Credit Event vs Liquidity Event: Telling Them Apart

Two different questions, not two points on one scale. A liquidity event asks where the cash is and when it will arrive; the promise itself is untouched. A credit event asks whether the promise still reads as written, and whether somebody entitled to say so has said it. Which label attaches to a set of facts is settled by the document, not by how alarming the facts sound.

One arrangement sits under everything below, and most readers already have it. Naming it before the definitions start is what makes it possible to take apart. Meeting these two terms together, almost everybody lays them on a single line running from mild at one end to severe at the other, with liquidity near the mild end and credit near the severe end. The severity line is comfortable, it is easy to remember, and it misleads on the first real set of facts it meets. A credit event and a liquidity event are not two positions on one scale; they are the answers to two questions that were written down separately because separate consequences hang off them. Stating that the two words are different does not go far. The pair of questions is what is worth having. A question can be put to a set of facts and a label cannot.

And here is the part that reads like a legal detail and is in fact the whole of the material. Quiet afternoons did this drafting. Which occurrences count, who may state that one has occurred, what becomes claimable and from whom, and in what order competing claims get met, were all fixed months or years ahead of anybody needing them, on days when the payments were arriving and none of it felt pressing. The timing of that drafting is not incidental to how well it works. Two sides can settle the meaning of a word while neither of them can yet tell which way the word will cut. Put the same wording up for agreement after the money has run short and nobody is defining anything any more: each side simply needs the word to mean whatever helps it, and bargaining of that sort finishes exactly where it started. Which is also why a reader wanting to classify a set of facts has to go and read something rather than form an impression.

What does a liquidity event actually put in question?

The easier of the two to feel comes first, and slowly. Either term may be new. A liquidity event is an occurrence in which the money and the moment fail to coincide. Nothing beyond that. Consider everything the sentence leaves out. How much is owed: absent. The dates that were agreed: absent. The rate that was struck, and who is entitled to what: both absent. Every term a holder lent against still reads exactly as it read on the day of signing. Timing is what is in doubt, together with reach, and nothing further.

Now the version from the street, and it is the one worth keeping in mind for everything that follows. A small courier firm has signed contracts with four office buildings, has invoices out against all four, and is owed money that lands next week. On Tuesday morning its account is empty and the diesel bill is due. Nothing about what that firm owes has changed since Monday. Nothing about what it is owed has changed either. The order book is full. The signatures are all in place. One date sits ahead of another date, and the gap between them has to be crossed with cash the firm does not have on Tuesday. A shortage of cash is a statement about a calendar, not a statement about an obligation.

Move that across to a borrower with a bond outstanding and nothing changes in the shape of it. The amount owed stays where the document put it. The dated payments stay on their dates. The rate written into the document stays written into the document. A reader who hears that a borrower is short of cash has learned something about a calendar and has learned nothing yet about the promise. Whether the promise is also in question is a second enquiry, run against a different document, and everything below is about that second enquiry and about how the two are kept apart.

Whether a shortage of cash is temporary or otherwise is a judgement about a business, and it needs far more of that business than a two-line example ever carries. The courier firm above has an order book and a diesel bill and nothing else. Two facts are all the timing point needs.

What does a credit event put in question instead?

Now the other one, given the same full treatment. These two terms get their meaning from being set out separately rather than by comparison. A credit event is an occurrence that was written onto a list before it ever happened. Being on that list gives a named party the standing to state that it has occurred. The statement then switches on something specific, drafted long beforehand, for the benefit of whoever the drafting favoured. Three things are doing work in there, and a reader who keeps only the first of them will be lost on the first real set of facts.

Thing one is a list. Somewhere in the written terms sits an enumeration, agreed by both sides while nothing whatever was going wrong. Whatever is not on that list is not one of these, however alarming it looks and however many people are worried about it. Thing two is a named party. Lists do not operate themselves; somebody has to state that an item on it is answered, and only certain people are permitted to. Thing three is the written consequence. Being on the list is not the object of being on the list; the object is that something specific was fastened to it in advance, and that fastened something is what a holder actually receives.

The difference from the previous section, in one line: this question is settled by reading a list, and that one is settled by looking at a bank balance. Neither reading says anything about the other. A reader who has the balance has answered the cash question completely and has not started on the promise question. A reader who has the list and the clauses has answered the promise question and knows nothing about whether anyone is short of money this Tuesday.

Set side by side, what does each of the two actually ask?

Both definitions are now on the table, so they can be put next to each other, and the useful arrangement is not two definitions but two questions. A definition sits there. A question can be carried into a room and asked. The preference for questions is not a matter of style; it is the difference between a reader who can classify a set of facts and one who cannot.

The liquidity question runs: is the cash where it needs to be, on the day it needs to be there? The liquidity question is answered by looking at an account and a calendar, and it has no mechanism for touching the amount owed. Whatever the answer turns out to be, the amount owed stands where the document put it.

The credit question runs: does the promise still read as it was written, and has a party entitled to say otherwise said it? The credit question is answered by reading the list of conditions and the clause naming who may declare. Whatever the answer turns out to be, it says nothing whatever about how much money is in anyone's account this morning.

