Hedge Risk or Counterparty Risk: Two Different Gaps

Two different things can let a covered holder down. Paperwork that was never a close match to the holding leaves a gap the holder keeps: that is hedge risk. The other side not doing what it agreed leaves the holder with no cover at all: that is counterparty risk. Fixing the paperwork does not improve the other side, and a faultless other side does not improve the paperwork.

A contract is two objects at once. One object is a set of terms, and the other is a party who has agreed to those terms. Each of the two can let a holder down entirely on its own, on a different day, in a completely different way, and satisfaction about one of them settles nothing whatever about the other. That is the whole of the distinction, and everything below is the working.

A household books a hall for a wedding in November and pays in advance. Two separate things can go wrong with that booking and only one of them is about the hall. The booking might be written for the wrong Saturday, which is a fault in the form and is legible on the form the moment it is signed. Or the hall might be shut before November arrives, which has nothing to do with the form at all: the form is perfect and there is no hall. Nobody who checks the date on the form has learned anything about whether the hall will still be standing, and nobody who visits the hall has looked at the date. The same picture, worked in rupees, follows.

One holder, one unit and one contract price

One holder has one unit of an invented reference asset, and it is worth Rs 2,000.00/- today. Financing costs 6.50 per cent a year. Multiply Rs 2,000.00/- by 0.065 and Rs 130.00/- of carry drops out; set that back on top of the spot price and a contract for delivery one year out is priced at Rs 2,130.00/-. The reference asset hands over nothing at all across the twelve months. Had it paid anything, that payment would come straight off the Rs 130.00/- of carry and the contract price would sit lower than Rs 2,130.00/-.

The holder writes one contract on the short side at Rs 2,130.00/- against the unit held. When the price of the reference asset travels 4.0 per cent, that is Rs 80.00/- on one unit. Where the contract is cleared instead of agreed privately, the holder lodges Rs 160.00/- of collateralProperty or cash lodged with somebody else so that a promise does not have to rest on trust alone. The property stays lodged for as long as the promise is open. against one unit, being 8.0 per cent of the Rs 2,000.00/- exposure. The 8.0 per cent collateral figure was invented for teaching and appears in nobody's rulebook.

Rs 2,000.00/- and Rs 2,130.00/- are prices: one is what a unit changes hands at today, the other is what a contract fixes for a later date. Rs 80.00/- is a payoff, being what one side owes the other once the reference asset has travelled four per cent. A payoff is not a premium. A premium is a fee paid up front to buy a right, and no fee is paid to enter a contract of this kind. A payoff is not a profit either. A profit is what is left once a cost has been subtracted from a payoff, and no cost has been subtracted. Rs 2,000.00/- of exposure is what a single unit actually puts at risk, and an exposure is not a notional: a notional is a fixed face amount written into a contract, and a contract over one unit carries no such fixed amount.

What is hedge risk, for a reader arriving with none of the background?

Hedge risk is what a holder still carries after writing a contract that does not match the thing it was written against. Hedge risk is not a fault in anybody's behaviour. Nobody has broken a promise, nobody has been careless, and nothing has gone wrong in any ordinary sense. The paperwork simply describes something slightly different from the holding, and the difference is what the holder keeps.

There are exactly three places where the description and the holding can part company, and each of them is a line on the contract. Start with the date. A holding stops mattering on some day, a contract stops on some other day, and the stretch between those two days is bare at one end or covered at the other for no reason. Then the quantity. Contracts are written in whole units of whatever size the venue publishes, so a holding of an awkward size leaves a few units standing behind no contract at all. And then the thing itself. Sometimes nothing has been written over what a party actually holds, so the nearest available contract is written over something near it and the holder accepts whatever daylight sits between the two.

