Liquidity Requirements: What Must Be Available When

A liquidity requirement states what cash a holder must have available and on what date, and it constrains a portfolio before any view about markets does. A requirement of that kind is a claim about timing rather than about value: a portfolio can be worth a great deal and still fail one. Meeting a call also moves every remaining weight. A second constraint quietly starts to bite there.

Most readers meet this idea as a sentence in a policy statement: the portfolio will keep enough in liquid form to meet foreseeable needs. The sentence sounds settled because there is a cash line in the allocation table. A liquidity requirementHow much cash the holder must be able to produce, and by when. An obligation in the mandate rather than a preference. is only a constraint once somebody has worked out at what size of payment meeting it starts to break something else in the same document.

The running example is the Anantara Multi-Asset Portfolio, an invented discretionary mandate of Rs 500 crore run by Faiz Ahmad Ansari for an invented charitable endowment whose committee is chaired by Rukmini Deshpande. Its stated shape is equity Rs 300 crore, fixed income Rs 150 crore and cash Rs 50 crore. On Rs 500 crore those are 60.0, 30.0 and 10.0 per cent. The mandate permits equity between 50 and 70 per cent of the portfolio.

The record carries no spending rate for this endowment and no payment schedule. So the number of years the cash sleeve lasts cannot be computed here: the two inputs that arithmetic needs are not in the record, and inventing them would teach a false precision rather than the mechanism. The record does support the size of a payment, so everything below works in call sizes: given a call of this size, what happens to the shape of the portfolio?

Three things a mandate can write down behave differently when a payment lands. A fixed rupee cap stays exactly where it is. A share of the total holds its percentage and moves in rupees the moment the total moves. An obligation on a date does something neither of the others does: it moves the total itself. Only the third kind changes the denominator, and a change in the denominator drags the other two along behind it. The obligation on a date has to be settled first.

What does a liquidity requirement actually state?

Two facts, and it is not a requirement until both are present. An amount, and a date. The pair sounds too simple to be worth a heading until it becomes clear how often a policy statement writes down only one of the two.

Take the amount on its own. A paragraph saying the endowment will need Rs 50 crore is a fact about size. Given long enough almost anything can be turned into Rs 50 crore, so every portfolio satisfies it eventually. No portfolio can fail it. A date on its own constrains nothing either: until somebody writes a figure beside the last day of the third quarter, there is no quantity to be ready with. The requirement is the pair rather than either half of it.

The everyday version. A household knows it will need money for a wedding; that is not yet a plan. The household knows the wedding is in November; that is not a plan either. The plan begins at the sentence carrying both: about Rs 8,00,000/- has to be in the bank by the first week of November.

Each half is a different kind of object. The date belongs to a calendar and sits outside the portfolio entirely: a building instalment, a grant the endowment has committed to, a fee that falls due whether the market is calm or not. The amount belongs to the obligation rather than to the holdings. Neither half is a statement about prices. A liquidity requirement can therefore be settled before anybody has an opinion about markets.

How is this different from the portfolio simply falling in value?

These are not the same kind of event. One is about what the holdings are worth. The other is about what the holder has promised to hand over. A fall and a call can arrive in the same week, they have different remedies, and confusing them is how a liquidity paragraph ends up written as a risk paragraph.

Put the two side by side. Inside the stated twelve month period the portfolio recorded a worst drawdownThe largest fall from a high point to a following low point inside a stated window. Move the window and the figure moves. of 9.7 per cent, against 8.1 per cent for the unnamed composite benchmark over the same window. On Rs 500 crore that is Rs 48.5 crore of value, very nearly the whole cash sleeve. And it required no cash at all: not one rupee had to leave the portfolio, and the committee's obligation to anybody was unchanged.

Now the other event. A payment of Rs 50 crore falls due on a stated day. Whether the portfolio rose 14.2 per cent that year or fell 9.7 per cent from its peak makes no difference: Rs 50 crore has to leave the portfolio and reach somebody else's account. A payment of that kind is a cash callA payment the holder must make out of the portfolio on a stated date. It reduces the total rather than the value., and its remedy is not patience. Patience answers a fall in value. A fall is a change in what somebody would pay today, and today is not the only day. Patience is no answer to a date.

