Endowment Effect

The endowment effect is the tendency for a person to value something more highly once it is theirs. Richard Thaler introduced the term in "Toward a Positive Theory of Consumer Choice," published in the Journal of Economic Behavior and Organization in 1980, and the sentence that names it is more specific than the way the term is usually reported: "Henceforth, I will refer to the underweighting of opportunity costs as the endowment effect." The two descriptions are one idea. If money paid out is felt as a loss and money forgone is felt as a smaller thing, a gain not taken, then anything already in hand is priced above the same thing not yet in hand. Thaler drew the valuation consequence in the same paragraph, noting that, all else equal, goods included in a person's endowment "will be more highly valued than those not held in the endowment."

The three cases Thaler opened with are all price gaps. A man who bought wine at about $5 a bottle in the late 1950s refuses $100 a bottle from his merchant, though he has never paid more than $35 for a bottle. A man mows his own lawn rather than pay the neighbor's son the $8 he would charge, and would not mow the neighbor's identical lawn for $20. Asked what they would pay to cure a one-in-a-thousand risk of a fatal disease, and what they would accept to take on the same risk, many people give answers that differ by a factor of ten or more. Nothing about the object changes between the two questions. What changes is which side of it the person is standing on.

The opportunity-cost formulation is the sharper of the two, because it also covers cases where nothing is owned yet. Thaler's own example is a price difference presented as a cash discount rather than as a credit-card surcharge: his 1980 paper records that when legislation on the point was before Congress, the card lobby turned its attention to which of those two words would be used. The money at stake is identical either way. One version asks the customer to give up a gain and the other asks them to absorb a loss.

The evidence hardened in a set of market experiments run by Daniel Kahneman, Jack Knetsch and Thaler in 1990. Subjects traded induced-value tokens for several rounds, with prices and trade volumes announced each time, and showed no endowment effect at all, which is what economic theory predicts. Half were then given coffee mugs to sell to the other half, and the mug and pen markets produced sizeable gaps that showed no tendency to shrink with more rounds. For both goods the median price at which owners would sell was about twice the median price at which buyers would buy. A later review assembled 337 estimates of that ratio from 76 studies and put the geometric mean at 3.28, while finding that it varies systematically by the kind of good: highest for public and non-market goods, and lowest for goods with a well-known monetary value.

The contested side is worth knowing, because it is unusually specific. Charles Plott and Kathryn Zeiler ran a protocol that gave subjects substantial practice at buying and selling before their prices were elicited, and found no significant gap for mugs, which suggests part of what earlier studies measured was unfamiliarity with the task. John List found the effect among amateur collectors recruited at a sports-card market and no effect among the professional dealers working the same floor. Others attribute the gap to ownership itself rather than to loss aversion. The Royal Swedish Academy of Sciences, reviewing this literature when Thaler received the 2017 prize in economic sciences, treats loss aversion as still the leading explanation while recording those alternatives. So the direction of the effect holds up, while its size depends heavily on the good, on the procedure, and on the experience of the person holding it.

In household finance the effect concentrates in a few places, and the well-known-value finding says which. A listed fund or a share of stock has a posted price, which is the condition under which measured gaps are smallest, so an unwillingness to sell an inherited holding is more often about what the holding means than about what it is worth. The larger exposure is anything without a quoted price: a house priced at the number the seller needs rather than at what comparable homes fetched, a small business, a collection, a second property that has been in the family. In each of those the owner's price and the market's price can sit far apart for years, and the cost is measured in the sale that does not happen.

The mechanism runs in three steps. Ownership sets the reference point. Parting with the object is then scored as a loss, while acquiring it would have been scored as a gain. Because those two are not weighted equally, the selling price comes out above the buying price for the same thing, and the person holding it experiences that gap as information about the object rather than about their own position.

A hypothetical, and the arithmetic is the point. Ivan inherits 300 shares of a fund trading at $48, so the position is worth $14,400. Asked what he would sell at, he says nothing under $60 a share, which prices the same position at $18,000. Asked separately whether he would put $14,400 of cash into that fund today, he says no. His two answers price one identical holding at $18,000 and at less than $14,400, a spread of at least $3,600 created entirely by which side of the trade he is on. The test that collapses the gap is the second question: would this be bought today, at today's price, with today's money?

Pros

• Reluctance to trade is not always an error. It damps down churn, and the trading costs and taxes that come with it.
• The effect is systematic rather than random, so it can be checked with a single question rather than merely regretted later.
• Attachment to a long-held asset sometimes preserves a position that would otherwise have been sold in a bad week for no good reason.

Cons

• It prices a holding by the owner's history rather than by what the holding will do next, and history is not information about the future.
• It sets asking prices above market on exactly the assets that are hardest to value, which is where the resulting delay costs the most.
• It makes an honest comparison difficult, because the sell question and the buy question about one object get different answers from the same person.
• It compounds with inertia, so a concentrated or inherited position can sit unexamined for years while the reason for holding it goes unstated.