Owning something raises the price you would accept for it

A gap that appears as soon as ownership does

Ask someone what they would pay for an object and ask another person what they would accept to give up the same object, and the second figure is typically higher. The gap opens as soon as possession does, and it’s generally called the endowment effect. It is one of the more reliably reproduced observations in the study of how people value things.

The usual explanation connects it to loss aversion. Giving up something you hold registers as a loss, and losses weigh more heavily than equivalent gains, so the compensation demanded exceeds what would have been paid to acquire it. Other explanations exist, including simple habit and the effort of imagining life without the thing, and the debate about mechanism is not settled.

Whatever the mechanism, the practical consequence is consistent: the price at which you would sell is higher than the price at which you would buy, for the very same object.

Where it costs a household money

The clearest case is selling. Items are listed at prices reflecting what they are worth to the owner rather than to the market, fail to sell, and depreciate further while waiting. The eventual sale price is lower than the offer that was refused a year earlier, and the difference is the cost of the gap.

It applies to arrangements as well as objects. An account, a supplier, a plan — anything held tends to be valued above the alternative, which is part of why switching happens less often than the numbers would suggest. The incumbent gets a premium simply for being the incumbent, which is worth remembering when a switch looks marginal on the household’s own numbers.

And it interacts with storage. Keeping things has costs — space, moving them, insuring them, maintaining them, finding what you need among them — which are diffuse and rarely counted, while the value of the object is specific and easy to picture.

Trials and possession are the same mechanism used deliberately

The commercial applications are straightforward. Home trials, test drives, approval periods and generous returns all create possession before payment is final. Once the item is in the house, the decision is no longer whether to acquire it but whether to give it up, and those two questions produce different answers.

The same logic applies to free periods on services. Cancelling at the end of a trial is a decision to relinquish something currently held, which is a harder decision than declining to start it would have been.

None of this is deception; the item can be returned and the service cancelled. It is simply an accurate use of a known feature of how people value what they hold, and knowing about it is most of the defence.

A question that partly dissolves the gap

The most effective reframing is to ask whether you would buy the item today, at its current market price, if you did not already own it. That converts a decision about giving something up into a decision about acquiring it, which is the frame that produces the lower and more market-consistent number.

It is a close relative of the question that dissolves sunk cost, and for a similar reason: both work by removing history from a decision that should only depend on what happens next.

It is not a complete answer. Some things are genuinely worth more to their owner than to anyone else — anything with a history attached, anything fitted to a particular purpose, anything whose replacement would involve substantial hassle — and those premiums are real rather than cognitive errors.

Where the effect is not an error at all

Attachment is not always a distortion to be corrected. Objects carry associations, and a household that ruthlessly liquidated everything to which it had a sentimental premium would be poorer in a way no balance sheet records. The point of noticing the effect is not to eliminate it.

There is also a transaction-cost defence of holding on. Selling takes time, involves strangers, and carries risk, and an owner who declines a modest offer because the process is not worth the difference is making a sound judgement rather than exhibiting a bias.

The bias is worth attending to specifically where the gap is large, the item is genuinely fungible, and the delay is costing money — which is a much narrower set of situations than the effect itself covers.