Mental Accounting: Why $20 More on a Shirt Can Feel Worse Than $20 More on Dinner

Money is fungible; the way people track it is not

In standard economics a dollar is a dollar. It buys the same amount whether it goes on dinner, rent or a shirt, so there is no reason for $20 to feel different depending on where it is spent. Richard Thaler's theory of mental accounting, one of the contributions cited when he received the 2017 Nobel Memorial Prize in Economic Sciences, starts from the observation that people do not behave this way. The Royal Swedish Academy's scientific background to the prize describes mental accounting as "a psychological theory of how limited cognition affects spending, saving, and other household behavior," in which people group spending into categories such as housing, food and clothes, and each category "has its own budget and its own separate reference point."

The result is what the Academy calls "limited fungibility between the accounts": the value a person puts on a given amount of money depends on which account it is assigned to. The background document tells the story, from Thaler and Sunstein's Nudge, of an actor who kept cash in labelled jars for rent and utilities and asked a friend for a loan because the jar marked for food was empty.

One real-world test comes from gasoline. Justine Hastings and Jesse Shapiro studied household fuel purchases when gasoline prices swung sharply around 2008. Their NBER working paper reports that when prices rose, households switched to lower-octane fuel "to an extent that cannot be explained by income effects," and that across specifications the authors "consistently reject the null hypothesis that households treat 'gas money' as fungible with other income." The Nobel background notes that when prices fell by about half, the shift toward premium gasoline was 14 times what a standard demand model predicted, with no similar shift in other product categories. A cheaper gas bill freed up money, but people spent it upgrading the gas.

Why $20 on a shirt can feel larger than $20 on dinner

Mental accounting does not claim to explain every feeling about spending, and the popular "dinner versus shirt" framing goes further than the research. But three findings from the literature fit the pattern well.

Separate budgets. If "going out" and "clothes" are different accounts, each with its own budget, then $20 more on a meal is judged against the eating-out budget and $20 more on a shirt against the clothing budget. Someone who spends freely on restaurants and rarely on clothes can find the second $20 harder to justify, simply because the clothing account is smaller and more closely watched.

Reference prices and transaction utility. Thaler split the value of a purchase into acquisition utility, the value of the item minus its price, and transaction utility, the difference between the price paid and the price the buyer expected or thought fair. A purchase can feel bad even when the item is worth its price, if the price looks like a "bad deal" against the reference point. The Nobel popular summary gives the reference point as, for example, "the lowest price we find when searching on the internet." Clothing is easy to price-compare, often side by side on the same page. A restaurant meal is harder to compare directly, so there is less of a gap between what was paid and a visible alternative.

Relative, not absolute, differences. In a scenario from Tversky and Kahneman's 1981 paper in Science, quoted in the Nobel background, most people say they would drive 20 minutes to save $5 on a $15 calculator but not to save $5 on a $125 calculator, even though the saving is the same. The Nobel popular summary gives the same pattern with a watch that is 100 krona cheaper in another shop. Applied to everyday spending, a $20 premium is 50 percent of a $40 shirt but 20 percent of a $100 dinner, and relative thinking makes the shirt premium loom larger.

Base price	$20 as a share of the price
$40 shirt	50.0%
$60 shirt	33.3%
$100 dinner	20.0%
$125 dinner	16.0%

Experiences and possessions

A separate line of research compares experiences with things. Leaf Van Boven and Thomas Gilovich, in "To Do or to Have? That Is the Question" (Journal of Personality and Social Psychology, 2003), found in two surveys that respondents from various demographic groups said experiential purchases, those made mainly to acquire a life experience, made them happier than material purchases. In a laboratory experiment, participants felt more positive after thinking about an experiential purchase than a material one. The authors' explanation is that experiences are "more open to positive reinterpretations, are a more meaningful part of one's identity, and contribute more to successful social relationships."

That fits the dinner side of the comparison. A shared meal is an experience with a social component, and on the authors' account experiences are more open to positive reinterpretation afterwards. A shirt is a possession, and it stays available for comparison with the next shirt. The research measures happiness with purchases, not regret about prices, so it supports the pattern without proving the popular explanation for it.

Where the popular story overreaches

A common claim is that a durable item causes lasting guilt because it stays in view and keeps reminding its owner of the price. The mental accounting literature points partly the other way. The Nobel background describes Prelec and Loewenstein's 1998 "prospective accounting" model, in which the pain of paying and the pleasure of consuming interact, and "coupling" is the degree to which consumption brings payment to mind. In Shafir and Thaler's study of wine collectors, as summarised by the Academy, advance purchases were thought of as investments, and a bottle bought earlier and opened as planned was often treated as "free," or even as saving money. Separating the payment from the later use reduced the pain of buying.

