The Sunk Cost Fallacy in Everyday Life

What the Sunk Cost Fallacy Is

The sunk cost fallacy is the tendency to continue a course of action because of past investment — money, time, or effort — even when continuing is no longer the rational choice available to you from today forward. The reasoning goes: "I've already put £X into this, so I should keep going to get my money's worth." The economic error sitting underneath that reasoning is simple to state and surprisingly hard to internalise: money, time, and effort already spent cannot be recovered by any decision you make next. It is sunk — gone regardless of what happens from here. The only question that should determine your next move is whether continuing is the best use of your resources starting now, with no reference at all to what has already gone.

This is one of the best-documented biases in behavioural economics, studied since the late 1970s by researchers including Richard Thaler and Daniel Kahneman, and it shows up with almost identical structure across completely unrelated domains — gym memberships, failing relationships, government infrastructure projects, stock portfolios, half-finished DIY jobs, and multi-billion-pound corporate ventures all fall prey to the same error, for the same underlying psychological reasons.

Understanding why the fallacy happens — not just that it happens — is what makes it possible to catch yourself in the act, because the pull of sunk cost reasoning is strong enough that simply knowing the rule ("ignore sunk costs") is rarely sufficient on its own. It has to be paired with specific, practical decision tools, which this chapter builds toward.

Why Our Brains Do This

Two psychological mechanisms do most of the work in producing sunk cost reasoning. The first is loss aversion: extensive research, beginning with Kahneman and Tversky's prospect theory, shows that people feel the pain of a loss roughly twice as intensely as the pleasure of an equivalent gain. Walking away from a past investment means accepting, explicitly, that the money or time is gone — crystallising a loss that, while you keep going, can be comfortably left abstract and unrealised. Continuing feels like postponing the loss, even when continuing is actually making the eventual loss larger.

The second mechanism is commitment consistency — a well-documented drive to behave in ways that stay consistent with decisions you have already made and publicly or privately committed to. Abandoning a course of action can feel like admitting the original decision was wrong, which triggers a specific kind of discomfort psychologists call cognitive dissonance. Continuing, even against the evidence, temporarily resolves that discomfort by preserving the story that the original decision was sound — "I haven't wasted the money, I just haven't finished benefiting from it yet."

A third, quieter factor is the social dimension: sunk cost decisions are often visible to other people — a partner who knows what the gym membership cost, a boss who approved the original project budget, friends who know about the losing investment. Reversing course in front of an audience adds a layer of perceived judgment on top of the financial loss itself, which pushes further toward continuing even when the private, unobserved version of the same decision would more easily go the other way.

Financial Examples

The gym membership. You paid £600 for a year-long gym membership in January. By March, you have not been in six weeks and realistically know you are unlikely to go regularly for the rest of the year. The sunk cost thinking says: "I've already paid, so I should force myself to use it." The rational thinking recognises that the £600 is gone whether you attend from here or not — the only live question is whether forcing attendance from today onward is actually the best use of your time and motivation, independent of the money already spent.

The losing investment. You bought shares at £10 each. They now trade at £3. The instinct to hold "until it gets back to at least what I paid" ties your decision to an entirely arbitrary reference point — your personal purchase price — that the market has no knowledge of and no reason to respect. The only useful question is: "Is this £3-a-share holding the best place for this capital right now, compared with everything else I could do with it?" If the answer is no, the rational move is to sell, regardless of the number you originally paid. This specific pattern has a name in behavioural finance — the disposition effect — the well-documented tendency for investors to hold losing positions too long (hoping to "get back to even") while selling winning positions too early (to lock in the good feeling of a gain).

The subscription. You pay £15 a month for a streaming service you have not used in three months, because you originally signed up for a deal that felt too good to pass up. The sunk cost — the value of that original deal — is already gone, captured or not. The ongoing £15 a month is a completely separate, live, future cost being paid for a benefit you are not receiving. Cancelling costs you nothing that the original deal already delivered.

The half-finished renovation. A kitchen refit goes over budget by £4,000 partway through. Finishing it "properly" would cost another £3,000 the budget never allowed for. The sunk cost pull says the £4,000 already spent demands finishing at full spec to "justify" it. The rational question is simpler: given where the money sits today, is spending a further £3,000 the best use of that specific £3,000, compared with a cheaper finish, a pause, or redirecting the money elsewhere? The £4,000 already gone has no bearing on that answer either way.

The non-refundable holiday. You have paid £900, non-refundably, for a trip. A family emergency now means going would be stressful and poorly timed. Many people go anyway "because it's paid for" — but the £900 is spent regardless of whether you travel. The only real question left is whether travelling, given current circumstances, is the better of the two remaining options: going, or not going. The £900 does not belong in that comparison at all; it is identical under both outcomes.