One more thing about that 9.7 per cent. Take a different twelve months around the same path and the peak and the trough move, and the figure moves with them. The moving peak and trough are why the window is quoted every time the figure appears, and a drawdown offered without its window is not a measurement anybody can check. How a drawdown behaves as a risk measure belongs to the monitoring sequence.

How Liquidity Needs Affect Portfolio Design, and what has to be settled first?

Liquidity needs set a floor. Nobody can promise a floor on value; the floor is on how much of the portfolio has to sit in a form that can be turned into cash by the date on the requirement. The floor is decided when the sleeves are decided, before a single holding has been chosen. A liquidity requirement is a design constraint rather than a monitoring one: it is satisfied by the shape of the portfolio, not discovered in a report the week the payment falls due.

The sequence matters. Work out what has to be paid and when. Size the part that answers it. Here that part is the cash sleeveThe part of a portfolio kept in cash or in something behaving like cash, so a payment can be met without selling. at Rs 50 crore, or 10.0 per cent of Rs 500 crore. Then let the rest of the design happen inside what is left. Reverse those steps and the liquidity paragraph describes a portfolio that already exists. A description of a portfolio that exists is not a constraint on one about to be built.

Now the honest part. The obvious question to ask of a Rs 50 crore cash sleeve is how long it lasts: the sleeve divided by what the endowment spends in a year. The division cannot be done here. The record locks no spending rate and no payment schedule. Where a record locks nothing, say so and change the question rather than supply a figure that looks like knowledge. So the question becomes the one the record does answer: given a call of a stated size, what shape is the portfolio left in?

A second figure people reach for belongs to a different question. Over the stated twelve months the portfolio recorded turnoverThe share of a portfolio replaced by trading over a stated period. It measures activity rather than obligation. of 34 per cent. On Rs 500 crore that is Rs 170 crore of traded value, or 3.4 times the entire Rs 50 crore cash sleeve. There is nothing to conclude from that. Turnover measures how much trading the mandate did and the cash sleeve is sized against what has to be paid out, so setting one against the other compares two quantities that answer different questions and happen to share a unit.

What happens to every other weight when a cash call is met?

This runs the opposite way to intuition. Meet a call from cash and the cash falls by the amount paid. Nothing else moves in rupees. But the total falls by exactly the same amount, and every remaining sleeve is now a share of a smaller total. So every non-cash weight rises, and it rises without a single unit of any risk asset having been bought.

Work it with a call of Rs 50 crore. Cash goes from Rs 50 crore to nil. Equity is untouched at Rs 300 crore and fixed income at Rs 150 crore, so the total is Rs 450 crore. The equity weight is Rs 300 crore over Rs 450 crore, or 66.7 per cent. The cash was what got spent, so equity sits six and a half points closer to the 70 per cent equity ceiling.

Name what did the work. A weight is a fraction. Meeting a call from cash leaves the numerator where it was and shrinks the denominator. Every weight moved and the only thing that changed was the number underneath the line. A shrinking denominator is why a report can show the equity share up sharply in a week when the dealing desk did nothing, and why the movement is not evidence about anybody's view on equity.

The same arithmetic at a size a reader can hold in their head. A household has Rs 10,00,000/- put by: Rs 6,00,000/- in shares, Rs 2,00,000/- in a savings account and Rs 2,00,000/- in a bank deposit, so shares are 60.0 per cent. A school fee of Rs 2,00,000/- comes out of the savings account. The savings account was there for exactly that. The shares are untouched at Rs 6,00,000/-, the total is Rs 8,00,000/-, and the shares are 75.0 per cent. Nobody bought a share, and the household is more exposed to the equity market on Tuesday than on Monday because it paid a school fee.

Does the equity weight simply keep rising as the call gets bigger?

