Mis-Selling: What It Is, and How It Differs From a Market Loss

Mis-selling is a failure that happens when something is placed with a household: it was told something untrue, or something that mattered was left out, or what was placed did not fit what the seller already knew about it. The test is what happened at the sale, and an outcome arriving later is neither necessary to it nor enough for it.

The word has been worn thin by use. The word gets attached to a holding that fell, to a fee somebody resented, to a claim that came back smaller than expected, to a rate that moved the wrong way. Every one of those is a real disappointment and not one of them is the thing the word names. Mis-selling names a specific failure at a specific moment, and the moment is the sale rather than anything that came later.

The distinction between the sale and the outcome is not pedantry, and it is not a defence of sellers. The precision runs the other way, and it protects households. A household that calls every loss mis-selling has spent the only precise term it had. When something genuinely was mis-sold to it, the sentence it reaches for is the same sentence it has already used four times about things that were nobody's fault. The word stops carrying weight exactly when the household needs it to.

Why does the moment have to be the sale rather than the outcome?

The question runs better the other way round. Suppose outcomes were allowed to decide it. Then two households who were sold the identical thing, in the identical conversation, by the identical person, on the identical afternoon, would have been mis-sold or not mis-sold depending on what happened afterwards. Neither the seller nor the household could see any of that at the time. A test that gives two different answers about one conversation is not a test of the conversation at all.

The everyday version. A wedding caterer tells a family the hall holds four hundred people. The hall holds two hundred and twenty. The family books it, three hundred guests arrive, and half of them stand outside in June. The mistake happened in the conversation, not in June. Change one thing: the caterer said two hundred and twenty, the family booked it knowing that, and then it rained. Standing outside in the rain is a bad afternoon and nobody misdescribed anything. The rain is not evidence about the conversation, and the conversation is not excused by good weather.

Financial products stretch that gap out over years. The stretch is the only reason the confusion survives. The conversation happens on a Tuesday and the disappointment arrives four monsoons later, by which time the conversation is a memory and the disappointment is a number on a statement. The number is the thing in front of the household, so the number is what gets reached for. The tests do not allow it, and that restraint is what makes them tests.

There is a second reason, and it is the practical one. The sale can in principle be established: there were words, there was a document, there was a form, there was a person with a record of what they knew about the household. An outcome years later has a thousand causes, almost none of them the seller's, and no amount of examining it tells anybody anything about the conversation. The tests point at the sale because the sale is the only place where evidence about the sale can be found.

How to Recognise a Mis-Sold Financial Product: what are the four tests?

Four questions. The four tests run on any product anybody has ever placed with a household, and they run on paperwork already held, without asking anybody's permission and without spending anything. The order matters: each question is cheaper to answer than the one before it is expensive to be wrong about.

None of the four asks what happened afterwards, and that is the single design decision that makes them useful. Equally important is what they do not require. The four tests do not require proof that anybody was dishonest. Three of the four are satisfied by ordinary carelessness, by a seller who was pressed for time, by a form filled in and never read back, by a person who explained the part they understood and skipped the part they did not.

Test one: was something said that was not true?

Test one is the cleanest of the four and the rarest. Something was stated as a fact at the point of saleThe moment at which the thing was placed with the household, which is the moment all four tests ask about., the fact was not so, and the document itself shows it was not so. A cover described as paying in full when the schedule limits it. A deposit described as available on demand when its own terms lock it. A charge described as absent when it is printed on the fourth sheet of the schedule.

The reason it is rare has nothing to do with honesty and everything to do with how sales conversations are actually shaped. A confident seller rarely says a false thing outright. A confident seller says a true thing warmly, in a way that leaves a false impression, and then moves on. An outright false statement is easy to catch and easy to avoid, so the failure that actually happens to households is almost never this one.

Take the Bhosale household, an invented family of three, and its health claim. On that record, nothing suggests test one. The room limit of Rs 4,000/- a day was on the schedule. The co-payA stated share of an eligible amount that the household pays itself, here 10 per cent, with the insurer paying the rest. of 10 per cent was on the schedule. The list of non-payable items was on the schedule. Nobody, so far as anything written down shows, said any of those were absent. An unmet test one is not a weakness in the household's position. Test one simply lands where it lands, and a test that landed everywhere would identify nothing.

Test two: was something material left unsaid?

