Mental Accounting: Money Treated by Bucket, and What That Costs

Mental accounting is treating money as though it arrived in labelled buckets that cannot be netted against each other. The labels are supplied by the person, the rupees are identical, and the separation carries a price that can be written down in rupees. Mental accounting explains house money, break-even chasing and narrow framing at once, and it is why a deposit can sit beside an expensive borrowing for years.

The whole subject rests on a fact so plain that it sounds like a trick question when said out loud. Money is fungibleInterchangeable, so one unit spends exactly like another whatever its source.. A rupee that arrived as salary and a rupee that arrived as a booked gain are the same rupee, and once the two sit in the same place nothing can separate them again. Mental accounting is what happens when a careful, ordinary person behaves as though that plain fact were false. Not a careless person, and not a person short of arithmetic. A separation of this kind is usually built by somebody being sensible.

Why does money being fungible matter before anything else?

Start away from finance entirely. A vegetable seller outside a bus stand keeps the day's takings in three tins: one for stock tomorrow, one for the room rent, one for the child's fees. A tin makes it slightly harder to spend the rent on stock, so the arrangement is a good one and the tins do real work. Now suppose the fee tin holds Rs 4,000/- while the seller borrows Rs 4,000/- from a lender at a rate that would horrify anybody who wrote it down, in order to buy stock. Nothing is wrong with the tins. Nothing is wrong with the borrowing on its own. The cost is created entirely by the fact that the two tins are not allowed to look at each other.

The tins show the shape of the whole subject, and the shape repeats at every scale. Two arrangements, each of them defensible, and a price that lives in the space between them because a wall was put up and never questioned again. The seller does not need a lesson in interest rates. The seller needs one question asked out loud: what does that wall buy? Nobody ever asks that about a tin.

The claim being made is narrower than it looks. Separating money is not silly, and a person who keeps tins is not confused. The claim is only this: the money itself supplies no reason to treat two piles differently, so any difference in how they are treated has to be justified by the person. Once that is accepted, every effect below is just a catalogue of the reasons people supply without noticing they are supplying them.

What is a mental account, and where do the labels come from?

A mental accountA pot of money a person keeps separate in their head, with its own rule for what enters and its own moment of being judged. is a pot of money a person keeps apart in their head, complete with a rule for what may enter it, a rule for what it may be spent on, and a moment at which it is closed and judged a success or a failure. The idea was set out by Thaler in Mental Accounting and Consumer Choice in Marketing Science in 1985, and taken further in Mental Accounting Matters in the Journal of Behavioral Decision Making in 1999. The description that has lasted is deliberately unflattering: people run a small internal book-keeping system, and it is not a very good one.

Notice the three parts. Each one causes a different problem later. An account has an opening balance, and the opening balance is where a reference level comes from. An account has an entry rule, and the entry rule decides what counts as belonging to the account. And an account has a closing moment, the moment at which the account is judged. An account that has been opened has to be closed, and nobody enjoys closing one below where it started, so the damage is done almost entirely by the closing moment.

Where do the labels come from? Three sources, mostly. The first is where the money arrived from, so a bonus gets treated unlike a salary of the same amount. The second is what the money was set aside for, so an education pot resists being touched. The third is the container the money physically sits in, so a deposit with a separate certificate feels less spendable than the same amount in a current account. Every one of these three is a fact about the container and not a fact about the rupee.

What is house money, and why is it spent differently?

House moneyA gain treated as though it still belonged to wherever it came from rather than to the person holding it, and therefore risked or spent more freely. is the name for a gain that gets filed in its own account and is then treated as though it never quite became the holder's own. Thaler and Johnson set out the effect in Management Science in 1990. A prior gain changes how the next choice is taken, and that is exactly what should not happen if only the total matters. The phrase comes from a gambling table, where a run of winnings is described as playing with the house's money, and the tell is that nobody ever calls their salary the employer's money.

Take the case. On 12 October Meera Sundaram sells Suvarna Chemicals Limited whole at Rs 4,60,000/- against a cost of Rs 4,00,000/-. The booked gain is Rs 60,000/-, or 15.0 per cent on what she paid. The Rs 60,000/- now sits in her bank account, indistinguishable from the Rs 25,000/- that arrives every month by standing instructionAn order left with a bank to move a fixed amount on a fixed day without being asked again.. If the Rs 60,000/- gets spent on something the Rs 25,000/- would never have been spent on, the label has changed the decision and the money has not changed at all.

There is a second, quieter version of the same thing, and it is worth naming because it does more damage than the spending version. A gain filed as winnings gets risked more readily than a wage would be. The account it sits in has a cushion in it, so a loss inside that account does not feel like a loss at all, it feels like giving back something that was never really held. An account with a cushion in it is an account in which the next decision is taken less carefully, and the cushion is entirely a matter of book-keeping.