No, and this is where policy work quietly goes wrong. Push the call past Rs 50 crore and there is no cash left. Every further rupee comes out of something that was not cash, and this mandate's order of sources sends it to equity. Equity now falls in rupees and the total falls with it, so the weight that had been climbing turns and comes back down.

At a call of Rs 100 crore, Rs 50 crore came from cash and Rs 50 crore from selling equity, so equity is Rs 250 crore against a total of Rs 400 crore, or 62.5 per cent. At Rs 125 crore, equity is Rs 225 crore against Rs 375 crore, or 60.0 per cent, so the portfolio has come back through its own policy weightThe share the committee decided each sleeve should carry. It is a chosen point, and the actual share drifts away from it. on the way down. At Rs 200 crore, equity is Rs 150 crore against Rs 300 crore, exactly 50.0 per cent. The path has a peak in the middle rather than an end, and a policy tested only at the largest call imaginable walks straight past where the strain is greatest.

The turn is not a quirk of these particular numbers. While cash is being spent, the numerator of the equity weight is frozen and only the denominator falls, so the fraction rises. Once cash is exhausted the numerator falls too, and faster in proportion than the total. Equity is carrying the whole reduction while fixed income sits still. The turn happens exactly at the rupee where the cash sleeve runs out.

One consequence catches committees out. While the equity weight travels, the fixed income weight climbs steadily and never turns: 30.0 per cent at the start, 33.3 per cent once the cash has gone, 40.0 per cent at Rs 125 crore, and 50.0 per cent at Rs 200 crore, where it meets equity coming the other way. A sleeve nobody touched has ended up carrying two thirds more of the whole than the committee wrote down for it.

How large a call does it take before another constraint is touched?

Here is the number a committee needs, and it is almost never in the liquidity paragraph. Met from cash first and then from equity, the answer is Rs 200 crore: equity at Rs 150 crore against a total of Rs 300 crore, exactly the 50 per cent range floorThe lowest share a sleeve is permitted to fall to under the mandate. Going under it is a breach whether anybody traded or not. the mandate wrote for equity, and 40 per cent of the whole portfolio. The Rs 200 crore turns a paragraph about intentions into a constraint that can be checked, and it falls out of the allocation table with nothing added.

The ceiling, on the other hand, cannot be breached this way. The highest the equity weight reaches along the whole path is the 66.7 per cent peak at a call of Rs 50 crore, against a ceiling of 70 per cent. The gap is 3.3 percentage points of clearance at the worst point, worth checking rather than assuming.

A second figure comes free. The headroom left at any moment is Rs 200 crore less whatever has already been called: Rs 150 crore after a Rs 50 crore payment, Rs 75 crore after Rs 125 crore. A committee that carries one number into the meeting and subtracts from it has a live liquidity constraint rather than a paragraph re-read once a year.

The same finding reads more usefully in rupees. A range written as a percentage is a rupee band that moves whenever the total moves. At Rs 500 crore the 50 to 70 per cent range permits equity between Rs 250 crore and Rs 350 crore. Meet a Rs 50 crore call and the total is Rs 450 crore, so the band is Rs 225 crore to Rs 315 crore, and the untouched Rs 300 crore of equity has Rs 15 crore of headroom under the ceiling. The band walked towards the holding, and no dealing desk was involved.

The Rs 200 crore was the answer for this mandate's order of sources, cash and then equity. Change the order and the answer changes. Sell only equity and leave the cash alone, and the floor arrives at Rs 100 crore instead, exactly half the size. Take it pro rata from all three sleeves and every weight stays where it was, so no wall is reached at any call size. Take it from cash and then from fixed income, and the portfolio runs at the ceiling rather than the floor: at a call of about Rs 71.4 crore, equity is still Rs 300 crore against a total of about Rs 428.6 crore, or 70.0 per cent. The same portfolio, the same obligation, and four different answers depending on a sentence about which sleeve is spent first.

