stock. The price gaps up. If it gaps far enough it hits the daily limit, which on NEPSE is now 15 percent either way after the amendment to the Securities Trading Operation Regulations took effect in April 2026, as ShareSansar and Himal Press reported. Unfilled buy orders pile up behind the limit. The next session opens with that backlog, and the move continues.
Nobody needs to coordinate anything for this to happen. The same NPR 2 crore that barely moves Company A can send Company B through its circuit for days. From the outside, that looks exactly like an operator campaign.
The reverse is also true, and it is the half retail forgets. When sentiment turns, the thin float works in the other direction. Few buyers stand underneath. A modest amount of selling pushes the price to the lower limit, and holders who want out find no one to sell to. Our explainer on how liquidity works on NEPSE covers why exit risk matters more than entry price in names like these.
So is it manipulation or is it float?
Often it is both, and the honest answer is that thin float is the precondition for most manipulation.
It costs far less to move a stock when only a small slice of it trades. A group that wants to engineer a breakout in a company with 50 lakh tradable shares needs serious capital. In a company with 7 lakh tradable shares, the same campaign is a fraction of the cost. So when you see a suspected pump, the low float is not a separate issue from the manipulation. It is the reason the manipulation was affordable.
But that cuts both ways for how you think. If you assume every sharp move in a small-float stock is an operator, you will miss the many cases where it is just ordinary demand meeting very little supply, after a good quarterly report, a dividend announcement, or a sector-wide rerating. You will also miss the opposite case, where a stock looks calm and “safe” only because almost nobody is trading it.
The better question is not “who is behind this?” It is “how much stock is actually available, and how much money is chasing it?” You can answer the second question from public data. You can almost never answer the first. If you want to test whether a move is broad or concentrated, learning to read a NEPSE floorsheet is the right next step.
The real float is smaller than the official float
Even the “public” shares are not all tradable in practice. Some of them belong to long-term holders who never sell. Some sit with mutual funds and institutions that trade rarely. Some are old physical certificates that were never dematerialized and so cannot be sold through TMS at all. Some belong to IPO allottees who applied for ten units years ago and forgot about them.
None of them count as locked in on any official breakdown. They behave as if they were.
This is exactly the gap ShareSansar pointed to in NEPSE’s float indices: the exchange can strip out promoter and government shares, but it has no method to measure how much of the public holding actually trades. So the official float is an upper limit. The effective float, the shares that genuinely turn over, can be much smaller.
We cannot give you a precise figure for any company here, because no public source measures it. But you can estimate it yourself with a crude but useful check. Compare a stock’s average daily traded volume with its public shareholding. If a company has 10 lakh public shares and trades 2,000 shares on a typical day, the active pool is far smaller than 10 lakh. That is an estimate, not a fact, and you should treat it that way. It is still better than assuming the whole public slice is live.
Lock-in expiry: the float event retail ignores
If thin float pushes prices up, a float expansion is the event that can bring them down. On NEPSE, the most predictable float expansion is the end of a lock-in period.
The dates are not secret. Companies publish lock-in expiry notices, and ShareSansar maintains a promoter lock-in list. Yet many retail investors buy small hydropower and microfinance stocks without ever checking when the promoter and local shares unlock.
There is a structural reason this matters more in some companies than others. In many hydropower companies, promoter and public shares have shared the same identifier, so once the lock-in ends, the formerly locked shares can be sold through the market. CDSC proposed a directive in 2025 that would give promoter and ordinary shares separate identification numbers and stop automatic conversion after lock-in, but it ran into opposition from the Nepal Chamber of Commerce and power producers, as Kantipur reported. In July 2026, Ratopati reported that SEBON chairman Gopal Bhatta had signaled the board would not advance the dual-identifier system, arguing it does not follow international practice. So, for now, the existing structure stays. Our deeper look at how promoter shares trade and convert on NEPSE covers the conversion rules and the discount promoter stock tends to carry.
Unlock does not guarantee selling. Promoters who believe in the project may hold for years. Others, who funded construction and have waited a long time for liquidity, may be eager to sell. You cannot know in advance which group you are dealing with. What you can know is the date, and the size of the supply that could arrive on it. A stock that has traded on a 15 percent float for three years and is about to open up to several times that pool is carrying a risk the chart does not show.
The same logic runs the other way for Nepal Telecom. If the government does sell down to 66 percent, the public float would expand from under 9 percent to roughly a third of the company. That would make NTC a more liquid, more index-relevant stock. It would also add a large amount of new supply. Both change how the stock trades, and both are knowable before they happen.
How to check free float before you buy
- Find the shareholding split. Company profiles on NEPSE and the major portals show promoter and public percentages. Annual reports and IPO prospectuses give the detail. Write down the public percentage.
- Adjust for locked-in shares. For companies that listed recently, check whether promoter, employee and local shares are still under lock-in, and when that ends.
- Compare float with daily turnover. Look at average traded volume over a few weeks against the public share count. A tiny ratio tells you the effective float is thinner than the official one.
- Mark the calendar. Note lock-in expiry dates, planned FPOs, government divestments, and right share issues. Each one changes the float.
- Size your position to the float, not your conviction. In a thin-float stock, assume you may not be able to exit quickly at the price on your screen. Hold less of it than you would of a liquid name.
The verdict
Free float is one of the most useful numbers on NEPSE and one of the least checked. It explains why some stocks can run through their circuit on modest news, why others look stable only because nobody trades them, and why a company’s “value” on a portal can rest on trading in a sliver of its shares.
The takeaway is not that thin-float stocks are bad. Some are good businesses with patient owners. The point is that a thin float changes what the price means. A price set by trading in 15 percent of a company is a weaker signal of value than a price set in 60 percent of it. It moves further on less information, and it can be overturned by a supply event printed in a notice months in advance.
So before you blame operators for the next wild move, run the float check. Sometimes you will find manipulation. More often you will find a small pool of shares and a lot of money chasing it. And occasionally you will find a lock-in expiry date sitting a few weeks away, which is the one thing in this market you were told about in advance.
This is analysis, not financial advice.
