Liquidity vs Volatility: Two Properties, Not One Number

Liquidity is how much can be transacted without pushing the price about. Volatility is how much the price has already moved around its own average over a stated period. One is a property of transacting, the other a property of a price series. Liquidity and volatility share no inputs, and all four combinations of high and low occur, so neither figure reveals the other.

Both of these are covered separately. How much of a holding can actually change hands, and what happens to the quoted price when more is moved than the day can absorb, belong to the liquidity measures. Volatility takes what a movement figure measures and, more usefully, what it refuses to measure. Set side by side, the two turn out not to be two readings of one thing.

Why does the distinction need setting out at all? Because in ordinary conversation both get called risk, both arrive as a single number with a percentage sign or a rupee sign attached, and both feel like they are answering the question how dangerous is this. The shared feeling of danger does a lot of quiet damage, and it costs real money when a holding turns out to be one that cannot be exited at the price that made it look sensible.

What does liquidity mean, before volatility is mentioned at all?

Liquidity is the ability to transact a stated size, near the price currently quoted, inside a stated stretch of time. Every one of those three parts has to be there. A share is not liquid or illiquid in the abstract, the way it might be large or small. Liquidity is always relative to a size, and it is a property of the holding being moved rather than a property of the company.

Think of a household with two things of equal stated worth: money sitting in a savings account, and a set of gold bangles. On paper they might both be worth two lakh rupees. But one of them turns into money this afternoon at exactly the figure printed on the statement, and the other one turns into money only after somebody is found, a weight is agreed, a making charge is argued about, and a discount is swallowed. Nobody says the bangles are worth less. The household says the bangles are harder to move. Ease of moving something, rather than its stated worth, is the whole idea, and shares behave in the same way.

Three quantities do most of the work in assessing it, and every one of them is counted rather than modelled. There is the value transactedThe rupee worth that went through over a period, with each parcel counted at the price it actually went through at. Read off the exchange record rather than estimated. in a day. There is the part of that day which was taken to deliveryThe portion of a day where the shares are actually moved into the buyer's account, rather than bought and sold back inside the same session. Reported separately by the exchanges., meaning the part where the shares genuinely moved into somebody's account rather than being bought and sold back within the session. And there is the quantity sitting visibly near the current price. Market depthThe quantity waiting to transact within a narrow band of the current price at any given moment. A separate quantity from a whole day of activity, and covered separately. is the name for that quantity. Depth is set out separately and used here rather than rebuilt.

What does volatility mean, before liquidity is mentioned at all?

Volatility is the dispersionHow widely a set of numbers is spread out around its own centre. A set can have the same average as another and be far more spread out, and dispersion is the word for that difference. of returns about their own average, measured over a stated period, and measured symmetrically. Symmetrically matters: a day up four per cent and a day down four per cent contribute exactly the same amount to the figure. The measure has no opinion about direction. Nobody calls a four per cent gain a risk that materialised, and that is why calling the figure risk is such an awkward fit.

Volatility is a property of the price series and of nothing else, and no holder enters it. Two people can hold the same share in wildly different sizes and face wildly different liquidity problems. Both read the identical volatility figure. The record of prices it was computed from contains neither of them.

The everyday version sits in any ordinary market. The vendor's tomato price wanders about week to week: some weeks it is thirty rupees a kilo, some weeks it is ninety. A packet of salt on the shelf behind him has cost the same amount for a year. The difference in how far the number wanders is what volatility measures, and it says nothing whatever about how easy either one is to buy.

Why do these two keep getting mistaken for one another?

Four reasons, and they stack. The first is that both get filed under the single word risk in ordinary speech, and once two things share a label people stop looking for the seam between them. The second is that both arrive as one number, so they look interchangeable on a screen where every column holds one number. The third is that both feel like answers to one silent question: how much trouble could this cause me.

The fourth reason contains a grain of something true, and that grain is what does the real damage. Thin shares that also jump about are genuinely common, and the reason is easy to see. When very little is waiting to transact near the current price, a single ordinary order has to reach further to find the other side, and the printed price moves further to get there. The link between thinness and movement is the strongest reason the two get confused, and it is still not a rule.

