Skip to content
Menu
Finabase Learn

Investor psychology Lesson 5 of 15

Articles · Behavioral finance

Mental accounting: how labels on money change what investors do

What mental accounting is, the experiments behind it, how house money, dividend income and per-position thinking distort a portfolio, and how to add the accounts up.

On this page
  1. The experiments
  2. How it shows up in investing
  3. Adding the accounts up: a worked example
  4. What helps
  5. How Finabase helps
  6. Frequently asked questions
  7. Summary
  8. Check your understanding
  9. Exercise

Mental accounting is the set of habits people use to organize money in their heads: sorting it into separate "accounts" by where it came from or what it is for, and then treating each account by its own rules. The term is Richard Thaler's. A tax refund is spent more freely than a paycheck. Winnings are risked more readily than savings. A stock is judged against the price paid for it, alone, as if the rest of the portfolio did not exist.

A dollar is a dollar wherever it sits. Economists call this fungibility, and mental accounting is the many ways people fail to treat money as fungible. Some of those ways are useful: a labeled savings account really does help people save. This lesson is about the ways that cost investors money, the experiments that documented them, and a worked example that adds the accounts back up.

The experiments

The same $5

Amos Tversky and Daniel Kahneman asked people to imagine buying a jacket for $125 and a calculator for $15. The salesman mentions that the calculator is on sale for $10 at a branch 20 minutes away. Would you make the trip? 68% of 93 respondents said yes. A second group got the same story with the prices swapped, so the calculator cost $125 and was $120 at the other branch. Only 29% of 88 would go.

Both groups were offered $5 for 20 minutes. They answered differently because they filed the $5 against the price of the item, where it is a third off in one case and 4% off in the other.

The result has been retested. The Many Labs 2 project repeated it in 2018 with 7,228 participants and updated items and prices. It replicated in direction, at less than half the original size: 49% would travel for the saving on the cheap item against 32% for the costly one.

The lost ticket

In the same paper, people were asked to imagine arriving at a theater. One group had lost a $10 bill on the way: would they still pay $10 for a ticket? 88% said yes. The other group had already bought the ticket and lost it: would they pay $10 for another? Only 46% said yes.

The loss is $10 in both versions. The lost ticket is charged to the "theater" account, which makes the play cost $20. The lost bill is charged to general funds, and the play still costs $10.

House money and breaking even

Richard Thaler and Eric Johnson studied how a prior gain or loss changes the next risky choice. Told they had just won $30, 70% of subjects accepted a 50-50 bet to win or lose $9. Told they had just lost $30, only 40% accepted the same bet. Winners were gambling with "the house's money". The winnings sat in a separate account where losses did not count as losses of their own money.

Losers were not simply more cautious, though. In Thaler and Johnson's experiments, subjects who had just lost mostly turned down a fair coin flip, yet a large majority, around 70% in their reports, preferred a long shot that could win back exactly what they had lost to a sure amount with the same expected value. The long shot offered something the fair bet did not: a chance to close the losing account at zero. This is the break-even effect described in loss aversion.

How it shows up in investing

One account per stock. Hersh Shefrin and Meir Statman built their explanation of the disposition effect on this. When an investor buys a stock, a mental account opens at the purchase price. Selling above it closes the account at a profit. Selling below it closes it at a loss, permanently. So losers stay open. The portfolio as a whole, which is the only thing that pays for retirement, is not the unit being judged.

House money in a rally. After a strong year, gains feel like winnings, and investors take risks with them that they would not take with "their" money: a concentrated bet, an option position, a larger size. The gain is as spendable as any other dollar in the account, and losing it is a real loss.

Dividends as income, capital as untouchable. Shefrin and Statman argued in 1984 that investors prefer cash dividends partly as a self-control device: "spend the dividends, never touch the capital" is an easy rule to keep. Malcolm Baker, Stefan Nagel and Jeffrey Wurgler later confirmed the behavior in brokerage and household data. Investors were far more likely to spend from dividends than from capital gains, and net withdrawals rose about one-for-one with ordinary dividends. A stock that pays a $1 dividend falls by about $1 when it does, so a shareholder who needs $1 is in the same position whether the company pays it or she sells $1 of shares, taxes aside. Total return explains the arithmetic. Treating the two differently leads people to choose investments by yield rather than by total return.

Buckets and layers. Shefrin and Statman's behavioral portfolio theory describes how people actually build portfolios: as a pyramid of layers, each tied to a goal, with a safe layer to avoid poverty and a risky layer for a shot at riches. The layers are managed separately, so nobody checks how they combine. A household can hold cash earning 4% in the safe layer while paying 24% on a credit card, because the two are in different accounts.

