Derivative Research: Describing Without Recommending

Derivative research describes what a contract obliges and what a figure actually is, and it stops before naming an action. A market observation states what is on the screen. A trade signal asks the reader to do something. Performance marketing displays a past result to attract. Writing that names every assumption, every base, every period and every missing number stays in the first of those three.

The record used all the way through, an invented one with no exchange behind it, no clearing corporation, no contract specification, no real party and no observed price, exists so that four runs of arithmetic can be worked in the open rather than merely described. The skill being taught is a way of writing and a way of reading.

Here is the situation this guide is built for. Somebody sends an analyst a note about a contract. The arithmetic in it is correct. The tone is measured. Nothing in it looks like a pitch. And the analyst still cannot tell, by reading it, whether it is describing something or asking for an action. Three kinds of writing can sit on exactly the same figure, and a reader cannot separate them by tone, by length or by how technical they sound. The demand each one makes on the reader is what separates them: nothing, an action, or their money.

The sorting question does the work all the way down, and it is why the separation is usually settled inside a single clause, most often the last clause of the note.

What is derivative research, and what are the three kinds of writing it sits among?

Derivative researchWritten work about a contract whose value is read off something else, such as a forward, an option or an arrangement with two legs. is written work about a contract whose value comes from something else. Its job is to make two things legible to somebody who was not in the room: what the contract obliges, and what each figure attached to it actually is. The job is narrower than it sounds, and most of the skill in it is refusing to do a wider one.

A reader who meets these three one at a time will quietly merge the second and the third, so meet all three at once. A market observationA statement of what a figure is and where it came from, which asks nothing of the person reading it. states what is on the screen. A trade signalA statement of an action to take, or of a view from which exactly one action follows. states an action to take, or a view from which exactly one action follows. Performance marketingA display of a result already achieved, put up in order to attract. displays a past result in order to attract. The sorting question never changes: what does this sentence ask of the person reading it? An observation asks nothing. A signal asks for an action. Performance marketing asks for money.

Three people describe the same sky on the same morning. The first says the cloud is low and the wind is from the west. The second says to leave now in order to stay dry. The third puts up a poster of last season of clear skies with their name printed under it. Nobody has said anything false and the sky has not changed. Three times over, what has changed is what each sentence wants from the person hearing it. The three kinds of writing about the underlyingThe thing whose price a contract's value is read from. Here it is one invented reference asset with no issuer, no ticker and no market. work exactly the same way.

What is a market observation, and what does one look like written down?

A market observation states what a figure is, where it came from, and what it would take for it to mean anything more than that. An observation is easier to recognise than to define, so here is one, written out on the invented record used all the way through.

The invented reference asset has a spot price of Rs 2,000.00/-. Financing costs 6.50 per cent a year. The asset pays nothing at all while it is held. A payout during the holding period would change every forward figure below, and that fact has to be said in the same breath as the price. Rs 2,000.00/- carried for one year at 6.50 per cent a year is Rs 2,000.00/- plus Rs 130.00/-, and that comes to Rs 2,130.00/-. Rs 2,130.00/- is the one year forward price.

Four features make that an observation rather than something else, and the four are worth setting out. Every figure says which kind of figure it is: the Rs 2,000.00/- is a price, the Rs 130.00/- is a cost of carrying, and neither is called a return. The rate carries its period, so it reads 6.50 per cent a year rather than 6.50 per cent of nothing in particular. The assumption is stated where it is used rather than in a note at the foot. And the arithmetic is multiplied out in the open rather than asserted, so a reader can disagree with the multiplication instead of having to trust the result.

An observation is allowed to be interesting. An observation is not allowed to be interesting because of an implication it declines to state. The second half of the definition is the harder one, and it is where careful writers go wrong. If a paragraph is arranged so that the reader supplies the conclusion the writer left out, the writer has still made the claim. The writer has only made the claim deniable.

What is a trade signal, and which clause turns writing into one?

A trade signal states an action to take, or states a view from which exactly one action follows. A signal does not have to contain the word buy or the word sell, and a signal does not have to be loud. A signal has to leave the reader with somewhere to stand.

Take the paragraph from the section above, change nothing in it, and add six words at the end. The spot price is Rs 2,000.00/-, financing costs 6.50 per cent a year, the asset pays nothing while held, so carried for one year that is Rs 2,130.00/-, a carry of Rs 130.00/-, so the long side looks well placed here. Not one figure changed. The multiplication is the same multiplication. The rate still carries its period. The assumption is still named. And the note is now a different kind of writing from the one it was a line earlier.