Now the property that makes hedge risk unusual, and it is the one worth carrying away. Hedge risk is a property of the terms rather than of anything that happens afterwards, so it is present from the moment the contract is written and knowable at that moment. Every input it depends on is printed on the ticket. Nobody has to wait for a price to move, a year to pass or a party to be tested. A holder who reads the ticket on the day of signing, and compares each of its three lines against the holding, has already found out exactly how much of this risk there is going to be. The holder cannot say on that day what the gap will cost in rupees. The cost depends on where the price goes. The presence of the gap is knowable on day one. The size of the bill it eventually produces is not.

The rupees each of those three joints leaves behind are worked through where the joints themselves are taken apart. All three joints exist from the day of signing, and all three are readable on that day.

What is counterparty risk, for that same reader?

Counterparty risk is what a holder carries because the party on the other end of the contract may not do what the contract says. Every contract has a second side. The second side is an organisation with affairs entirely of its own, and the holder neither controls those affairs nor necessarily sees them. It has its own borrowings, its own obligations to people the holder will never meet, and its own bad week. None of that is written on the ticket, and none of it can be read off the ticket.

Counterparty risk takes two shapes, and the two shapes are not the same problem, so each is treated on its own.

The first shape is money owed and not paid. The contract has moved in the holder's favour, an amount is due, and it does not arrive. The holder is now an unsecuredOwed with nothing lodged behind it, so payment depends entirely on the payer's own condition on the day. creditor of somebody else, standing in a queue whose length and order the holder had no say in. The holder is left with recourseThe route somebody has to being paid when the other side does not pay, and the person or the property that route ends at., and recourse is not the same as a payment.

The second shape is worse and gets a fraction of the attention. The cover itself disappears. A contract with a party that is no longer there is not a contract at all, so the holder who believed they were covered is uncovered, and the way they find out is that the day arrives when it mattered. Nothing on the ticket has changed. No line was struck out. The entire value of a promise sits in whether the promiser is there to keep it. The paperwork is still in the file, still perfectly drafted, and it protects nothing.

Now the property that separates this from hedge risk. Counterparty risk is a property of somebody else's condition rather than of the terms, so it is not knowable in advance in the way hedge risk is. A holder can investigate, ask for accounts, ask for collateral, form a view. A holder cannot read the answer off the contract. The answer is not on the contract and never will be. The contract records what was agreed. Whether it gets done is settled somewhere else entirely, on a day nobody has picked yet.

What do these two risks have in common?

One property joins the two risks, and the shared property explains why they turn up together and why they are so easily confused for one heading. The holder had neither of them on the morning before the contract was written. Both arrived with it. Not as a side effect, not as fine print, but as the direct consequence of the arrangement the holder chose.

Look at what actually changed. Before the contract, the holder was exposed to one thing: the price of the reference asset could fall, and a fall would be worn in full. After the contract, that particular exposure has been given up. In its place the holder now holds a contract that may not fit, and stands opposite a party that may not perform. For anybody who keeps a written list of what they are exposed to, writing a contract takes one line off that list and puts two on.

None of that is an argument against covering an exposure, and none of it is an argument for one. The count is unwelcome mostly because the word hedge carries a sense of things having been tidied up. Nothing was tidied up. One exposure was exchanged for two, and the two are of a different kind: the original one was to a price, and both new ones are to arrangements. Whether that exchange suits any particular party turns on facts outside these figures.

What actually separates the two?

Four differences, and they are worth holding as four rather than as an impression. Take them one at a time before the grid puts them side by side.

The subject each is about. Hedge risk is about fit: does the description on the contract match the thing being held? Counterparty risk is about performance: will the other side do what the contract says on the day it says to do it? A contract can fit beautifully and be held by somebody who will not pay. A contract can be held by the soundest party on earth and describe the wrong dates.

Where each one lives. Hedge risk lives in the terms, in a document the holder has a copy of. Counterparty risk lives in the party, somewhere the holder has no copy of anything. Two different places mean two different sets of people looking, and the split between those people is what produces the practical mess described below.

When each one becomes visible. Every input hedge risk needs is on the ticket, so hedge risk is visible the day the contract is written. Counterparty risk is visible on the day it is tested and, in the ordinary case, on no earlier day. Between signing and testing the holder may form views, but a view is not a reading.