How It Distorts Investing Specifically

Investing is one of the areas where the sunk cost fallacy does the most lasting financial damage, because the stakes compound over time and the temptation to "wait for it to recover" can persist for years rather than resolving within a single decision. A few specific patterns are worth naming directly.

Averaging down to "rescue" a position. Buying more of a falling stock specifically to lower your average purchase price, with the primary motivation being to make the original loss easier to recover from, is sunk cost reasoning wearing a strategy's clothing. Averaging down can occasionally be a sound decision — but only when it is justified by the investment's current prospects, not by a wish to improve the optics of a past purchase price.

Refusing to sell at a loss inside an ISA or pension for "tax reasons" that do not actually apply. Some investors conflate the (legitimate) tax treatment of gains and losses outside tax-advantaged accounts with a general reluctance to "realise" a loss anywhere, including inside an ISA or SIPP where crystallising a loss carries no additional tax consequence at all. The reluctance in these cases is pure loss aversion, dressed up as a tax consideration that does not actually exist for that account type.

Treating a fund or platform switch as a sunk cost. Some investors stay with an expensive platform or an underperforming fund because of the effort already invested in opening the account or researching the original choice. The effort is sunk. The ongoing fee differential and performance gap are live, future costs, and should be judged purely on whether switching today is worth the switching effort, not on how much research went into the original decision.

Beyond Money: Careers, Relationships, and Projects

The same mechanism drives decisions well outside personal finance, and recognising the pattern in these areas often makes it easier to spot in money decisions too, because the emotional structure is identical.

Careers. Staying in a career path, or a specific job, primarily because of years of training, a degree, or professional exams already completed is sunk cost reasoning applied to time rather than money. The training is done and cannot be undone either way. The only live question is whether this is the best use of your remaining working years, starting from today, regardless of the path that got you to this point.

Relationships. "We've been together six years, I can't just leave now" ties a present decision to past time invested, rather than to the relationship's current and likely future quality. This is one of the more painful applications of sunk cost reasoning precisely because the emotional stakes are so much higher than a gym membership — but the underlying logical error is the same one.

Government and corporate projects. The most famous public example is the Anglo-French Concorde programme, which continued receiving funding for years after cost overruns made its commercial failure effectively certain — a pattern so well documented in the economics literature that "the Concorde fallacy" is sometimes used as another name for sunk cost reasoning at scale. Large organisations are, if anything, more exposed to this than individuals, because reversing a publicly announced, heavily invested project carries political and reputational costs on top of the financial ones, which pushes toward continuation even in the face of overwhelming evidence that the project should stop.

Escalation of Commitment — The More Dangerous Cousin

Escalation of commitment is what happens when sunk cost reasoning does not just preserve a bad decision but actively drives more resources into it, in an attempt to justify or recover what has already been lost. It is the mechanism behind "chasing losses" in gambling — betting progressively larger amounts specifically to recoup previous losses, rather than because the new bet is independently a good one — and it appears in milder form in everyday financial life whenever a losing position is topped up primarily to make the loss feel less final.

The defence against escalation is the same decision rule as for ordinary sunk cost reasoning, applied more strictly: before adding any further resource to something already underwater, ask explicitly whether you would make that specific new commitment today, starting completely fresh, with no history attached. If the honest answer is no, the additional commitment is being driven by the past loss, not by the merits of the decision in front of you.

How to Make Better Decisions

The simplest and most reliable tool is to mentally remove the past investment from the decision entirely and ask: "If I had not already committed anything to this, would I start it today, from scratch, knowing what I know now?" If the honest answer is no, the sunk cost fallacy is very likely what is keeping you in it rather than any genuine ongoing merit.

For investment holdings specifically, a related and very practical habit is to zero-base your portfolio periodically — say, once or twice a year. For each holding, ask: "If I did not already own this, and had this same amount of cash sitting uninvested today, would I buy this specific asset at this specific price?" If the answer is no, that is a meaningful signal worth acting on, entirely independent of what you originally paid for it.

It also helps to pre-commit to decision rules before emotion is involved, precisely because sunk cost reasoning is strongest in the moment and weakest in the cold light of a rule set in advance. A stop-loss level on a trade, a maximum number of months you will "give" an underperforming side project before closing it, or a specific review date for a membership or subscription all work the same way: they move the hard decision to a moment before the emotional pull of sunk cost has a chance to operate.

