Liquidity Risk vs Funding Risk: Selling or Raising the Cash

Liquidity risk is whether an asset can be turned into cash near its value. Funding risk is whether money can be raised or rolled at all, and at what price. At Vindhya Commercial Bank Limited, invented, the one ladder limit is within and both funding limits are in breach, and the two words are not one word.

Most people meet these two phrases inside the same sentence and reasonably conclude that somebody is repeating themselves. The two phrases are not one phrase. The two describe faults on opposite sides of the balance sheet, they are paid for in different currencies of pain, and at Vindhya Commercial Bank Limited they point in opposite directions at the same moment. One ambiguity has to be settled first. Without it the rest of the comparison reads as an argument with its own heading.

Which of the two meanings of the word liquidity is set against funding risk?

The word liquidity is used two ways in this subject area, and both of them are correct. In the broad sense it names the whole question of cash arriving and leaving an institution: the ladder, the buffer, the thirty day computation, the survival horizon, all of it. The broad sense is why a whole sequence of related subjects carries the word in its name. In the narrow sense the word names one specific thing: whether an asset can be turned into cash near its value, quickly enough, when the cash is actually wanted. The narrow sense is the one set against funding risk. Hold the broad sense in mind instead and the comparison appears to contradict itself on every second line.

Here is the everyday version of the same trouble. The word traffic names both the entire subject of how a city moves and one specific thing: the queue a driver is sitting in. A sentence saying traffic is fine can mean the city has good roads or that this particular junction is clear, and the two are not the same claim at all. Nobody proposes to fix the English language. A careful person says which one they mean, every single time, and that habit is the whole of the discipline.

So take the two definitions apart and keep them apart. Asset liquidity riskThe risk that an asset cannot be turned into cash near its value, quickly enough, when it is needed. is a statement about things the institution already has. Funding riskThe risk that money cannot be raised or rolled at all, or can only be raised at a price that changes the economics. is a statement about people the institution does not control. One is a question about buyers. The other is a question about lenders. Everything that follows comes from that single sentence.

What exactly is asset liquidity risk a statement about?

Asset liquidity risk is a statement about the distance between what something is worth and what it fetches when the cash has to be in hand. The claim is not that the valuation is wrong. The claim is that valuation and realisation are two different events, and that realisation has a price attached. To monetiseTurn an asset into cash by selling it or by borrowing against it, and the two routes carry different costs on the same asset. a holding is to accept that price.

Three things decide how big that price is, and none of the three is about the value being wrong. The first is size against market depthHow much of an asset can be sold before the price moves, which is what makes one asset liquid and another merely valuable.. A holding small enough to disappear into the ordinary buying of an ordinary day costs almost nothing to sell. The same asset in a size the available buyers cannot absorb moves the price against the seller, and the seller pays the move. The second is speed. Three weeks of patience is a completely different sale from this afternoon, and the haste is the part of the cost the seller chooses rather than inherits. The third is who else is selling. One buyer is a negotiation, no buyer is a discount, and everybody trying to sell the same thing on the same morning is a wider discount still.

The everyday version is a scooter and a flat, and it is exact. A used scooter is worth what a used scooter is worth, and a seller who needs the money by Friday will get close to it. There are many buyers and one scooter is a small thing. A flat is worth far more and is far harder to turn into cash by Friday. A household that needs money quickly therefore borrows against the flat rather than selling it. Nobody in that story thinks the flat has fallen in value. The household has met the cost of realising a value it already had, and a household that must move fast meets a bigger cost. The flat is asset liquidity risk in a sentence, and the bank version differs only in the number of zeroes.

Why is asset liquidity risk not the same thing as price risk?

Confusing the two costs people more than any other mistake in this subject. Both faults show up as money, and both show up on the same bad morning. Price riskThe risk that the value of a position changes, which is a different subject from the cost of realising a value that has not changed. is the value moving. Asset liquidity risk is what it costs to realise a value that has not moved at all. If a holding is marked at Rs 100 crore and the mark is correct, and Rs 2 crore is still handed over to get the cash today, nothing about the Rs 100 crore was wrong. The Rs 2 crore bought immediacy, and the buyer who took the position off the seller's hands charged for taking it.