Here is the test, and it is something to do rather than something to avoid. The test is to read the last clause on its own, with everything above it covered, and ask whether a reader could act on that clause alone. If they could, the whole note is a trade signal whatever the paragraphs above it say. The clause does not inherit the care of the paragraphs. The paragraphs inherit the demand of the clause.

There is a second route in, and it catches more writing than the first. A view stated with no alternative reading beside it is a signal wearing the clothes of a description. A sentence that offers one way to read a figure, and does not mention that other readings exist, has decided for the reader. Such a sentence never names an action, so it walks past the first test untouched, and it fails this one.

What is performance marketing, and why is it a third thing rather than a louder signal?

Performance marketing displays a result in order to attract. Its subject is not the contract at all. Its subject is the person or the service that produced the result, and the contract appears only as evidence.

A signal is a claim about what is to come, and it can be argued with. Performance marketing is a claim about what has already happened, and it invites belief rather than argument. The structural difference is why performance marketing is a third category rather than a fourth degree of the second. A signal has reasoning inside it, so a signal's reasoning can be disputed. A display of past results has arithmetic inside it, often arithmetic that is entirely correct, and no reasoning to dispute at all. The only response left available to a reader is to believe it or not.

No example of it appears here, and that turns out to be the point. The invented record behind everything here holds no outcome for any position anywhere: no result, no run of results, no record of what anything ever did. There is nothing to display. And yet the whole of performance marketing has just been described without one, and that shows how little of it was ever about the contract.

The rules on what may be said about past results in anything published to the public are set by the Securities and Exchange Board of India (SEBI) at sebi.gov.in. Those rules attach to who is publishing as much as to what is published, and they move, so the current requirement is to be confirmed at that source.

Why do three different figures here all read Rs 2,000.00/-?

Before the checks, one piece of housekeeping that would otherwise nag at a reader working through them. Three quantities below are all Rs 2,000.00/- and they are three different things. The three agree by construction, not because one of them was copied into another.

The first is the spot price of the invented reference asset: what it is quoted at today. The second is the strike of the option pair used in the third check, and it equals the spot price because that pair is struck at the money. At the money means exactly that. The third is the exposureThe amount of the underlying a position actually stands against, as against the notional, which is only a multiplier. that one unit of a position stands against. One unit of the underlying at its price is the spot price again by definition. Three different kinds of quantity, one number, and a reader who was not told would be right to wonder which had been pasted over which.

The rest of this guide is four checks. Each one takes a few seconds, each one is worked here on the invented record rather than merely quoted, and each exists because of one specific misreading it catches. The four are habits rather than rules. Each is therefore written as something to do rather than as something not to do.

What is the check that stops a forward price being read as a forecast?

Carry today's price at the financing rate for the period, then see whether the whole of the gap is already accounted for.

On the invented record: Rs 2,000.00/- carried for one year at 6.50 per cent a year is Rs 2,000.00/- plus Rs 130.00/-, and that comes to Rs 2,130.00/-. The carryThe cost of holding the underlying from today until the agreed date, which on this record is financing and nothing else. of Rs 130.00/- is the entire difference between the spot price and the forward price. The asset pays nothing while it is held. There is no residue. Nothing is left over for a view about direction to live in.

The check catches a financing cost being read as an opinion. A reader meets Rs 2,130.00/- against Rs 2,000.00/- and concludes that somebody expects a rise of 6.5 per cent over the year. The forward price says nothing of the kind. The figure says only that buying the asset today and borrowing the money to do it costs Rs 2,130.00/- by that date, so anybody offering to sell forward for materially less is offering money away. The identical forward price would arrive if every party involved were certain the price was about to fall.

The check is best run before it is needed. A forward price nobody has asked about can be taken and the subtraction done anyway. Running it cold matters because the subtraction is then already familiar by the time somebody puts a number in front of the reader and wants a comment on it inside a minute.

One thing this record cannot show. If the asset paid something out while it was held, the payout would subtract from the carry and the forward price could sit below the spot price. The asset on this record pays nothing, and that case therefore cannot be worked.

What is the check that stops a notional being read as a size?

Ask of every headline quantity what sum would actually change hands in the period, then divide the headline by that sum.

The invented arrangement has a fixed payer and a floating payer and no names beyond those two roles. Its notionalThe amount the payments under an arrangement are multiplied by. It is a multiplier and it never changes hands. is Rs 1,000 crore. The fixed rate written into the agreement is 7.20 per cent a year, and the floating benchmark reads 6.00 per cent a year for the first period. Work both legs rather than reading the net straight off. Rs 1,000 crore at 7.20 per cent a year is Rs 72.00 crore gross for the period. Rs 1,000 crore at 6.00 per cent a year is Rs 60.00 crore gross. The two legs differ by Rs 12.00 crore, and that is what the fixed payer hands the floating payer.