Test two is the one that catches households, and it is the one the Bhosale household's Rs 24,000/- sits under. Something was materialSomething that would have changed what the household did, which is what makes an omission count rather than merely being a gap. if knowing it would have changed the decision. Nobody mentions everything, and a schedule that mentioned everything would be unreadable, so the test is not whether the seller mentioned everything. The test is whether the omission was the one that mattered.

The arithmetic is published to the rupee, so every step of the claim can be rebuilt rather than taken on trust. The bill came to Rs 1,42,000/-. Four deductions were applied. Non-payable items of Rs 8,400/-. Room rent above the limit of Rs 8,000/-. A proportionate deductionA reduction applied across every other payable component in the same proportion by which a stated limit was exceeded, rather than to the exceeded item alone. on everything else of Rs 24,000/-. And a co-pay of Rs 10,160/-, being 10 per cent of the Rs 1,01,600/- that remained. The household paid Rs 50,560/-. The insurer paid Rs 91,440/-. The two add to Rs 1,42,000/-, and 50,560 out of 1,42,000 is 35.6 per cent of the bill.

Now split what the household paid. Three of the four deductions, totalling Rs 26,560/-, follow directly from terms printed in the schedule the household holds: the non-payable list, the room limit itself, and the co-pay rate. The fourth, Rs 24,000/-, is 47.5 per cent of everything the household paid, and it follows from a mechanism nobody described: that exceeding a room limit does not merely reduce the room charge, it reduces every other payable component in the same proportion. The term was disclosed and its consequence was not, and those are two different objects that most schedules only carry one of.

The everyday version makes it obvious. A tailor tells a customer a suit costs Rs 4,000/- and that going up a fabric grade adds Rs 800/-. The customer chooses the better fabric expecting to pay Rs 4,800/-. In that shop, choosing a dearer fabric also moves the stitching, lining and finishing onto a higher rate card, so the bill arrives at Rs 6,600/-. The fabric price was true. The rate card exists. Nobody said the one moved the other, and the number the customer was budgeting against was never the number the customer was going to be asked for.

Test three: was what the seller knew inconsistent with what they placed?

Test three is about fit, and the word for fit is suitabilityWhether what was placed matched what the seller had been told, or had collected, about the household's position and needs.. Test three asks a narrow question, and keeping the question narrow matters. Not whether the product was good. Not whether it turned out well. Whether the seller had information about this household in front of them, and placed something that does not fit that information.

The reason the test exists is that sellers frequently do collect that information. A form asks income. A form asks how many people depend on it. A form asks what else is held and what is owed. Once those answers exist on paper, they are not decoration: somebody wrote them down, and a placement that contradicts them is a placement made in the face of what was known rather than in ignorance of it.

Here is where the household example lands honestly. On this record, test three is not established either way. Nothing in what has been published shows what was collected about the Bhosale household when the floater was taken, and nothing shows it was not. A cover of Rs 5,00,000/- with a room limit of Rs 4,000/- a day is not on its face an odd thing to place with a household of three running on a take-home of Rs 39,800/- a month. Not established is a real answer, and turning it into an accusation because the household is sympathetic would do to sellers exactly what careless sellers do to households.

Documents would change the answer. If a form exists showing what was recorded about this household at the time, it would be readable against what was placed. Whether such a form exists cannot be told from the record. Many households do not hold their own proposal paperwork years later, and that is ordinary rather than a failing.

Test four: would the household have acted differently?

The fourth test asks a counterfactualA question about what would have happened had things been otherwise, here what the household would have chosen had it known the mechanism.: knowing what was not said, would the choice have been different? People assume test four is hopeless. Who, after all, can say what they would have done? And on this record the arithmetic answers test four most cleanly. The result is genuinely surprising and worth working through slowly.

The choice at issue was a room. A room within the Rs 4,000/- a day limit was available and a dearer one was taken. How much did the room actually cost? Hold the bill at Rs 1,42,000/- so that the room decision is isolated from everything else on it, and run three versions.

The middle row comes first. The middle row is the one the household was living in. A household reading a room limit of Rs 4,000/- a day sees one consequence: go over and the excess falls on the household. On that reading it was accepting a cost of Rs 8,000/-, and its share of the bill would have been Rs 28,960/-. The household actually paid Rs 50,560/-. The gap between what the visible term implied and what the invisible mechanism delivered is Rs 21,600/-, and no line anywhere in the schedule prepares a household for it.