Splitting evenly across whatever is on the menu. Shlomo Benartzi and Richard Thaler found that retirement-plan participants tend to spread contributions evenly across the funds offered, so the menu decides the allocation. The pilots' plan at TWA offered five stock funds and one bond fund, and participants held 75% in stocks. The University of California plan offered one stock fund and four bond funds, and participants held 34% in stocks. In an experiment with University of California employees, groups shown different menus chose 43% or 68% in stocks depending on how many of the funds were stock funds. Each fund was treated as an account deserving its share.

Adding the accounts up: a worked example

An illustrative household keeps its money in five places and thinks about each separately.

Mental account Balance The year Gain or cost
"Safe" savings $10,000 +4.0% +$400
Index fund (retirement) $20,000 +10% +$2,000
"Play money" stock picks $10,000 −12% −$1,200
Old employer stock $6,000 −15% −$900
Credit card balance −$4,000 24% APR −$960
The household $42,000 −$660
One year, five mental accounts, one household: what each line earned or cost “Safe” savings (4.0%) +$400 Index fund +$2,000 “Play money” stock picks −$1,200 Old employer stock −$900 Card balance (24% APR) −$960 The household as a whole −$660 Illustrative numbers from the worked example on this page.
The worked example: what each mental account earned or cost over one year, and the total. Illustrative numbers.

Account by account, the story this household tells is comfortable. The savings are safe. The retirement fund had a good year. The play money is down, but it is only play money. The employer stock "will come back". The card is a separate matter.

Added up, a household with $42,000 of net worth lost $660 in a year in which a plain index fund rose 10%. Three decisions that mental accounting made invisible account for all of it:

  1. Holding 4% cash while owing 24%. Paying the $4,000 card balance from savings would give up $160 of interest and save $960: $800 a year, with no risk.
  2. "Only play money." The label made a 12% loss on a quarter of the household's assets feel smaller than it was. $1,200 buys the same groceries whichever account it left.
  3. The stock with a history. The employer shares are held against the price at which they were received. Held against nothing, they are $6,000 in one company.

None of this says the household should have no savings buffer or no fun with a small account. A cash reserve has a purpose that an interest rate does not capture, and labels help people stick to plans. The exercise is to see the total at least once a year, so that each label is something the household chose to pay for.

What helps

There is good evidence that mental accounts exist and little evidence on how to manage them well, so these are reasoned practices, not tested cures.

  • Keep one statement of everything. Assets, debts and the year's change in each, on a single page, as in the table above. The comparison between lines is the point.
  • Judge by total return. Price change plus dividends, for the whole portfolio. Income is one way a return arrives, and the label does not make it safer.
  • Set risk for the portfolio, then check each layer against it. If the plan says 60% in stocks, the play account and the employer stock are part of that 60%.
  • Treat gains as yours. A rule for position size that uses the account's current value (see position sizing) does not distinguish between the money you deposited and the money you made.
  • Use the bias where it helps. Labeled accounts and automatic transfers make saving easier, and nothing in the research says to give that up.

How Finabase helps

A Finabase portfolio lists every position with its cost, market value, day gain and total gain, with the totals on the same screen, which is the "one statement" view for the investments you track there. The dividends tab on a symbol page lists each payment, and the header above it shows the price change over the year, so yield and price change can be read together. The paper trading page shows account value, cash, market value, unrealized gain and realized gain in one row: a small model of looking at the whole instead of the parts.

Frequently asked questions

What is mental accounting in simple terms?

It is treating money differently depending on where it came from or what you have labeled it for, even though every dollar buys the same things. Spending a bonus more freely than salary is the everyday example.

What is the house money effect?

It is taking more risk with recent winnings than with other money, because the winnings feel as if they belong to someone else. In Thaler and Johnson's experiment, 70% took a coin-flip bet after a win and 40% after a loss.

Is mental accounting always bad?

No. Earmarking money for rent, an emergency fund or retirement helps people control spending and stick to goals. It becomes costly when the labels hide a comparison, such as low-interest savings beside high-interest debt, or when each investment is judged alone.

Are dividends better than selling shares for income?

Before taxes and trading costs they are equivalent: a dividend lowers the share price by about the amount paid. Preferring dividends is a mental-accounting habit that can help with self-control, but it is not a source of extra return, and it can skew a portfolio toward high-yield stocks.