None of this is an argument for recklessness or constantly abandoning things at the first sign of difficulty. Sometimes the right answer, even after properly removing the sunk cost from the reasoning, is still to continue — a holding may genuinely be a good buy at today's price; a career path may genuinely be the right one from here. The point is not to always walk away. The point is to make sure the decision is actually being driven by future value, not by an unrecoverable number in the past.

Two Worked Scenarios From Everyday Life

Scenario one — the car that keeps needing repairs. A car has already had £1,800 spent on repairs over the past year, and the gearbox has now failed, with a quote of a further £1,100 to fix. The instinct many people have is to think "I've already put £1,800 into this car, I can't just scrap it now" — treating the next repair as a continuation of a project already heavily invested in. The rational question ignores the £1,800 entirely: is spending £1,100 on this specific car, today, the best use of that money compared with putting it toward a different car, a short-term rental, or public transport while saving toward a replacement? If a mechanic would value the car, post-repair, at less than the cost of two further repairs like this one, the sunk £1,800 is irrelevant to that comparison — only the next £1,100 and what it buys actually matters.

Scenario two — the course you paid for but are not enjoying. Someone pays £450 upfront for a ten-week evening course, attends three sessions, and realises it is not useful or enjoyable, with seven weeks of sunk time and money ahead of them either way. The sunk cost pull says finish it to "get your money's worth." But the £450 is identical whether the remaining seven sessions are attended or not — it has already been spent and cannot be clawed back by sitting through more evenings that do not serve the original goal. The only live question is whether those seven remaining evenings are better spent finishing a course that is not delivering value, or doing something else entirely with that time. Framed this way, the decision usually becomes much clearer, and often different from the one sunk cost reasoning would have produced.

Practical Pre-Commitment Tools

Because sunk cost reasoning is strongest exactly when you are most emotionally invested, the most effective tools are ones you set up in advance, when you are not yet attached to a specific outcome.

1. Set a review date, not just a review trigger. "I'll reconsider the gym membership if I haven't been in a month" is weaker than "I will review this on the 1st of every month regardless." A fixed date removes the temptation to keep pushing the review back.
2. Write the exit criteria down before you start. Before beginning a side project, a renovation, or a new investment position, write down in advance the specific conditions under which you would stop — a cost ceiling, a time limit, a performance threshold. Deciding this before any money is sunk removes the bias that affects the decision once some has been.
3. Use a neutral second opinion for big decisions. Someone with no history in the specific decision — a friend, a partner, a financial adviser — is not subject to your personal sunk cost pull and can often see the "would I start this today?" question far more clearly than you can from inside it.
4. Separate the decision from the emotion with a short delay. Where possible, avoid making the continue-or-stop decision in the same moment you are confronting the loss. A short pause — a day, a week — allows the acute loss-aversion response to settle slightly before the actual decision is made.
5. Reframe the language you use. Replace "I'd be wasting the money I already spent" with "the money is already spent either way — what's the best next move?" The first framing keeps the sunk cost active in your reasoning; the second removes it entirely. This sounds trivial, but the words you use to describe a decision measurably shape which factors your brain treats as relevant to it.

None of these tools require you to become a different kind of decision-maker overnight. They work precisely because they take the decision away from the moment of maximum emotional pull and put it somewhere calmer — on a form you filled in before you were attached to the outcome, or in the hands of someone who was never attached to it in the first place.

Common Mistakes

Confusing sunk cost avoidance with perseverance. Perseverance is continuing through difficulty when the expected future outcome is still genuinely positive. Sunk cost continuation is staying when the future outcome is negative, purely because of what has already been invested. Telling the two apart requires an honest, current assessment of future prospects — not a tally of past effort.

Treating "I've come this far" as an argument on its own. Distance travelled is not evidence about distance remaining, or about whether the destination is still worth reaching. It feels like relevant information; it is not.

Letting other people's awareness of the original decision drive the next one. Reversing course in view of people who know what you paid or committed adds social discomfort on top of the financial one, but the people watching are not the ones who have to live with the ongoing cost of continuing.

Applying the fallacy's lesson too bluntly. "Ignore sunk costs" does not mean ignore all context — a position with embedded skills, relationships, or knowledge genuinely built through past investment can be a legitimate reason continuing is still the better option today. The test is always whether that context makes continuing the right call now, not whether stopping would "waste" what has already gone.

Waiting for a "clean" exit point that never arrives. Some people recognise a sunk cost situation clearly but keep waiting for a neater moment to act — a round number, a specific date, a tidier justification to offer others. In practice, the cost of continuing rarely pauses while a cleaner exit point is awaited, and the search for one is often, on close inspection, simply another form of the same avoidance the fallacy produces in the first place.