The remainder each one leaves behind. A gap is a gap in both directions, so hedge risk leaves a remainder that can land either side of nil: the mismatch can leave the holder a little worse off or a little better off, depending on which way the price went. Counterparty risk does not leave a remainder. Failure to perform removes what was there. There is no version of a party failing to perform in which the holder ends up marginally ahead.

The error the whole contrast exists to prevent is filing both of these under one heading called the risks of hedging. Under one heading they get one owner, one review and one tick, and the tick gets put in the box after whichever of the two that owner happens to understand. The two cases below show exactly what that costs.

Why does reducing one do nothing at all for the other?

The claim is easy to state and easy to nod along to, so it is worked twice in figures, changing one thing each time. Run the same holder and the same unit through both.

Case one: a flawless fit and a party that does not perform

The contract is written on the very reference asset held, for exactly one unit, ending on exactly the day the unit is sold. There is no date joint, no quantity joint and no item joint, so nothing is left over anywhere in the terms. On the day, the reference asset price gives up 4.0 per cent. A fall of 4.0 per cent is Rs 80.00/- off the unit held. The contract is on the short side, so it moves Rs 80.00/- the other way, and Rs 80.00/- is due to the holder from the other side.

The Rs 80.00/- does not arrive. The fit was flawless and the holder now stands exactly where an uncovered holder would stand: down the full Rs 80.00/- on the unit, with nothing coming the other way. Note carefully what that Rs 80.00/- is. The amount is a payoff owed on the contract and never received. A payoff is not a loss on the holding: the loss on the holding is a separate Rs 80.00/- fall in the value of the unit. A payoff is not a profit either, and nothing has been netted off anything. Two amounts of the same size, doing two different jobs, and only one of them ever moved.

Case two: a party that performs faultlessly and a contract that does not fit

Now hold the party still and move the terms. Every amount owed on this contract is paid on the day it falls due, without a reminder, for the whole life of the arrangement. There is nothing to complain about on that side at all. But the holding stops mattering at six months while the contract runs to twelve, so the holder closes out against a contract price that still has unexpired financing sitting inside it, and a remainder is left behind. The rupee arithmetic of that remainder is worked out where the three joints are taken apart.

Put the two cases beside each other and the conclusion falls out: neither answer contains any part of the other. In case one the fit answer was perfect and the holder was completely uncovered. In case two the performance answer was perfect and the holder still had a remainder. The two cases together are what it means to say these are two risks and not two views of one. A party that checks either alone has finished half a job, and has no way of telling from the half it did how the other half would have come out.

What does clearing change, and what does it change counterparty risk into?

Start with what actually happens. Where a contract is cleared rather than agreed bilateralSettled straight between the two sides, with nobody in the middle of it and nobody else keeping the record.ly, a clearing corporationThe institution that takes on the two sides of a cleared contract as two separate obligations, so each side deals with it rather than with the other side. stands between the two sides, each side lodges collateral before it may carry the position at all, and there is a stated order in which resources get drawn on when a member fails to meet what it owes. The order is settled while nobody yet knows which party will be the one needing it, and being fixed in advance is exactly what makes it a protection.

The steps of that order, and the thresholds inside it, are written by the Securities and Exchange Board of India (SEBI) at sebi.gov.in, and they change.

Now the part that gets skipped most often. Clearing does not delete counterparty risk. Clearing converts it. Work through what the holder actually ends up facing, and count.

The holder no longer faces one specific named party. The removal of that named party is real, and it is the whole point of the arrangement. But in place of that party the holder now faces three things. There is an arrangement, and unlike a party an arrangement cannot be investigated: a holder can read a counterparty's accounts and form a view, and cannot do the equivalent here. There is a clearing memberAn institution with standing to deal directly with the clearing corporation, through which a participant's position reaches the cleared system. now sitting between the holder and that arrangement, itself a party with affairs of its own. And there is a standing requirement to lodge Rs 160.00/- of collateral on one unit, at the 8.0 per cent invented for teaching, and to keep topping it up for as long as the position stays open.

