The Sunk Cost Fallacy: Why You Finish Books You Hate

You are two hundred pages into a novel you actively resent. You are not curious about the ending. You are not learning anything. You keep going anyway, because stopping now would mean those two hundred pages were wasted. So you spend another four hours to avoid wasting the previous six.

Said out loud it is obviously wrong, and almost everyone does it. This is the sunk cost fallacy, and books are only its most harmless venue.

What a sunk cost is

A sunk cost is any investment you have already made that you cannot recover: money spent, hours worked, effort given, reputation attached. The defining feature is that it does not change based on what you do next. Whether you finish the book or close it, those six hours are equally spent.

Which means, coldly, that a sunk cost carries no information about what you should do now. The only relevant questions are forward looking: what will this cost me from here, what will it give me from here, and what else could I do with the same resources? The fallacy is letting the unrecoverable past sit in the calculation anyway.

The classic demonstration

The tidiest evidence comes from Hal Arkes and Catherine Blumer in 1985. In one study they sold theatre season tickets at three prices: full price, a small discount and a large discount, assigned at random. Everyone got the same seats to the same plays. The people who had paid full price attended noticeably more performances in the first half of the season.

Nothing about the plays differed. Only the amount already spent, which was gone for all three groups, differed. The larger payment created a felt obligation to extract value from it.

The same paper used a scenario people still quote: you have paid for a hundred dollar ski trip and then a much better fifty dollar trip on the same weekend, and both tickets are non refundable. Many people choose the trip they expect to enjoy less, because the more expensive ticket would otherwise feel wasted. You have already lost the money either way. The only remaining choice is which weekend you get.

Arkes and Blumer argued the driver is not stinginess but a desire not to appear wasteful, including to yourself. That is why the fallacy is so resistant to being explained: it is powered by a value most of us hold.

Where it shows up in ordinary life

• Books and shows. Finishing something you dislike so the hours count for something, when the hours are exactly what you cannot get back.
• Courses and degrees. Two years into a path that no longer suits you, the two years become the argument for a third, when the honest question is whether the remaining time buys anything you want.
• Apps and subscriptions. An annual plan you barely use gets opened out of guilt rather than interest, and the guilt sessions teach nothing.
• Queues and waiting. Twenty minutes on hold makes the twenty first minute feel compulsory, even after you have concluded the call will not resolve anything.
• Relationships and projects. The oldest version of this trap, and the one where the language of investment does the most damage to clear thinking.

Organisations have a bigger version of the same problem, usually called escalation of commitment: a project acquires defenders whose reputations are tied to it, and each further spend makes abandonment more embarrassing.

How it differs from healthy perseverance

Not every long slog is a fallacy, and the distinction matters because this idea is easy to misuse as permission to quit anything mildly hard.

Perseverance is forward looking: the reward is still there, you still want it, and the difficulty is the ordinary cost of getting it. A learning plateau is the honest example. Progress goes flat while consolidation happens underneath, and quitting there is usually a mistake, which is why it is worth understanding why learning plateaus happen before treating one as a verdict.

Sunk cost thinking is backward looking: the reward has faded, changed or turned out to be something you never wanted, and the only remaining reason to continue is the amount already spent. One useful test is the stranger's question. If someone handed you this project today, at this exact stage, with nothing invested, would you take it on? If the answer is no, the investment is doing all the work.

Why the pull is so strong

Several things reinforce each other. Losses hurt more than equivalent gains please, so writing off a cost feels like taking a fresh loss rather than acknowledging an old one. We prefer to look consistent, and abandonment reads as inconsistency. And unfinished things nag: the mind keeps an open loop on interrupted work, which is the Zeigarnik effect, and closing the loop by finishing feels better than closing it by deciding.

There is also a plain resource problem. Deciding to quit takes deliberate thought, and deliberate thought is exactly what runs low late in the day, which is when most of these choices get made. That is one practical face of decision fatigue: continuing is the default, and defaults win when you are tired.

How to quit well

1. Name the cost as gone

Say it explicitly: I spent nine hours and I am not getting them back. Grieve the amount rather than the decision. People find quitting far easier once the loss has been acknowledged out loud, because the loss stops being something quitting would cause and becomes something that already happened.

2. Set decision points in advance

Before you start, write the condition under which you will stop. A hundred pages. Three chapters. Two weeks of trying. Pre commitment moves the choice to a moment when you are calm and have nothing invested, which is the only moment you can judge it cleanly. This is the same machinery as any other commitment device, pointed at exits rather than starts.

3. Ask what else the time could buy

The real cost of finishing a bad book is not the book. It is the good one you did not read. Comparing an option against your best alternative, rather than against zero, reframes persistence as the expensive choice it often is.

4. Count the information as your return

An abandoned project is rarely a total loss. You learned that this genre is not for you, that this field is not the one, that this method does not suit your week. Treating that knowledge as the return converts a write off into a purchase, which is both more accurate and much easier to accept.

The two minutes you get back

The reason a psychology writer keeps returning to this is that quitting well is one of the few habits that hands you time rather than asking for it. Closing a book you resent does not just end an annoyance. It frees the slot.

What goes in the slot is up to you. We built MindSnap, which is our app, for exactly that kind of gap: a couple of minutes of a real story rather than a chapter of obligation, with five flagship story collections plus unlimited topics you can request, and you can browse topics to see the range. The general point stands whatever you use. A small deliberate choice beats a large reluctant one, and the hours behind you have no vote.