Sunk Cost Economics: When Past Investment Predicts Future Waste

The sunk cost fallacy is not a bug in human decision-making—it is a feature of how we construct narratives about commitment and identity.

We know the logic: money spent is gone. Future decisions should depend only on future costs and benefits. Yet organisations routinely pour resources into failing projects because they have already invested heavily. A pharmaceutical company continues a drug trial that shows no efficacy. A retailer expands a store format that underperforms. A technology firm maintains a platform that bleeds users. The past investment whispers louder than the future forecast.

Behavioural economists have documented this pattern exhaustively. Arkes and Blumer's classic study showed that people who paid more for a theatre ticket were more likely to attend a show they no longer wanted to see. The ticket price was sunk—irrelevant to whether the evening would be enjoyable. Yet it shaped behaviour anyway. The investment became a justification for continued commitment, even when that commitment made no rational sense.

What gets less attention is why this happens so reliably, and what it reveals about how organisations actually function.

The sunk cost fallacy persists because it solves a real problem: how to signal resolve. When a leader commits substantial resources to a direction, that commitment becomes evidence of conviction. Abandoning the project early signals weakness, poor judgment, or fickleness. The cost already spent becomes a form of credibility. Walking away means admitting the investment was wasted—and admitting you made a poor decision. The sunk cost is not really about the money. It is about reputation and identity.

This is why the fallacy is so resistant to correction. You cannot simply tell a decision-maker that past costs are irrelevant. You are asking them to absorb a reputational loss that feels very real, even if the economics are sound. The rational choice—cut losses and redeploy resources—reads as failure.

Organisations amplify this dynamic through structure. Once a project has a budget line, a team, and a quarterly review cycle, it develops institutional inertia. The people managing it have careers invested in its success. Stakeholders have made public commitments. Cancellation requires not just acknowledging a bad decision, but creating visible losers—people whose projects are terminated, whose roles are eliminated. The sunk cost becomes embedded in the organisation's politics.

The consequence is predictable: capital gets trapped in low-return activities. A business unit that should be wound down instead limps forward, consuming resources that could generate better returns elsewhere. The cost of this misallocation compounds over time. A company that tolerates sunk cost reasoning in one domain will tolerate it in others. The practice becomes normalised.

There is a second, subtler cost. When organisations continue investing in failing projects to justify past spending, they signal something about how decisions are actually made. Employees learn that rational analysis matters less than narrative consistency. They learn that admitting error is costly. They learn to defend positions rather than update them. This shapes how information flows, how risks are assessed, and how new opportunities are evaluated.

The antidote is not willpower or better spreadsheets. It is structural. Some organisations build explicit review gates where projects are evaluated as if they were new decisions, with past investment explicitly excluded from the analysis. Others create psychological safety around project termination—treating it as learning rather than failure. A few separate the people who initiated a project from those who evaluate its continuation, removing the reputational stake.

But the deepest fix is cultural. It requires treating the ability to change course as a sign of strength, not weakness. It requires rewarding the leader who says "we were wrong, and here is what we are doing instead" rather than the one who doubles down. It requires making the cost of sunk cost reasoning visible—showing what else could have been built with those resources.

Until then, past investment will continue to predict future waste. Not because people are irrational, but because the incentives are perfectly rational. The sunk cost fallacy is not a cognitive error. It is a rational response to an environment where admitting error is expensive.