Opinion · Psychology (Behavioral Finance)
Most people who have ever owned a stock or a coin know the feeling. The position is down, the chart keeps sliding, and every time you open the app you tell yourself the same thing: “I’ll sell once it gets back to what I paid.” The number you paid becomes a kind of finish line, and until the price crosses it, selling feels like admitting defeat.
That feeling has a name. It is the sunk cost fallacy, and it is one of the most studied mistakes in decision-making. It is not really about money, and it is not only about investing. It is about how badly we want the past to have been worth it, and how often that wish quietly makes the next decision for us.
What the sunk cost fallacy actually is
A sunk cost is money, time, or effort that is already spent and cannot be recovered, whatever you do next. The fallacy is letting that unrecoverable cost influence a decision about the future. In theory, a rational decision only looks forward: what will I gain or lose from here? What I paid yesterday is gone either way, so it should carry no weight.
In practice, it carries enormous weight. Selling a losing position is where most people feel it most sharply. The loss already exists on paper, but pressing “sell” turns it into something final. Holding keeps a small hope alive that the loss is not really a loss yet. So we hold, not because we believe the asset has a bright future, but because selling would force us to accept a past mistake.
The key point is subtle. The sunk cost fallacy does not say “always sell” or “always quit.” It says the purchase price should not be part of the decision at all. If you set that number aside and the outlook is still strong, holding can be perfectly reasonable. The mistake is not holding. The mistake is holding because of what you paid.
The classic evidence comes from a 1985 study by psychologists Hal Arkes and Catherine Blumer. They sold theater season subscriptions at randomly assigned prices, some at full price and some at a discount. Over the next six months, the people who had paid full price attended more plays. The shows were identical. The only difference was how much each person had already spent, and that past spending pulled them out of the house.
Why “locking in” a loss feels so bad
Two other ideas help explain why this pull is so strong. The first is loss aversion: losses tend to feel heavier than gains of the same size. The second is what behavioral economists call mental accounting. We keep an imaginary account for each purchase, and we hate closing an account in the red. As long as the position is open, the account is technically still “in progress.”
In investing, this combination has its own name: the disposition effect. Economists Hersh Shefrin and Meir Statman coined the term in 1985 to describe investors’ tendency to sell winners too early and ride losers too long. In 1998, finance professor Terrance Odean tested it on roughly 10,000 brokerage accounts. For most of the year, investors realized gains far more readily than losses. The proportion of gains realized was 14.8%, compared with 9.8% for losses. The telling detail is what happened next. Over the following year, the winners people sold went on to beat the losers they kept by about 3.4 percentage points on average. Holding on to the losers was not a quiet bet on a comeback. On average, it was the worse choice.
There is also an anchor at work here. The price you paid becomes the reference point that every later price is judged against, much like the first number in a negotiation. I covered that mechanism in my piece on the anchoring effect. The sunk cost fallacy is what happens when that anchor is not just a comparison point but a line you refuse to cross.
Three places the same trap shows up
The pattern becomes clearer when you look at cases that have almost nothing in common on the surface.
The theater seat. The Arkes and Blumer subscribers show the pattern at its smallest scale: a personal, low-stakes choice about how to spend an evening. No one’s reputation was on the line. The money was spent either way. The only thing the extra trips to the theater “recovered” was the feeling that the ticket had not been wasted.
The supersonic jet. At the other extreme is the Concorde. The British and French governments kept funding the aircraft for years as its commercial case weakened, partly because so much had already been poured in. The case became famous enough that the sunk cost fallacy is sometimes called the “Concorde fallacy,” a label biologists also borrowed for animals that keep investing in a doomed effort. Here the stakes were national, the decision-makers were experts, and the trap worked anyway.
The long wait. Between those two sits a very modern version: the pre-order that never seems to arrive. When I looked at Tesla’s $50,000 Roadster deposit, the interesting question was why people keep waiting for a product years past its promised date. Part of the answer is that the deposit already paid makes walking away feel like losing it, even long after the original promise has slipped.
Scale, expertise, and stakes all differ across these three. The structure does not. In each case, a cost that could not be recovered was treated as a reason to keep going, and the question “is this still worth it from here?” was replaced by “I can’t let what I’ve already spent go to waste.”
The cost that never shows up on the statement
I have been on the wrong side of this more than once. When a position I owned turned against me, I did exactly what the research describes. I held, told myself I would sell once it recovered, and checked the price far more often than was useful. Looking back, the times I cut the loss and moved on turned out better far more often than the times I waited.
But the reason was not always the price. It was the stress. Watching a losing position day after day is its own kind of expense. It takes your attention, it follows you around, and if it goes on long enough, it starts to wear on your body. My own rough rule became this: if holding on is costing me sleep and health, that is a real price too, and one that can eventually show up in places like a doctor’s bill. No rebound in the chart is guaranteed to cover it.
That is the part the sunk cost framing sometimes misses. We usually talk about it as a purely financial error, measured in dollars not recovered. Yet the decision to keep holding has its own ongoing costs, and they never appear on a brokerage statement. Time, attention, and stress are all spent in the present. Unlike the purchase price, they are not sunk yet. They are exactly the costs a forward-looking decision should count.
A better question to ask
Knowing about a bias does not make you immune to it. The sunk cost pull is strong precisely because it feels like prudence rather than error. “Don’t waste what you’ve already put in” sounds like good advice. The fix, then, is not willpower but a better question, one that removes the past from the frame.
The most useful version is simple: If I didn’t already own this, would I buy it today, at today’s price? The same question works outside investing. If I hadn’t already bought this ticket, would I go out tonight? If we hadn’t already spent three years on this project, would we start it now? If the answer is yes, keep going, and do it with a clear conscience. If the answer is no, what you already spent is not a reason to stay. It is only the reason it hurts to leave.
And when you run the numbers, count the costs that don’t have a ticker symbol too. The past is already paid for. The stress of carrying it forward is not.
Editorial note: This piece grew out of studying a psychology concept rather than following a trending news story. It describes a behavioral pattern, not financial advice; decisions about specific investments depend on your own circumstances.