The link is not a rule. Thinness pushes movement in one direction only, on some shares, sometimes, and how often is not established. A thin share can also sit perfectly still for weeks. A heavily transacted share can move about violently. Neither of those is unusual, and the next section shows all four possibilities on the same company.

Can a share be liquid and volatile at the same time?

Yes, and it can be any of the other three things too. Set the two properties on separate axes and four regions appear, and every one of them is occupied by real shares in every market.

A share can be liquid and volatile: enormous quantities change hands and the price still swings hard. A heavily followed company looks exactly like that on the morning it reports something surprising. A share can be liquid and stable: a great many people are transacting, none of them is learning anything new, so enormous quantities change hands and the price barely moves. A share can be illiquid and volatile: hardly anything transacts, and each thing that does transact shifts the printed price a long way, for the reason given above.

The fourth combination, illiquid and stable, is the one worth slowing down for. A price that has not moved because nothing transacted is not a stable price in any useful sense, and this corner is where the difference between the two properties becomes impossible to argue with.

The everyday version of that fourth corner is a plot of land in a small town where three houses were sold in the last four years. Ask what a plot is worth and everyone quotes the same figure they quoted in the year before, and the year before that. The number has not moved. Nothing has been settled either; there simply has not been a transaction to move it. Calling that stability is a description of the paperwork, not of the plot.

What do both look like on one company, with nothing shared?

Take Sarvani Coatings Limited and set out both properties, using only its own illustrative record as at 28 August 2026, and stopping short of a verdict on either. Laying them out together reaches no conclusion. Each row shows what it was built from, and that is the thing worth noticing.

Read the right hand column. Eight of those ten rows were counted from quantities that changed hands, two were computed from a series of prices, and not one input appears on both sides of that split. Every price in the record could be deleted and the first eight rows still rebuilt from the exchange volume file; every volume could be deleted and the last two still rebuilt from the price history. The split is not a coincidence of this company; it is what the two measures are.

The rest of the record sits around those figures and does not disturb the split. Sarvani Coatings has 24.00 crore shares in issue at Rs 486/-, so its capitalisation is Rs 11,664 crore, and its free floatThe slice of an issuer sitting outside the promoter group, and therefore the slice available to change hands. Both its definition and its disclosure come from the rules named below. of 47.6 per cent gives Rs 5,552 crore. A day of Rs 42 crore is about 0.76 per cent of that float. Every one of those is still a quantity or a price, never a mixture, and none of them touches the 28.0 per cent.

Can one of them move and leave the other exactly where it was?

One test settles the argument, and it can be run on the figures above. The test runs twice, once in each direction.

First, hold Sarvani Coatings' price path exactly as it is, and imagine the value transacted was a tenth of what it is: Rs 4.2 crore a day instead of Rs 42 crore. Every closing price is untouched, so the daily figure is still 1.77 per cent and the annualisedRestated onto a one year footing so that figures measured over different lengths of time can be put beside each other. Restating a daily figure onto a one year footing is set out under volatility. figure is still 28.0 per cent, to the last decimal. The Rs 250 crore holding, meanwhile, has gone from about 5.95 days of everything that transacts to about 59.52 days of it. At 59.52 days it is no longer a holding that can be taken.

Now the other direction. Hold the value transacted at Rs 42 crore and imagine a price path that wandered half as far each day: 0.885 per cent daily instead of 1.77 per cent. Half of 1.77 per cent annualises to about 14.0 per cent. The liquidity picture has not moved by a rupee. The same Rs 42 crore transacts, the same 18,000 shares sit near the price, and the same holding still takes about 5.95 days.

Being able to move each figure to a new value while the other one stays at the identical number is the demonstration that they are two separate properties, not two views of one. If they were the same underlying thing wearing different clothes, this would be impossible.

Which one limits what can be done next, and which only reports?

Here is the organising idea, and it is worth more than any of the arithmetic above. Liquidity constrains what can be done. Volatility describes what already happened to the price. Neither can ever stand in for the other in a decision.