Now read the bottom row, the counterfactual test four actually asks about. Had the room been within the limit, the Rs 8,000/- excess and the Rs 24,000/- proportionate deduction both vanish. The two deductions total Rs 32,000/-, and that sum is what the room decision put in motion. The household's own outlay does not fall by the whole Rs 32,000/- though, and the reason is worth stating because it is the kind of detail that gets waved through. Returning Rs 32,000/- to the eligible amount raises the co-pay base, so the 10 per cent co-pay rises by Rs 3,200/-, from Rs 10,160/- to Rs 13,360/-. The household's outlay therefore falls from Rs 50,560/- to Rs 21,760/-, a difference of Rs 28,800/-.

So put the counterfactual to a household whose buffer is Rs 31,320/-, covering 0.73 of one month of the Rs 42,770/- that leaves in an ordinary month. Would it spend Rs 28,800/- on a better room? Rs 28,800/- is 0.67 of a month of everything the household spends. Rs 28,800/- is more than nine tenths of the entire buffer, and paying it leaves Rs 2,520/- standing between this household and the next thing that goes wrong. Nobody in that position chooses a nicer room, and the arithmetic says so without anybody having to be believed.

The arithmetic is what makes test four answerable. Test four is not asking anybody to remember a state of mind. It is putting a number against a buffer and observing that the number is larger than the buffer. Where a counterfactual can be settled by arithmetic rather than by recollection, it stops being the weakest of the four tests and becomes the strongest.

Does a loss have to have happened at all?

No, and this is the cleanest proof that the tests are about the sale. Something can be mis-sold and then, by luck, do perfectly well. Something else can be sold impeccably and then lose money. Mis-selling and losing money are two independent things that people collapse into one because they usually meet the word for the first time on a day the second has happened.

The practical consequence is that a household holding something unsuitable does not have to wait for it to hurt before the sale is worth examining. If a form was filled in on the household's behalf, if a fact was stated that the document contradicts, if a consequence was never described, that already happened. Whether it costs anything is a separate question with a separate answer, and the second one may take years to arrive or may never arrive at all.

The reverse consequence is the one this guide keeps returning to. A statement showing less than was put in is a statement showing less than was put in. Nothing in it is evidence about a conversation. Reaching for the word mis-selling on the day a number goes down is reaching for the wrong tool, and the cost of using it is that the right one is no longer available.

What does a signature actually settle?

Somebody signs a form, watches money go the wrong way, says something about it, and is told that they signed. The exchange happens constantly, and what a signature settles has two halves, both of them exact rather than comforting.

A signature settles something real. A signature is not nothing and it should not be waved away. Signing a confirmation of having received and read a scheduleThe document setting out the specific terms of one arrangement, such as the amount of cover, the limits and the stated rates. makes the printed terms of that schedule part of the arrangement between the household and whoever issued it. A room limit of Rs 4,000/- a day is then an acknowledged term. A co-pay of 10 per cent is an acknowledged term. Arguing later that they were never seen is arguing against a document with the household's own name at the bottom, and that argument is a hard one.

The acknowledged terms are exactly why the Bhosale household's complaint asks for Rs 24,000/- and not Rs 50,560/-. Rs 26,560/- of its own loss follows from terms that were printed and acknowledged, and the household does not contest that part. A complaint that asks for the part it can evidence is a different document from one that asks for everything, and the difference is not modesty, it is precision.

Now the second half, the part almost nobody is told. A signature settles what the document actually said. A signature does not settle what the document did not say. No document can bind anybody to a consequence it never states. Test two is entirely about what was not said. So a signature confirming that a schedule was read establishes nothing at all about a mechanism the schedule does not contain.

The limit of a signature is not a loophole and it is not a clever argument. It is the plain difference between a termWhat a document states. A limit, a rate, a list of exclusions, each written down and readable. and the consequence of a term. The schedule says the limit is Rs 4,000/- a day. The schedule does not say what happens to the other Rs 1,33,600/- of a bill once that limit is passed. The information required to arrive at Rs 24,000/- is not on the schedule, so a household reading that line carefully, twice, with a pen would still not arrive at it.