The third of those is the honest summary of the whole exchange: cash where there was trust. Under the private arrangement the holder was relying on somebody's willingness and ability to pay, and it cost nothing to fund. Under the cleared arrangement the holder is relying on a set of default arrangementsThe written sequence in which money and assets are drawn on when a member cannot meet what it owes. that are published rather than promised, and paying for that with a permanent claim on cash. Something was given up in both directions.

Whether the second position is smaller than the first is not something the exchange itself answers. Neither position has a size until somebody supplies one. The exchange names what goes and what arrives, and naming is not measuring.

What can move hedge risk, and what cannot touch it?

Hedge risk lives in the terms, so only the terms move it. The claim is not a rule of thumb but the definition read backwards, and it means the list of things that can reduce hedge risk is exactly three items long. The three levers are the three joints: a contract ending nearer the day the exposure ends, a quantity closer to the number of units held, or a contract written over the thing actually held rather than over something near it. There is no fourth lever, and there is no lever that works on all three at once.

Now the hard limit, and it is where most of the disappointment lives. Every one of those three levers depends on a contract with those terms actually being available to write. A holder who wants an end date three weeks earlier can only have it if a contract ending three weeks earlier exists. A holder who wants a quantity that divides the holding evenly can only have it if the contract carries that many units. Which contracts exist, what dates they run to and how many units each one carries are decided by SEBI at sebi.gov.in, working through the venues. So the three levers are real and they are also frequently jammed, and a holder who finds all three jammed has hedge risk they cannot reduce at all.

And now what does not touch hedge risk at all. A stronger party on the other side changes nothing about fit, and clearing changes nothing about fit either. Clearing works on who the holder faces. Clearing has no opinion about what the contract says. A perfectly cleared contract written on the wrong dates is still written on the wrong dates, and the remainder it leaves behind on the day the holding is sold is the same remainder to the paisa. Clearing arrives with a great deal of machinery, machinery reads as thoroughness, and thoroughness in one place is regularly mistaken for thoroughness everywhere.

Which of the two is the bigger problem?

A comparison invites the question, and the figures above cannot answer it. The reasoning follows, and can be checked rather than taken on trust.

A ranking is arithmetic done on a record, and no record stands behind these figures. How often either of the two has actually bitten is not written down anywhere in them. Nor is there a likelihood to attach to either, and the only sizes present were invented to give the carry arithmetic something to work on. So there is no frequency to state, no rupee figure that stands as a typical loss for either, and no basis for saying which of the two costs parties more.

Behind every number above sits a short list, and it is shorter than it looks: one made-up thing, one price for it, the cost of carrying that price forward twelve months, and a collateral percentage somebody chose to give the margin arithmetic a figure to work on. No failure appears anywhere in that set. Nothing in it says how often. Nothing in it says how large.

A control would have to run along a size, and neither of these two risks has a size in these figures. Sliding one would invent exactly the ranking that the evidence does not support. The shape of each risk remains, and a grid that carries into a real arrangement where the sizes are somebody else's to supply.

Is the cleared route the safer one?

There is a pull, at the end of a comparison like this, towards saying that the cleared route is the safer one. Nothing in these figures supports it. Clearing hands the holder something that cannot be investigated in exchange for something that could be, and it puts a permanent claim on the holder's cash for as long as the position stays open. The privately agreed route leaves the other side visible and leaves the holder's cash alone, and leaves the holder alone with that other side as well.

Which of those trades suits a particular party turns on how much cash it can raise at short notice, how much it can find out about whom it is dealing with, how long it means to hold the position, and what SEBI at sebi.gov.in and the Reserve Bank of India at rbi.org.in ask of each route. Not one of those four sits in the figures here.

A distinction is a thing to see with, not a thing to act on. The distinction is handed over, and what to do with it belongs to the party holding the contract.

The reference asset, its holder and the party standing opposite are invented.
Educational material. Not advice on any investment, tax, budget or market position.