The mechanism is the bid-offer spreadThe distance between what a buyer offers and a seller asks, which is the immediate cost of turning a position into cash. and its behaviour at size. The buyer is not doing the seller a favour and is not predicting a fall. The buyer is being paid to hold something they did not want until somebody who does want it turns up. Price risk is answered by valuing the position. Asset liquidity risk is answered by pricing what it costs to stop holding the position, and no valuation of the position says anything about that cost. Valuing a position, measuring how sensitive that value is and the methods for both belong to the market risk sequence.

What exactly is funding risk a statement about?

Funding risk is a statement about other people. Funding risk asks whether the money the institution runs on will still be there tomorrow morning, and what it will cost if it is. Nothing on the asset side answers that question. The answer sits in the decisions of depositors, of the counterparties who hold the bank's own paper, and of the institutions that lend to it overnight. An institution can hold assets everybody wants and still fail, if the money that paid for those assets walks out faster than the assets can be turned around.

Three things decide how exposed an institution is. The first is who provides the money and how many of them there are. Twenty providers can be argued with and one provider cannot. The second is how often the money has to be replaced, called the rollReplacing a maturing borrowing with a new one from the same or a different provider, which is where funding risk actually shows up.: money that never has to be replaced can never be refused, and money that has to be replaced every ninety days offers four chances a year for somebody to say no. The third is why the provider is there at all. A provider who came for the rate leaves for a better rate. A provider who came because the salary lands in that account every month has a stickier reason to stay.

The stall outside the office building is the whole of funding risk at human scale. The vegetable seller takes stock from a supplier on thirty days of credit and settles at the end of the month. He is not worried about the value of his stock, and he is not worried about buyers. The stock is fine and the buyers arrive at lunchtime every day. The supplier can end him by deciding in month four that credit now comes at a price, or comes not at all. His assets are perfectly sound. His funding was a monthly conversation he had stopped noticing, and it is the same conversation Vindhya Commercial Bank Limited has nine times over with the counterparties holding its certificates of deposit.

How much does each one actually cost at this bank?

Vindhya Commercial Bank Limited prices each of these risks exactly once in its own records, and those two numbers are the only places in the whole account where the difference becomes money. Take the asset side first. Contingency action F2 raises cash by selling Rs 3,600 crore of available for sale securities in two to three working days at an assumed discount of 2.0 per cent. So Rs 72 crore never arrives, and Rs 3,528 crore does. The Rs 72 crore is not a loss on the securities. The Rs 72 crore is the price of having the cash this week rather than at maturity, and it is paid once.

The pair is more instructive than either action alone, so stand contingency action F1 next to F2. F1 raises Rs 4,800 crore against government securities the bank holds outside its buffer, on the same day, at no assumed discount at all. A repurchase arrangement and the way it settles belong to the fixed income subject area. The distance between an assumed 2.0 per cent on one asset and an assumed nothing on another, on the same balance sheet on the same morning, is the entire content of the phrase high quality liquid asset. Which assets an Indian bank may actually treat as liquid, and at what haircut, is set by the Reserve Bank of India at rbi.org.in.

The funding side is priced by an early warning indicator rather than by an action. Indicator W1 watches what the bank pays on its own certificates of deposit against the rate at which banks lend to each other, and it turns amber at 25 basis points over and red at 50. The two triggers are the bank's own working numbers. More than one thing in this account is Rs 2,880 crore, so turn the triggers into money and name the object every time. The certificates of deposit outstanding are Rs 2,880 crore across nine counterparties, and on those 25 basis points is Rs 7.2 crore a year and 50 basis points is Rs 14.4 crore. Applied to the whole of wholesale funding, being wholesale term deposits of Rs 11,520 crore plus borrowings of Rs 8,400 crore, being Rs 19,920 crore in all, a spread over the interbank rateThe premium an institution pays over the rate at which banks lend to each other, which is the market's price on that particular name. of 25 basis points is Rs 49.8 crore a year and 50 basis points is Rs 99.6 crore. Set against net interest income for the year of Rs 2,880 crore, a different Rs 2,880 crore entirely, those are about 1.7 per cent and about 3.5 per cent of the year's earnings.