Summary

Mental accounting is sorting money into labeled accounts and judging each alone. Experiments show people will drive 20 minutes to save $5 on a $15 item but not on a $125 item, replace a lost $10 bill but not a lost $10 ticket, gamble more freely with winnings, and reach for long shots when behind. In investing the same habit opens an account for every stock at its purchase price, treats dividends as spendable and capital as sacred, and lets the fund menu set the allocation. The worked example shows the cost of never adding up: five comfortable stories and a household that lost money in a good year. One page with everything on it, read once a year, is the simplest correction.

Check your understanding

5 questions on this lesson. Each answer comes with the reason, and nothing is scored.

  1. People would drive 20 minutes to save $5 on a $15 calculator but not on a $125 one. What does this show?

    1. People cannot do arithmetic
    2. The saving is judged against the item’s price, in that item’s mental account, not as $5 for 20 minutes
    3. Calculators are more important than jackets
    Show the answer

    B. The saving is judged against the item’s price, in that item’s mental account, not as $5 for 20 minutes The trade is identical in both versions. Filing the $5 against a $15 price makes it look large, and against $125 it looks small.

  2. After a 30% gain, an investor says “I’m playing with house money now” and doubles the size of the next trade. What is the error?

    1. None: gains should be risked more freely
    2. The gain is treated as a separate account that does not count as real wealth, though losing it reduces wealth exactly as much
    3. Position size should never change
    Show the answer

    B. The gain is treated as a separate account that does not count as real wealth, though losing it reduces wealth exactly as much In Thaler and Johnson’s experiment 70% accepted a bet after a win and 40% after a loss. A dollar of gain is as spendable as a dollar deposited.

  3. A retiree needs $4,000 a year from a $100,000 holding. Before taxes and costs, how does receiving a 4% dividend compare with selling $4,000 of a stock that pays none?

    1. The dividend leaves the retiree better off because the shares are untouched
    2. They are equivalent: the dividend lowers the share price by about the amount paid
    3. Selling shares is always worse
    Show the answer

    B. They are equivalent: the dividend lowers the share price by about the amount paid Either way $4,000 leaves the holding. “Never touch the capital” is a self-control rule, not a source of return.

  4. A household holds $10,000 in savings at 4% and owes $4,000 on a card at 24%. Roughly what does keeping them in separate mental accounts cost per year?

    1. About $160
    2. About $800
    3. About $960
    Show the answer

    B. About $800 Paying the card from savings gives up $160 of interest and saves $960 of interest: $800 net. The labels “safe savings” and “card” hide the comparison.

  5. One retirement plan offers five stock funds and one bond fund; another offers one stock fund and four bond funds. What did Benartzi and Thaler observe?

    1. Participants in both chose about the same share of stocks
    2. Participants held far more in stocks where the menu had more stock funds, because they spread money across the funds offered
    3. Participants ignored the menu and copied their coworkers
    Show the answer

    B. Participants held far more in stocks where the menu had more stock funds, because they spread money across the funds offered The plan with mostly stock funds had 75% in stocks and the other 34%. Treating each fund as an account that deserves its share lets the menu choose the allocation.

Exercise

One task to do in Finabase. Nothing is graded or saved.

Open Paper trading (the sample account when signed out, or your own) and read the summary row: Account value, Cash, Market value, Unrealized gain and Realized gain. First look only at the Positions table and say how the account is doing. Then use the summary: what share of the account is cash, and what is the gain on the whole account compared with the amount funded?

Open Paper trading
What you should find

The Positions table shows only what is invested, so it usually looks better or worse than the whole. Cash that earns nothing pulls the account’s return below the positions’ Gain %. The honest figure is Account value against the amount funded. That is the single-statement view: every account, including the idle one, counted together.

Sources and further reading

  1. Thaler — Mental accounting and consumer choice, Marketing Science 4(3), 1985
  2. Thaler — Mental accounting matters, Journal of Behavioral Decision Making 12(3), 1999
  3. Tversky and Kahneman — The framing of decisions and the psychology of choice, Science 211(4481), 1981
  4. Thaler and Johnson — Gambling with the house money and trying to break even: the effects of prior outcomes on risky choice, Management Science 36(6), 1990
  5. Shefrin and Statman — Explaining investor preference for cash dividends, Journal of Financial Economics 13(2), 1984
  6. Baker, Nagel and Wurgler — The effect of dividends on consumption, Brookings Papers on Economic Activity 2007(1)
  7. Benartzi and Thaler — Naive diversification strategies in defined contribution saving plans, American Economic Review 91(1), 2001
  8. Shefrin and Statman — Behavioral portfolio theory, Journal of Financial and Quantitative Analysis 35(2), 2000
  9. Klein et al. — Many Labs 2: investigating variation in replicability across samples and settings, Advances in Methods and Practices in Psychological Science 1(4), 2018