A reader holding these two as questions can sort a set of facts in about a minute, and a reader holding them as severity labels cannot sort anything at all. Sorting facts in a minute is the practical payoff, and it survives testing: put both questions to any sentence about a borrower, and the sentence will usually answer one of them, hint at the other, and settle neither.

Can one set of facts be described both ways at once?

Take a single sentence, and make it a supposition rather than a report. One sentence is enough to show the whole difficulty. Suppose that a scheduled payment of Rs 91/- on Palash Cements Limited's invented five year bond does not arrive on its date. The supposition is the whole of the available information. Both questions can now be put to it.

Described as a cash question, the sentence says the money was not in the right place that morning. The cash description fits the facts perfectly. Described as a promise question, the sentence says a scheduled amount failed to arrive. Whether that answers a condition sitting on the document's list is a separate matter, and the promise description fits the facts just as perfectly. Neither reading is a stretch and neither reading is wrong.

The facts do not choose. This is the sentence to carry away if only one is carried away. The sentence does not contain what settles it, so two competent readers can look at the identical sentence, describe it two different ways, and both be describing it accurately. The settling facts sit somewhere else entirely: whether the occurrence answers a listed condition, and whether the entitled party has made its statement. Neither of those is visible in the sentence, and no amount of staring at the sentence will produce them.

Which of the two can move the amount that is owed?

Here the two separate arithmetically rather than verbally. The size of the difference is not something most readers expect, so this block is worth working slowly. The written terms of that bond run as follows. Rs 1,000.00/- of face amountThe sum standing at the far end of a bond, repayable on its last date, and doubling as the base on which the coupon rate is struck.. Five annual dates. A rate of 9.10 per cent a year written into the contract, putting Rs 91/- on each of those dates, with the Rs 1,000.00/- returning alongside the last one.

One sentence about the clock belongs beside the price rather than in a note underneath it. Annual compoundingOne discounting step per year. A future sum is reduced by the factor one plus the rate, once per year of waiting, never twice. governs every figure below, so a sum four years out is reduced four times over by the factor 1.0910 and a sum five years out is reduced five times. Handled that way the five dated amounts come to Rs 1,000.000000/- precisely. AT PAR is the name for that outcome, and it explains why one number is doing duty as both the contracted rate and the rate the discounting runs at. Suppress the convention and somebody working the sum twice a year finishes at a different price, having started from figures that print the same, and puts the discrepancy down to their own arithmetic.

Return to the supposition. A scheduled payment goes unmade. Two readings of the same morning follow, and a table separates them better than a sentence does.

Two properties of that pair deserve pinning down before the picture. Take the difference first: Rs 1,000.00/- separates the two rows, and Rs 1,000.00/- is the face amount itself. The unpaid coupon sits in both rows and cancels, leaving the principal as the sole addition, so the equality is forced rather than lucky. Take the split next: within the larger reading the unpaid coupon accounts for 8.3410 per cent and the principal for 91.6590 per cent, two shares closing on 100.0000 with no remainder to explain away. A clause and a party entitled to use it are between them worth nearly twelve times, and every other thing in the picture held still. Same borrower on both rows. Same trading. Same payments banked before any of this.

Can one of the two happen without the other?

Yes, in both directions, and the two directions together finish off the severity line for good. A borrower can be scraping around for money this month and still meet every dated payment out of whatever else it holds, with not one term of the written arrangement altered by any of it. Equally, a written arrangement can be torn up and replaced by mutual agreement at a borrower whose account is comfortably full. Somebody with no cash problem whatever has just answered the promise question in the affirmative.

Once both directions are accepted as available, the arithmetic of the thing is simple and unforgiving. Two questions, each with two possible answers, gives four combinations. All four are reachable. A line running from mild to severe has two points on it, and two points cannot hold four combinations, so the arrangement was never going to work no matter how carefully anyone placed the two words on it.

The four combinations are worth walking rather than merely counting. Everything intact and cash where it should be is the ordinary case, and it is the one almost every borrower is in almost all of the time. Cash short with every term still reading as it was signed is the courier firm from earlier, moved into a bond. A rewritten promise at a borrower with money in hand happens because the two sides agreed to rewrite it, and an agreement is what it needs rather than an empty account. And both at once is the case a reader tends to imagine when either word is used. Imagining only that fourth combination is the habit that makes the other three invisible.

Who assigns the label, if the facts do not?

Not the holder. Not the borrower. Not the reader, coming to it afterwards. The document decided in advance which occurrences sit on its list, and it named the party who may declare that one of them is answered. Everything else is commentary.

Instinct says the opposite of what follows. Occurrences do not announce themselves. A holder can be privately certain that things have gone badly and not a syllable of any written arrangement shifts, however well founded that certainty happens to be. Shifting it takes a formal statement, made by somebody the drafting picked out, under a clause the drafting picked out, to the effect that a defined condition has been met. Ahead of that statement there is an occurrence with no standing attached to it. Behind it, the standing has arrived, and everything fastened to that standing is now within reach of the parties the drafting favoured. Put two competent readers in front of one set of facts and they may well split on whether the condition has been met. The split does not matter. The entitlement to make the statement rests with one of them, and the drafting says which.