So an item used for months or years may call its price to mind less, not more, once the purchase is in the past. Which effect dominates for a given person and purchase is not something the research cited here settles. The honest summary is that separate budgets, reference prices and relative thinking explain why the decision to buy the shirt can feel harder. They do not show that owning it causes lasting guilt.

Cost per use, and what it hides

The usual practical advice is to divide the price of a durable item by the number of times it will be used. The arithmetic is simple, and it does put a coat and a restaurant meal on a similar footing. The table below shows how sensitive the answer is to the usage guess, which is the one number nobody knows at the time of purchase.

Item	Price	Low use	Middle use	High use
Winter coat	$220	30 wears: $7.33	90 wears: $2.44	300 wears: $0.73
Shirt	$40	10 wears: $4.00	30 wears: $1.33	60 wears: $0.67
Office chair	$450	250 days: $1.80	500 days: $0.90	1,000 days: $0.45

Price divided by uses. Usage counts are illustrative assumptions, not measurements.

The same $220 coat costs anywhere from 73 cents to more than $7 per wear, depending on an estimate made in a shop. The more useful form of the calculation is a break-even count. If a $40 shirt is expected to be worn 30 times, it costs $1.33 per wear, and a $120 shirt has to be worn 90 times to match it. If an $80 coat would be worn 60 times, a $220 coat has to be worn 165 times. Framed this way, the question becomes whether the more expensive item will really be used that much more, which is a question about behavior rather than about price.

Cost per use also leaves out things that matter: money spent now is not available to save or invest, a durable item may be replaced for reasons other than wear, and comparing a coat's cost per day with a meal's cost per evening mixes very different kinds of value. It is best treated as a way to check one assumption, not as proof that a purchase was a good one.

The same break-even logic applies to larger decisions. Our rent increase vs. moving cost calculator works out how many months a lower monthly housing cost takes to repay the one-time cost of moving.

Frequently asked questions

What is mental accounting?

It is Richard Thaler's theory of how people organise money decisions by grouping spending into separate categories, each with its own budget and reference point, instead of treating money as fully interchangeable. The 2017 Nobel background describes it as a psychological theory of how limited cognition affects spending, saving and other household behavior.

Is there real-world evidence that people do not treat money as fungible?

Yes. Hastings and Shapiro studied gasoline purchases and found that households shifted between fuel grades in response to gas price changes far more than income effects could explain, and they consistently rejected the hypothesis that households treat gas money as fungible with other income. The Nobel background reports a shift toward premium fuel 14 times larger than a standard model predicted when prices fell by about half.

Why might a $20 difference feel bigger on a shirt than on a meal?

Three findings fit: clothing and eating out are often separate mental accounts with different budgets; clothing is easy to compare against a visible reference price; and people judge savings relative to the price, so $20 is 50 percent of a $40 shirt but 20 percent of a $100 dinner. The research explains why the decision can feel harder, not that the purchase causes lasting regret.

Do experiences make people happier than possessions?

Van Boven and Gilovich's 2003 study found that respondents across demographic groups said experiential purchases made them happier than material ones, and that thinking about an experiential purchase produced more positive feelings in the lab. The authors attribute this to experiences being more open to positive reinterpretation, more tied to identity and more social.

How reliable is the cost-per-wear calculation?

Only as reliable as the usage estimate. A $220 coat is 73 cents per wear at 300 wears and $7.33 at 30. A more useful version is the break-even count: a $120 shirt must be worn 90 times to match a $40 shirt worn 30 times.

Sources

• Royal Swedish Academy of Sciences, Richard H. Thaler: Integrating Economics with Psychology, scientific background to the 2017 prize
• NobelPrize.org, popular information on the 2017 Prize in Economic Sciences
• Hastings and Shapiro, Mental Accounting and Consumer Choice: Evidence from Commodity Price Shocks, NBER Working Paper 18248
• Van Boven and Gilovich, To Do or to Have? That Is the Question, Journal of Personality and Social Psychology 85(6), 2003
• Tversky and Kahneman, The Framing of Decisions and the Psychology of Choice, Science 211, 1981
• Shafir and Thaler, Invest Now, Drink Later, Spend Never: The Mental Accounting of Delayed Consumption