Which brings the dignity point round. Almost everybody signs. The schedules are long, they are written for people who already know the machinery, and they are handed over at a moment when a household is thinking about a person who is unwell rather than about proportionality clauses. Signing a document that was not fully absorbed is the most ordinary thing in the world, and it is not the same as having agreed to something that document never described.

Why is mis-selling so hard to establish?

Because of one asymmetry, and once it is seen the whole difficulty makes sense. The evidence of what was written is complete. The document exists, it is dated, it says what it says, and it will say the same thing in ten years. The evidence of what was said is almost nothing. A conversation at a counter or across a kitchen table leaves no record, no timestamp and no second copy.

So the four tests sit on two very different footings. Test one asks about a statement and the statement was spoken. Test three asks about what the seller knew, and that usually lives on a form somebody else keeps. Test two, by contrast, asks about an absence, and an absence in a document can be shown by holding up the document: the mechanism is not in the schedule, and the schedule is right there. Test four asks about a decision, and where a household's own figures answer it, they answer it out of arithmetic rather than memory.

The asymmetry is why the Bhosale household's strongest ground is the two tests that lean on documents and arithmetic rather than the two that lean on recollection. The shape is not a coincidence about this household. It is general, and it is the single most useful thing to know before anything is written down about a household's own situation.

There is a second difficulty that has nothing to do with evidence. People treat a settled outcome as closed. Once a claim has been paid, once a figure has been transferred, once the matter feels finished, revisiting it feels like reopening something that was over. The pull to leave a closed matter closed, and the way whatever has been settled turns into the point from which every later gain and loss gets measured, is a finding of Kahneman and Tversky, and it explains why a great many grievances are never written down at all. Nothing about that is a criticism of anybody who let a matter rest.

What do the four tests actually say about this household's record?

Time to run all four at once, honestly, without deciding in advance what the answer should be. Test one: nothing on the record suggests anything untrue was said, so it is not met. Test two: the mechanism producing Rs 24,000/- was never described anywhere in the document the household holds, and knowing it would have changed a decision worth Rs 28,800/-, so it is squarely met. Test three: nothing establishes what was collected about this household or what was placed against it, so it is open in both directions. Test four: the arithmetic answers it, and it answers it against the room decision.

One met. One answerable. Two not established. One met, one answerable and two not established is what an honest application of four tests to a real record usually produces, and anybody promising a clean four out of four is describing a courtroom drama rather than a set of papers.

What is mis-selling definitely not?

How does a household actually use any of this?

Not by reaching a verdict. Nobody working from one household's own papers can reach one. By doing something narrower and more useful: sorting its own paperwork into what is contestable and what is not, before saying anything to anybody.

The four tests run over what a household holds, in writing, on one sheet. For each of the four, the entry is either the specific thing that can be pointed at or the words not established. Where the answer is a printed line in a document, the document and the line are noted. Where the answer depends on a conversation, that is noted too, and noted honestly as depending on a conversation. The sheet costs nothing, takes an evening, and needs no professional.

The sheet produces a shape. In the Bhosale household's case the shape was: three deductions totalling Rs 26,560/- follow printed terms and are not contestable, one deduction of Rs 24,000/- follows a mechanism that is not in the document, and the counterfactual behind it is worth Rs 28,800/- against a buffer of Rs 31,320/-. The shape is why the letter written on 20 June asks for Rs 24,000/- rather than Rs 50,560/-, and asking for the smaller number is the stronger move rather than the meeker one.

The reason is simple. A request for Rs 50,560/- asks whoever reads it to disregard terms that were printed and acknowledged. Nobody will do that, and the reply that comes back deals with the printed terms and never reaches the part that matters. A request for Rs 24,000/- names one component, points at its absence from a document anybody can check, and leaves nothing else to argue about. The reply that arrived on 4 July restated the terms in the schedule and declined. A perfectly ordinary reply, and exactly why a further rung exists.

Two things sit outside all of this, and they need saying flatly. Whether to complain is the household's own decision, and it depends on matters no general account can see, including whether the time can be afforded. The outcome at a forum sits outside all of this as well. The route exists and can be described; the result at the end of it is not something anybody can promise, and anybody who does promise one is stating something they cannot know.

The Bhosale household, Meghna Bhosale, Ashok Bhosale and Ira Bhosale are invented.
Educational material. Not advice on any investment, tax, budget or market position.