What is break-even chasing, and why must the account close at zero?

Break-even chasingRefusing to close a pot of money until it has returned to the level it started at. is what happens when the closing rule of a mental account meets a number that has gone the wrong way. The account was opened at a level. The account is judged at the moment it closes. So the account is not closed. The account stays open until it can be closed at zero or better, and staying open converts a decision about the future into a decision about a number from the past.

The case shows it exactly. On 12 October Meera keeps Kesari Logistics Limited, then standing at Rs 1,95,000/- against a cost of Rs 3,00,000/-, and says she will sell it when it gets back to Rs 3,00,000/-. Read that sentence again. Nothing in it says anything whatsoever about Kesari Logistics Limited. The sentence states only the level at which one mental account may be permitted to close. The Rs 3,00,000/- is a fact about a purchase, not a fact about a holding, and no amount of staring at it will make it become one.

Two things follow from that and both are worth stating plainly. The first is that break-even chasing is not stubbornness or greed. Break-even chasing is book-keeping. An account has been opened, and the person is simply refusing to write down a closing entry they do not like. The second is that the rule has a cost, and here that cost is countable. Between 30 September and 31 March following, Kesari Logistics Limited fell a further 20.0 per cent, from Rs 1,95,000/- to Rs 1,56,000/-, a further Rs 39,000/- gone while the account stayed open waiting for permission to close. One case is not evidence that any rule works, and it does not establish that closing the account would have been the better course.

What is narrow framing, and what does it cost?

Narrow framingJudging one decision on its own rather than together with the others it sits beside. is the habit of judging each account by itself instead of judging what the whole position did. Barberis and Huang set out the asset pricing version in the Journal of Finance in 2001, and the mechanism is the same one running underneath every effect above: an account has a boundary, and everything outside the boundary is invisible when the account is being judged.

Work it through with the four positions in the case at 30 September. Vindhya index scheme up Rs 36,000/-, Nilgiri mid-cap scheme down Rs 45,000/-, Suvarna Chemicals Limited up Rs 60,000/-, Kesari Logistics Limited down Rs 1,05,000/-. Judged one at a time, two of those four are painful enough that a person may well act on them. Judged as one set, the cost is Rs 13,00,000/- against a value of Rs 12,46,000/-. The whole set is Rs 54,000/- down, or 4.2 per cent. The offsetting between the four is real, it is arithmetic rather than opinion, and it is invisible to anybody who looks at one line at a time.

What is Asset Segregation, and what does it hide?

Asset segregationHolding things apart so that their gains and losses are never added together. is the structural version of everything above. Asset segregation holds two things in separate accounts, and their gains and losses are then never added together. The person never sees the one number that describes their position. Narrow framing is the habit of looking at one line. Asset segregation builds the accounts with no place where the lines meet, so looking at one line becomes the natural thing to do.

The same four positions make it visible. Segregated, the reading is two gains worth Rs 96,000/- in one account and two losses worth Rs 1,50,000/- in another. Integrated, the reading is Rs 96,000/- less Rs 1,50,000/-, leaving Rs 54,000/- down. Both readings are correct, so segregation is not an arithmetic error. Segregation is a refusal to do one further subtraction that only takes a second. Nothing about a segregated set of accounts is wrong on inspection, and that is exactly what makes segregation hard to spot.

Segregation has a cousin worth naming in the same breath. The cousin works the other way round. People will often integrate two losses so that the pair is felt once instead of twice, and separate two gains so that the pair is enjoyed twice instead of once. The direction changes but the machinery does not: an account boundary is being drawn where it makes the reading feel better, and the reading is then trusted as though the boundary had been there all along.

Mental Accounting vs Asset Allocation: what actually separates them?

Here is where this subject most often gets misread, and it is worth slowing down. Somebody who keeps a reserve apart from long-term money is doing something sound, and from outside it looks exactly like somebody who keeps a deposit apart from a borrowing. Two pots, one wall, no netting. The picture is the same. The two are separated not by the shape of the arrangement but by whether a decision was ever taken, and a decision is invisible from outside.

The test is one question, and it fits in six words. What is the wall for? A deliberate allocation has an answer to that question, and the answer names something the separation buys. A mental account has no answer. The wall was never put there on purpose; it grew out of a label. Asked why a reserve is kept apart, a person gives a reason. Asked why a deposit sits beside a card borrowing, a person usually pauses. The arrangement was never a decision at all.