Which cost is paid once and which one is paid every day?

One difference matters more than any of the definitions, and that one changes a decision. The asset liquidity cost is paid once. The sale happens, the discount is handed over, the cash arrives, and the matter is closed by the evening. The funding cost is paid continuously. A spread over the interbank rate is not an event, it is a rate, and it goes on being charged every single day the money is outstanding. A one-off cost and a continuing cost cannot be compared at all until somebody names the period on both of them, and almost every argument about which is cheaper is really an argument in which one side has quietly dropped the period.

The household version has the identical shape. Take it first. A wedding is coming and the money is short. One route is to break a fixed deposit early and pay the penalty. The penalty stings once and is then over. The other route is to borrow. Borrowing costs nothing today and costs something every month for as long as the borrowing runs. There is a length of time at which the second route has cost exactly as much as the first, and after it the second is the more expensive one. Nobody in the room usually works out where that point is, and the whole reason the question feels hard is that one side is a lump and the other is a drip.

Worked out at this bank, in its own numbers, a full year of paying 50 basis points more on all Rs 19,920 crore of wholesale funding costs Rs 99.6 crore. Rs 99.6 crore on a Rs 3,600 crore sale is a discount of 2.767 per cent. The 2.767 per cent is the crossing, and the crossing is the whole comparison in one line: a single afternoon at a 2.767 per cent discount and a whole year at 50 basis points over the interbank rate are the same money. On the milder trigger the same arithmetic gives 1.383 per cent: Rs 49.8 crore over Rs 3,600 crore is 1.383 per cent. Read the other way, the plan's own assumed 2.0 per cent discount, costing Rs 72 crore, is overtaken by a year at 50 basis points in about month 8.7. Rs 99.6 crore a year is Rs 8.3 crore a month, and Rs 72 crore divided by Rs 8.3 crore is 8.67 months. At 25 basis points it takes about 17.3 months to catch the same Rs 72 crore.

What happens when both risks arrive on the same morning?

Everything so far has treated the two as separate faults that can be measured one at a time. In a stress they are not separate, and they do not take turns. Both prices are set by the same nervousness in the same room, so the morning funding costs 50 basis points more is the same morning the securities sell at a wider discount. The institution is not choosing between paying once and paying every day. The institution is paying on both sides at the same time, and the bill on each side is bigger than the number written in the plan.

The measures do nothing about it, and that is the quiet part. The plan carries a fixed 2.0 per cent discount for action F2, and a fixed assumption cannot notice that everybody else is selling the same paper that morning. The thirty day computation applies fixed haircuts and fixed run-off factors, and a fixed factor cannot notice that the providers of money have all reached the same conclusion on the same day. Neither contains any mechanism at all for the other one moving. Neither omission is an accusation of carelessness against anybody: it is the ordinary consequence of measuring two things separately and then meeting them together.

Which of the two does this bank actually have?

Definitions are cheap. The useful question is which of the two faults is actually present at Vindhya Commercial Bank Limited at month 12, and the bank's own records answer it without any interpretation at all. Three of its twelve limits touch this subject, and every figure below is restated from the bank's own set.