On a bond the party usually picked out is the trusteeThe one party a bond's terms appoint to act for the whole body of holders together, so a single voice can speak where several thousand would otherwise be needed.. Routing this through one party rather than several thousand is not tidiness; several thousand separate voices reach no decision at all. Every duty attaching to that role, and the trigger starting each duty running, sits with the Securities and Exchange Board of India (SEBI) at sebi.gov.in, unstated above. Filings a corporate borrower must make once a dated payment goes astray sit in the same place. The classification and carrying of a holding that has ceased paying sits instead with the Reserve Bank of India at rbi.org.in. Three routings, none of them written out.

All of this apparatus predates the trouble it addresses, and that is not a quirk of drafting practice but the only reason any of it functions. Definitions left until the money runs short ask two sides with opposite interests to agree what a word means at the precise moment agreement has become impossible. Some arrangements go further still and add a cross defaultWording that treats difficulty on one borrowing as difficulty on this one, letting a condition met elsewhere reach through into these terms. clause, letting a condition met on a different borrowing reach into this one. Whether any given arrangement does that is a question about that arrangement's own wording.

What has to be read before the question can be answered at all?

Three things, in a fixed order, and none of them is a news report. Take them one at a time. Each one answers a question the next one depends on.

First, the list of conditions, to see whether the occurrence is on it at all. Reading the list is the step people skip, and skipping it produces every wrong classification described above. Second, the clause naming who may declare, to see whether anyone entitled to has actually done so. An occurrence on the list with nobody having declared is a fact with no status, and a great deal of confusion comes from treating the first step as though it were the third. Third, the clause naming what follows a declaration, to see what is genuinely available. Availability is a matter of drafting and varies from one document to another.

A reader with all three has an answer, and a reader with fewer than three has an impression. None of that is a comment on anybody's intelligence. The difference is a matter of where the information lives. The information lives in the document, and a reader who has not read the document is guessing whatever else they may know.

Three decisions sit just past the edge of what is written above: making a formal statement, taking up an offer put to a holder, and moving with a group rather than alone. Each of the three is taken under wording nobody here has seen, and each carries consequences invisible from the material above. The machinery is what stands in their place: the list of conditions, the identity of whichever party the drafting picked out, and an address for whoever settles the remainder.

Why does every rate carry a label?

Labelling rates is a discipline rather than a subject, and the whole argument falls over without it. The 6.90 per cent used above is the five year government SPOT rateThe rate covering money handed over now and returned on one named later date, as distinct from a rate covering a stretch that has not started yet., and a SPOT rate is the one attaching to money that leaves a holder's hands today and comes back on a single named date later on. A FORWARD rate is a different object altogether: it covers a stretch that begins later and ends later still. A FORWARD rate is nobody's forecast. The curve of SPOT rates is already carrying it, and arithmetic is what pulls it back out. The two SPOT rates that would have to sit beside such a number are not stated, so the FORWARD rate for the stretch beginning at the five year date is named above and carries no number.

Palash Cements Limited issues at 9.10 per cent a year against that 6.90 per cent standing at the five year government node, and the distance between the two is 2.20 percentage points, or 220 basis points. A hundred basis points make one percentage point, so those two figures measure the identical distance and neither is a translation of the other into a different quantity. The reason for labelling every single rate is that two objects with nothing in common print as very similar numbers, and the label is the only thing keeping them apart.

What can honestly not be said about either occurrence?

How often either occurrence arrives, and whether one tends to be followed by the other, are questions only a count of real occurrences answers, and no such count has been taken above. A comparison normally reaches for one instance of each and sets them beside one another. Two definitions, one stated supposition, and a visible gap where the instances would otherwise sit are what stand in place of that pair. The absence is a feature of the subject rather than of the treatment alone: how often a credit event arrives depends on which occurrences a given document happened to list, so a count run across documents listing different occurrences is counting different things.

The gap is worth naming precisely. A reader usually wants two numbers at this point: how often this sort of thing happens, and how much comes back afterwards. The first would need a history of comparable documents. The second would need a set of claims already resolved and paid out. Where those numbers would go there is an empty space with a label on it. A labelled empty space is a better thing to be handed than a figure somebody remembered.

Why hindsight makes the lender look careless

Somebody who lent and was not repaid did not miss something obvious. Outcomes read as plain once they have arrived, and were nothing of the kind while the decision was still open. Every clause described above got drafted for exactly that reason: sensible people fully expecting repayment nevertheless allowed for being wrong, and wrote the allowance down. How Palash Cements Limited runs itself does not bear on any of this. The machinery is the subject, and the machinery reads the same way whoever the borrower turns out to be.

Why a severity dial cannot be built

A slider built for this subject would need a dial reading how badly is the cash short, and the whole argument above is that no setting of such a dial settles anything. Sliding from mild to severe cannot put an occurrence on a document's list, and it cannot make an entitled party declare. The two readings therefore sit beside each other as fixed text, and the question below leaves the sorting to the reader.

Palash Cements Limited, its five year bond and the courier firm above are invented.
Educational material. Not advice on any investment, tax, budget or market position.