Fundamentals

Loss Aversion: What It Is and How It Shapes Our Decisions

By Daniel Sardá · Published on October 14, 2026

6 min read1,206 words

Loss aversion describes why a loss can weigh more heavily than a comparable gain. The key is the reference point from which we judge each change.

Imagine buying a mug for ten euros. Minutes later, someone offers you those same ten euros to take it back. Even though nothing material has changed, the offer may no longer seem sufficient. Before, you compared the mug with the money you would have to give up; now, parting with it feels like a loss.

That shift in perspective helps explain loss aversion: a pattern in which losing something relative to a reference point often carries more weight than gaining a comparable amount. It does not mean everyone reacts in the same way, or that keeping something is always irrational. It describes a common asymmetry in how people evaluate change.

A reference point changes what an outcome means

In the simplest economic theory, it may seem enough to compare the final situation: how much money, time, or well-being remains after a decision. Prospect theory, proposed by Daniel Kahneman and Amos Tversky in 1979, introduced a different observation: people often evaluate outcomes as gains or losses relative to a reference point.

That reference point may be the current situation, but it can also be an expectation, an aspiration, or the way an option is presented. As a result, receiving less than expected can be experienced as a loss even when total wealth has increased.

Consider someone who expects a 10 percent pay rise and ultimately receives 4 percent. Their income is higher than before, so they have not suffered a reduction in wealth in the strict sense. Yet against the expectation they adopted as their reference point, the outcome falls on the loss side.

Key idea: A psychological loss is not always a decline in wealth. It can be the gap between what one receives and what one expected to keep or attain.

Framing matters precisely because it can shift that dividing line. Describing an outcome as “keeping 90 percent” or “losing 10 percent” does not alter the final amount, but it can change which aspect becomes central to the evaluation. That does not automatically make the decision a mistake; it shows that preferences take shape within a context.

Loss aversion is not risk aversion

The two concepts are often confused because both arise in economic decisions, but they answer different questions.

Risk aversion describes a preference for a certain outcome over an uncertain one of comparable value. For example, someone may prefer to receive 50 euros for sure rather than enter a gamble with possible outcomes that are both better and worse.

Loss aversion, by contrast, concerns the difference in weight given to losing and gaining relative to a reference point. It can operate even when chance is absent. In 1991, Tversky and Kahneman developed a reference-dependent model of choice that also applies to decisions under certainty.

An ordinary sale illustrates the point. The owner of an object and a prospective buyer know the price and face no lottery. Yet the owner may demand more to give it up than the buyer would pay to acquire it. For the owner, the transaction means losing something already part of the reference situation; for the buyer, it means gaining something. This difference is related to the endowment effect.

Useful distinction: Risk depends on uncertainty among possible outcomes. Loss depends on where each outcome stands relative to a reference point. They can coincide, but they are not the same thing.

The distinction also explains why declining a bet does not by itself show loss aversion. It may reflect caution about uncertainty. And declining to sell something does not demonstrate risk aversion: it may stem from the object’s personal value, transaction costs, or information the buyer does not have.

Three ways to recognize the mechanism

Loss aversion becomes clearer when we first identify what counts as “mine,” “normal,” or “expected.”

Keeping an object

Someone may ask 30 euros to sell a book they would not buy for that price. The book they own is part of their reference point, and giving it up is coded as a loss. This case resembles the endowment effect, but it neither makes the two terms synonymous nor proves that every refusal to sell is a bias.

Falling short of an expectation

A company initially announces a bonus and later grants a smaller one. The employee receives additional income, yet may judge the change as a loss relative to what was announced. The relevant reference point is not necessarily the prior salary, but the expectation that was created.

Choosing between certainty and uncertainty

Now suppose there are two options: keep a certain amount or risk part of it to gain more. Both phenomena may be involved. Uncertainty raises questions about risk; the possibility of ending below what one already considers one’s own introduces a relative loss. Separating the two questions keeps all caution from being attributed to a single cause.

There is no universal ratio

Loss aversion is often summarized by saying that “losses hurt twice as much as gains.” The phrase can be a useful informal reminder of the asymmetry, but it is not a general law. The estimated magnitude varies with the task, method, context, and people involved. A 2024 meta-analysis of decisions under risk found considerable heterogeneity and an aggregate estimate lower than the classic figure.

It is also worth resisting an overly broad conclusion: not all resistance to change arises from loss aversion. Keeping a home, an investment, or a familiar practice may reflect stable preferences, switching costs, lack of information, or a prudent assessment. The concept helps formulate a hypothesis; it does not replace examining the specific reasons.

Nor should it be confused with the sunk-cost effect. This occurs when a past, unrecoverable investment influences the decision to continue. Endowment concerns what one owns; a sunk cost concerns what one has already spent. Both can favor the status quo, but through different mechanisms.

Caution: Identifying a reference point is not enough to call a decision irrational. First consider information, preferences, costs, and the real alternatives.

Individual decisions, exchange, and responsibility

Reference dependence has a modest implication for understanding exchange: two people can value the same good differently without either necessarily being mistaken. Property, expectations, and particular circumstances shape what each person experiences as a gain or a loss.

This is consistent with a view of voluntary exchange grounded in individual valuations. At the same time, the freedom to decide and bear risk entails responsibility: recognizing that framing or attachment to the current situation may influence us does not remove the obligation to weigh consequences.

Loss aversion offers neither a universal recipe nor an automatic political conclusion. Its usefulness is more modest and concrete: it lets us see that a decision depends not only on the final outcome, but also on the point from which it is measured.

Before judging a choice—our own or someone else’s—it is worth asking: what is the reference point that makes this outcome a loss? Sometimes it will be what one already owns; other times, an expectation or a goal. Making that reference point visible does not decide for us, but it clarifies what we are comparing and why one alternative weighs more heavily.