A household meets the same choice at a size anybody can check by hand. The same Rs 2,00,000/- fee out of the same Rs 10,00,000/-. From the savings account, and the shares stay at Rs 6,00,000/- of a Rs 8,00,000/- total, or 75.0 per cent. By selling shares, and the shares are Rs 4,00,000/- of the same Rs 8,00,000/- total, or 50.0 per cent. One fee, one total afterwards, and a twenty five point difference in exposure, decided by which pocket it came out of.

Why is being listed not the same as being liquid?

Because they answer different questions, and documents keep writing one and meaning the other. The Anantara mandate carries a line saying no unlisted holdings. The no-unlisted line is a genuine constraint doing useful work, and it has not addressed the liquidity requirement at all.

Being listedAdmitted to trading on a recognised venue. A register answers it yes or no. is a fact about a venue: a register answers it yes or no, today and next year. Being liquidAble to be turned into cash quickly and in size without moving the price much to find the other side. is a fact about depth: how much can be sold in a day without moving the price to find somebody on the other side. The first has one answer for all time and all sizes; the second has a different answer every day and again for every size of sale.

The street version is a Sunday vegetable market. Every stall being open is the listing. A request to one stall for two kilos is served in a minute. A request to the whole market for four hundred kilos before noon starts the prices moving. Depth was never the same thing as opening hours.

Whether a particular holding can be sold quickly, and what a large sale costs to push through a market, are facts about markets rather than mandates and are covered separately. The design consequence is what matters here: a rule about where a holding trades is not a rule about how much of it can be turned into cash by the date on the requirement.

What must a liquidity paragraph contain to be checkable?

Four things, and three of them are easy. The amount. The date. The order of the sources it will be met from. And the call size at which meeting it starts to touch something else in the same document. The first three describe an obligation and most paragraphs carry them in some form. The fourth is the only one connecting the obligation to the rest of the mandate, and almost no liquidity paragraph contains it.

The Anantara version is short. Rs 50 crore. On the last day of the stated quarter. Met from cash first, then by selling equity. And meeting it touches the equity range floor at a call of Rs 200 crore, or 40 per cent of the portfolio, with the ceiling untouchable along the way since the peak is 66.7 per cent. Four sentences, checkable by anybody with the allocation table in front of them.

Writing the fourth item honestly forces a second finding out: the mandate wrote a range for equity and none for anything else. Fixed income reaches 50.0 per cent against a 30.0 per cent policy weight, and cash reaches nil against a 10.0 per cent one, and neither breaches a stated line because the record contains none for either sleeve. The absence of a stated range is a finding about the document, and naming it beats quietly assuming a limit exists.

How does a committee actually use this on a Tuesday?

The committee carries three numbers into the room, one line each. How much can be paid without selling anything, the cash sleeve at Rs 50 crore. Where paying it leaves the shape, equity at 66.7 per cent. How large a call breaks something, Rs 200 crore. Rukmini Deshpande's committee needs no model for any of that. All three fall out of the allocation table the mandate already carries.

An analyst reading somebody else's mandate does the same arithmetic from the outside and learns something the document does not say. A private holder does it too, at a different scale and with the same steps: the school fee, the property deposit, the tax payment due in March. The arithmetic does not care whether the payer is a Rs 500 crore endowment or a household with one savings account. Either way it is arithmetic on a shrinking total. Only the size of the numbers changes.

One habit is where committees go wrong even when they do run the numbers: they test the biggest call they can imagine and stop. Here that is Rs 250 crore, reporting an equity weight of 40.0 per cent. The largest-call test is a real finding and the wrong test on its own. The greatest pressure on the ceiling happened at Rs 50 crore and the range floor was reached at Rs 200 crore, and a single probe at the far end saw neither.

The last habit of practice keeps the two ideas apart in the room. A fall in value and a call on cash can turn up in the same week, and neither causes or cancels the other. The portfolio can be down Rs 48.5 crore from its peak on the Tuesday and still owe Rs 50 crore on the Thursday. The remedy for the first is time and the remedy for the second is cash, and a committee with one plan for both has a plan for neither.