One more thing about that comparison, and it guards against a common overcorrection. The test does not say that a good separation is free. The reserve costs something too: money sitting in a reserve is not doing whatever else it might have done, and that is a real price. The price of a deliberate separation was accepted in exchange for something named. The price of a mental account is paid for nothing in particular. How money should be allocated to purposes is a separate subject with its own methods, set out under strategic asset allocation; the test above only tells a deliberate allocation apart from a mental account.

What does the separation actually cost, in rupees?

A cost that stays vague never changes anything, so the cost of this separation gets priced exactly. Meera holds a deposit of Rs 2,40,000/- earning 6.5 per cent a year. She also carries Rs 1,80,000/- standing on a card at 36.0 per cent a year. Setting the deposit against the borrowing would stop 36.0 per cent running on Rs 1,80,000/-, saving Rs 64,800/- a year. The same move would give up 6.5 per cent on that same Rs 1,80,000/-, costing Rs 11,700/- a year. The separation therefore costs Rs 64,800/- less Rs 11,700/-, being Rs 53,100/- a year, or Rs 4,425/- every month it stands.

People get this part wrong, so sit with what that arithmetic does not say. Nothing about either arrangement is unsound taken on its own. A deposit at 6.5 per cent is a perfectly ordinary deposit and there is nothing careless about holding one. A borrowing on a card at 36.0 per cent is an ordinary borrowing on ordinary terms, and it is the terms rather than the borrower that make the rate what it is. The Rs 53,100/- exists only in the space between the two accounts, and it is charged for a wall rather than for anything either arrangement is doing. The wall has a price; whether to clear a borrowing, break a deposit or rearrange anything at all is a separate decision entirely.

One more idea earns a name before the control below. The idea is what makes the number worth stating at all. The Rs 53,100/- is a carrying costWhat it costs each year to keep two arrangements separate rather than netted against each other.: it is not a one-off, it is not a loss that has already happened, and it does not fade. The carrying cost is a rent, charged annually, on a wall that nobody built on purpose. Most of the effects in behavioural finance cost money once. A carrying cost keeps costing money every year until the question gets asked.

The straight line is the reason the idea is usable rather than merely true. If the cost were a fixed penalty for having two accounts, the only response available would be an all-or-nothing one, and most people would take neither option. Because it scales, the arithmetic works on any position at all, including one that is only partly walled off. The rent on any wall comes out of one multiplication: the gap between the two rates, applied to whatever amount is sitting behind it.

When is keeping money in buckets correct?

Often, and this needs saying without hedging. A bucket is correct wherever the bucket buys something that the netting would destroy. The clearest case is a reserve. Meera keeps Rs 1,10,000/-, being two months of a Rs 55,000/- monthly outgo, apart from everything else. The reserve buys certainty that a bad month is met out of the reserve rather than by selling a holding at whatever the price happens to be that week. Net the reserve against anything and that certainty disappears. The wall was chosen, the purpose is stated, and the price of keeping the reserve was accepted in exchange, so the reserve is not mental accounting at all.

Two more buckets are worth defending in the same terms. A pot kept apart because the money is committed to somebody else, such as fees due next term, buys certainty that the commitment gets met. A pot kept apart because reaching it is deliberately awkward buys friction. Friction is a genuine purchase for anybody who knows what they are like at eleven at night. In every one of these cases the answer to the question exists, and it names something. The wall has become a mental account only when the answer is a pause.

How does an adviser or a lender read this on one sheet?

Three people read the same position, and each one uses it differently. Knowing how each of the three reads it matters to anybody who is one of them or is being read by one. An adviser such as Devika Rao at Palash Advisory Services Private Limited reads a position sheet across accounts rather than down one. Reading across is how she sees what a person holding one statement at a time cannot. She is looking for a rate on one side and a higher rate on the other with a wall between them, and the conversation she opens is not about what to do. The conversation opens on one question: what is this separation for? An adviser who asks that question has done the work; an adviser who arrives with a rearrangement has skipped it.

A lender reads it the other way. Faced with an application, a lender looks at what is held and what is owed together. A household holding Rs 2,40,000/- while carrying Rs 1,80,000/- at 36.0 per cent is describing how decisions get taken there, not only how much money there is. And a person deciding alone, with no adviser and no committee, can do the same reading in about four minutes: list every pot on one sheet, write the rate beside each, and look for the largest gap with a wall across it. The exercise costs nothing and needs no method beyond subtraction.

Meera Sundaram, Devika Rao, Palash Advisory Services Private Limited, the Vindhya index scheme, the Nilgiri mid-cap scheme, Suvarna Chemicals Limited and Kesari Logistics Limited are invented.
Educational material. Not advice on any investment, tax, budget or market position.