Now divide in both directions. The two quotients teach different halves of the same fact. Rs 12.00 crore divided by Rs 1,000 crore is 1.2 per cent, so the sum that moves is 1.2 per cent of the headline. Rs 1,000 crore divided by Rs 12.00 crore is 83.33, so the headline is 83.33 times the sum that moves. The notional never changes hands, and 98.8 per cent of it stays exactly where it is.

The check catches a multiplier written up as an amount at risk. Reading a notional as an amount at risk is the commonest way a written account of one of these arrangements ends up describing something nearly two orders of magnitude larger than the arrangement itself. A reader who runs that division on any figure presented to them as a size, in this subject and well outside it, has taken away something worth more than the fact that produced it.

The value of the arrangement today cannot be established from this record. A value would need a reading of the floating benchmark for every period still to come, and this record holds one reading for one period and nothing at all after it. So the first period net is computable and a value is not.

What is the check that stops a payoff being read as a profit?

Ask of every number on a payoff drawing whether it knows what was paid.

The invented option pair is struck at the money at Rs 2,000.00/-, and the call cost a premium of Rs 180.00/-. At a price of Rs 2,130.00/- the call's payoffWhat a contract delivers at expiry, ignoring what was paid to get into it. is Rs 2,130.00/- less Rs 2,000.00/-, or Rs 130.00/-. The profitThe payoff net of what was paid, which is a different number and often a different sign. at that same price is Rs 130.00/- less the Rs 180.00/- paid, or minus Rs 50.00/-. The profit does not reach nil until the price is Rs 2,180.00/-, the strike plus the premium.

The two numbers describe the same position at the same price, and one of them is positive while the other is negative. Everything hangs on which of the two the writer meant. The check catches a payoff written up as a profit, and that is not a rounding matter: it is a gain recorded on a losing position.

A question travels further than a rule, so keep the check as a question: does this number know what was paid? A payoff line and a profit line are the same shape with the premium between them, so the question bites hardest on a drawing whose vertical axis carries no label. With the axis covered there is nothing left to say which of the two is on the drawing. The premium here is given by the record and is derived nowhere in this guide. Pricing an option needs a measure of how much the referenced price moves about. No such measure sits on this record, and supplying one would mean inventing the most important input in the calculation.

What is the check that stops a small move being read as a small move?

Every move is restated as a share of what was posted rather than as a share of the exposure, with the base said out loud.

The initial margin of 8.0 per cent used below was chosen for teaching and is not a requirement that anybody set. On that invented figure, Rs 160.00/- is the marginWhat is posted against an open position, as against the exposure the position stands on. posted against Rs 2,000.00/- of exposure. A 4.0 per cent adverse move on that exposure is Rs 80.00/-. And Rs 80.00/- is 50.0 per cent of the Rs 160.00/- posted. The same two figures make the exposure 12.50 times what was posted.

Read the two descriptions again, side by side. A move of 4.0 per cent. Half of what was put down. Both are true, both describe one move of Rs 80.00/-, and they sound nothing alike. The gap between those two descriptions is the whole of why the base has to be said. A ratio without its base is not a small number or a large one. Without its base a ratio is not yet a number at all.

The amount actually posted against a position, and the way that amount is worked out, sits with SEBI at sebi.gov.in and with the clearing corporations working under its framework. The requirement varies by contract and by day, and it moves, so any real figure printed here would be out of date before it was read.

The relationship holds at every size, and it is worth seeing laid out rather than asserted. Each row below is the same arithmetic run again: the move as a share of the exposure, the rupees that produces on Rs 2,000.00/- of exposure, and those same rupees as a share of the Rs 160.00/- posted. The exposure stands at 12.50 times what was posted, so the second column of percentages is the first multiplied by 12.50 every single time.

Read the last row slowly, because it is the one the first three lead to. A move of 8.0 per cent on the exposure is the whole of what was posted. No likelihood is attached to any price on this record, so nothing in that row is a statement about how likely such a move is, and none is being smuggled in by arranging four rows in ascending order. The table is arithmetic on a base, and it stops there.

What does a reader who never writes a note do with any of this?

Most people who need these four checks will never publish a line. Those readers are on the receiving end: a treasury assistant reading a lender's morning note, an analyst reading a counterparty's write-up before a call, a household reading something forwarded to a phone by a cousin who works in markets. The checks work identically from the reading side, and they are faster there: the reader is checking one number rather than composing a paragraph around it.