Volatility is entirely backward looking in its construction. Every input is a price that has already printed. When somebody says a share is a twenty eight per cent share, the whole of that statement is about a stretch of time that is finished. The figure may be useful for framing what comes next, and people do use it that way, but the object itself is a summary of a record.

Liquidity is used forward. Nobody assesses liquidity out of historical interest. People assess it because there is something they want to do, at a size, within some period, and the assessment answers whether that thing is available to them. The past activity is only the evidence; the question is about a transaction that has not happened yet.

What does it cost to let one number answer both questions?

Two errors follow from conflating them, and they are mirror images. The first: a reader treats a low volatility figure as evidence that a share is easy to get out of, and sizes a holding accordingly. Nothing that went into the figure was a quantity, so the figure never supported that sizing. The holding turns out to be one that takes weeks to unwind rather than days.

The second: a reader treats a thin share as safe because the printed price has been steady, and has therefore mistaken an absence of transactions for an absence of movement. The small town land plot is back. Evidence of an absence of transactions looks exactly like evidence of a genuinely settled price, and that resemblance is what makes the mistake so comfortable.

Both errors run in the direction of overconfidence, never caution, and the one-sidedness is precisely what makes them worth naming out loud. A mistake in the direction of too much caution would surface as an opportunity missed and would sting for a week. The two errors above surface as a holding that cannot be left.

What does a fund manager do with each of the two, on a working day?

Watch where each figure actually lands in the working day and the difference stops being an abstraction. A manager running a pool of money uses the liquidity assessment before the holding exists. The question is: at the size this pool would need, how many days of everything that transacts is that, and is the answer acceptable given the mandate she has written? For Sarvani Coatings and a Rs 250 crore holding, the answer is about 5.95 days of everything, and about 18.60 days if she restricts herself to the part that goes to delivery. The 28.0 per cent cannot tell her how many days anything takes, so it never enters that conversation.

The volatility figure lands somewhere completely different. The figure appears in the reporting pack, in the risk committee papers, and in the conversation about how the pool behaved against what she told her investors to expect. Every one of those conversations is about a period that has closed. She may use it to frame what she says about the period ahead, but she is doing that by choice and by argument, not because the figure reaches forward on its own.

A research analyst uses the split the same way and earlier. Meghna Iyer, deciding whether to start covering a company at all, checks what transacts first. Six weeks spent on a business whose shares nobody can take a position in are six weeks wasted. The movement figure has no bearing on that decision. The movement figure belongs later, in the section of the note that describes how the share has behaved, where it is a description and is labelled as one.

And the household version is the one everybody already knows. Money in a savings account has a stable value and can be reached this afternoon. A plot of land has a value that has not moved in years and cannot be reached for months. The second one is stable and impossible to reach at the same time, and nobody in the household would be confused about it for a second. The confusion only appears once both properties are reduced to a percentage and printed in adjacent columns.

How are the two kept apart, every time?

The discipline is small, and it is closer to a reflex than to a method. Ask the two questions separately, in words, before either number is looked at. How much can be transacted, at what size, in what stretch of time? And what has this price done, over what period? The two questions have different denominators and different periods, and a figure built for one of them cannot be reshaped into the other. Never let one figure be the answer to both.

Look at what the denominators actually are. A liquidity figure is a size over a quantity that transacts: rupees over rupees a day, giving days. A volatility figure is a spread of returns over a stated stretch: a percentage a day, restated onto a year. There is no arithmetic that converts one into the other, no matter how the units are pushed around, and that impossibility is not a limitation of the measures; it is the reason both exist.

One last habit worth building. When a source presents one of the two figures and calls it risk, the translation is worth making before reading on: this is a figure about quantities transacted, or this is a figure about how far the price wandered. The translation takes a second, and it stops most of the damage. Almost every version of that damage begins with a reader accepting the word risk and never asking which of the two things was meant.

Sarvani Coatings Limited, Nandivarman Paints Limited, Kesaria Surface Solutions Limited, Thottam Chemicals Limited and Meghna Iyer are invented.
Educational material. Not advice on any investment, tax, budget or market position.