Limit L9 caps the negative gap in the first ladder bucket, LB1, at 28.0 per cent of that bucket's outflows. The gap is minus Rs 2,400 crore on outflows of Rs 9,600 crore, being 25.0 per cent, so utilisation is 25.0 over 28.0, or 89.3 per cent, and it is within. L9 is a position limitA limit on the shape of the balance sheet, such as a cap on a bucket's negative gap, as against a limit on who provides the money.: it is a statement about the shape of the ladder and not about who is providing the cash. Limit L10 caps wholesale funding at 20.0 per cent of total liabilities. Wholesale term deposits of Rs 11,520 crore plus borrowings of Rs 8,400 crore is Rs 19,920 crore, the cap on total liabilities of Rs 88,320 crore is Rs 17,664 crore, and Rs 19,920 crore over Rs 17,664 crore is 112.8 per cent, an excess of Rs 2,256 crore. Limit L12 caps the top twenty depositors at 12.0 per cent of deposits, being Rs 9,216 crore against Rs 76,800 crore of deposits. The top twenty hold Rs 11,136 crore, so utilisation is 120.8 per cent and the excess is Rs 1,920 crore. One position limit, comfortably within. Two funding limits, both in breach, both open for months, and both carrying Devendra Achar, head of treasury, as the named risk owner.

The early warning indicators tell the same story in a second, independent way. Vindhya Commercial Bank Limited runs seven of them, numbered W1 to W7, and the split between the two risks is four, two and one. W1 watches the cost of the bank's own certificates of deposit, W2 the share of those certificates rolled at each auction, W3 the daily net retail deposit flow and W4 the top twenty depositor share, and every one of those four is a question about whether money arrives and stays, and that question is funding. W6 watches the buffer against thirty day net outflows and W7 watches the survival horizon in days, and both are questions about whether the stock of cash lasts, and that question is position. W5 watches undrawn committed lines drawn by customers, and it is neither: it is a contingent outflow driver sitting on the asset side. At month 12 the red indicator is W4, a funding indicator, and the amber one is W7.

When does the difference actually change a decision?

A distinction that never changes a decision is a debate about words. The distinction between the two risks changes four decisions, and each of them is a different kind of change.

The first is sizing a stock of liquid assets. Sizing the stock is an asset liquidity decision from beginning to end, and it adds precisely nothing to funding stability. A larger holding of easily sold securities makes the institution better at turning things into cash and does not make one depositor more likely to stay. If the fault is on the funding side, buying more securities is an expensive way of not fixing it. The second is choosing between selling something and borrowing something. The choice is between paying once and paying every day, and it cannot be made honestly with only one of the two numbers on the table. The crossing at 2.767 per cent exists precisely so that the two can be compared at all.

The third is writing a limit. A limit has to say which of the two risks it caps, or it caps neither properly: L9 constrains the shape of the ladder and has nothing to say about who is providing the money, and L10 constrains who is providing the money and has nothing to say about the shape. A framework with one of each is not a framework with two of the same. The fourth is reading, and reading is where most people meet the distinction. When a summary line uses the broad word, the honest response is to ask which half it is describing, and the answer at this invented bank is that the green half was described and the red half was not.

Who has to hold the two apart in an ordinary week?

Devendra Achar, the head of treasury at Vindhya Commercial Bank Limited, is the named risk owner of both open funding breaches, and his week is spent on the funding side of the distinction: who is rolling, at what spread, and how many providers there are. When his committee asks whether liquidity is comfortable, he has to answer twice. The honest answer is comfortable on the position and not comfortable on the funding.

Sunanda Ravikumar, the chief risk officer, reads the same paper for the shape of the framework rather than the numbers, and the shape is plain: one position limit and two funding limits, and the two that are breached are the pair nobody can fix by buying securities. A credit analyst at another institution, looking at this bank as a counterparty rather than from inside it, does something similar and quicker, and reads the funding side first. An institution that has to replace a large share of its money often can be pushed by other people, and an institution holding assets it can sell can only be pushed by itself.

Girish Talwalkar, group treasurer of Nirjhar Industries Limited, invented, sits on the other side of the table and faces exactly the same pair. His asset liquidity question is whether the receivables and stock the company holds can become cash before the month ends. His funding question is whether the lines the company runs on will be renewed. The distinction does not belong to banks: it belongs to anything that has assets on one side and somebody else's money on the other. A household is the smallest version. The deposit that can be broken at a penalty is the asset liquidity side. The overdraft that comes up for renewal every year, at a rate somebody else sets, is the funding side, and no amount of care about the first has any effect on the second.