Take the household case, the one that matters most and gets written about least. A message arrives with a headline of Rs 1,000 crore in it and the word arrangement, and it reads as though something enormous is at stake. The second check turns that into one question: what actually moves in the period? On the invented record the answer is Rs 12.00 crore, or 1.2 per cent of the headline. The message may still be about something serious. The headline is now being read at its own size rather than at eighty-three times it.

An analyst uses the same four differently, as a filter on the writing rather than on the figures. Read the last clause of a note first, before anything else. If the last clause names somewhere to stand, everything above it has to be read as support for that clause rather than as description. The note is then evidence of something else. Reading the last clause first is not cynicism about the writer, but reading a document as the kind of document it is. A lender's credit team does the same thing with a valuation it did not commission, and for the same reason.

What does running a check not do?

A reader is most likely to draw the wrong conclusion here, straight after finding something useful, so the limit gets said plainly and early. A check catches a misreading. Catching a misreading is the whole of a check's work. A check does not alter what a position obliges, it does not reduce what is owed at any price, and running four of them does not make anything different to hold.

A reader who runs all four is better at reading and no better protected, and those are two different things. The Rs 80.00/- moves whether or not anybody named the base it should be measured against. The profit at a price of Rs 2,130.00/- is minus Rs 50.00/- whether or not anybody asked what was paid. The checks change what a reader understands about the arithmetic, and the arithmetic was never waiting on that understanding.

One more thing belongs here. Somebody who lost money on a contract they did not fully understand did not fail a test of care. Derivative instruments are difficult, a great deal of the writing about them is built to persuade, and an account of reading carefully sits one sentence away from implying that anybody harmed had been careless. Carelessness is not the explanation. The four checks are offered because they are useful, not as a standard somebody was supposed to have met already.

How Derivative Research Communicates Assumptions and Uncertainty, and where do the four placements go?

Communicating assumptions and uncertainty turns out to be four placements inside the writing itself rather than a paragraph of hedging at the end, and the difference is not stylistic.

The assumption is named where it is used. The arithmetic here assumes the underlying pays nothing while it is held, and that assumption is stated beside every carry figure rather than once at the foot. A reader meeting the same arithmetic on something that does pay will find a subtraction where this walkthrough has none. The base of every ratio is named in the same sentence as the ratio, so 50.0 per cent of what was posted, meaning Rs 160.00/-. The period of every rate is named in the same sentence as the rate, so 6.50 per cent a year, for one year. And the absence is named rather than quietly reached past.

The fourth placement is where most writers have it backwards, so it is worth being exact about. Hedging is a way of avoiding a commitment. Naming an absence is a commitment, and a checkable one. Two sentences make the difference plain. The sentence it may go either way says nothing at all: there is no fact in it a reader could go and verify, so it can never be wrong and can never help. The sentence no likelihood is attached to any price on this record, and nothing in it ranks one price above another commits to something checkable, saying exactly what is available and what is missing, and a reader can hold the writing to it afterwards.

Who sets the things not printed here?

Five requirements are touched by everything above. The authority is named for each of them and the value for none of them, and a sixth row holds the one invented percentage used here so that it can never be mistaken for something an authority set.

The value cell in each row is empty. Each of those is set by the authority printed inside the row, each of them moves, and a figure written out here would be wrong rather than merely old. The disclosure duties, the registration conditions, the margin and the position limits all sit with SEBI at sebi.gov.in. The arrangements under which an interest rate or currency contract may be entered into sit with the Reserve Bank of India at rbi.org.in. Conduct principles separating research from marketing across borders originate with the International Organization of Securities Commissions (IOSCO) at iosco.org, and what applies in India is SEBI's version of them. Every one of these is to be confirmed at its own source before it is relied on.

Should a reader who can run all four checks enter into one of these contracts?

The question arrives here, so it gets answered here. How these instruments work is one question, and whether anybody should hold one is a different question with a different answer for every holder. The distinction is not a hedge, it is not a formality, and it is not going to be softened in the next sentence.

The useful part is naming what somebody would have to know before that question could be answered at all. The exposure the position is being held against, if anything. The most that can be given up if the referenced price moves the wrong way. The whole range of prices that could arrive and how likely each one of them is. And whether this particular reader may take such a position at all, a matter set by SEBI at sebi.gov.in.

None of the first three is held here. No likelihood is attached to any price here. No spread is drawn around any figure. No position anywhere in this record is reported as having worked. No position anywhere in this record is reported as having done anything at all. The fourth belongs to an authority. Understanding a mechanism completely is not a reason to use it.

The reference asset, the option pair and the two-legged arrangement are invented.
Educational material. Not advice on any investment, tax, budget or market position.