Sunk Cost Fallacy

The sunk cost fallacy is when people keep going because they already spent time, money, or effort. They do not quit even when stopping would be better.

The sunk cost fallacy is a decision bias where prior irrecoverable investments unduly influence ongoing choices, leading to suboptimal commitment. It manifests as continued resource allocation despite marginal returns that fail to justify further expenditure.

A person buys a nonrefundable concert ticket for $80 but feels sick on the day of the show. Instead of staying home to rest, they drag themselves out because they "already paid." The $80 is gone either way, but the sunk cost makes them choose an option that leaves them feeling worse.

A pharmaceutical firm has invested $400 million in a late-stage drug candidate. Phase III trials return a hazard ratio indicating marginal efficacy and a rising adverse-event signal. A forward-looking net present value model—discounting projected revenues against remaining R&D, regulatory, and launch costs—yields a negative expected value. Nevertheless, the portfolio committee approves continued development because prior expenditures dominate the deliberation frame. The temporal accounting module treats the $400 million as a salient state variable, suppressing revaluation and blocking resource reallocation to higher-NPV pipeline assets. The resulting escalation of commitment persists until regulatory rejection forces termination, by which point an additional $120 million has been committed—capital that subsequent analysis attributes directly to sunk cost anchoring rather than any updated clinical signal.

Because someone already paid, they feel they must keep going. That feeling makes them ignore new facts showing it is worse to continue.

A temporal_accounting_systems structural element encodes prior expenditures as salient state variables, creating asymmetric weighting toward continuity when evaluating options. This weighting constrains revaluation processes and biases choices toward preserving past commitments.

Pause and ask if future gains beat future costs without thinking about past losses. Make choices based only on what will happen next.

Reframe decisions with forward-looking cost–benefit analysis that explicitly excludes sunk expenditures from utility calculations. Implement decision checkpoints and precommitment rules to override anchoring from prior investments.

Overcommitment to failing projects; Delay in reallocating resources; Escalation of losses

Adversarial actors can deliberately front-load investments — financial, emotional, or reputational — into a target party to create irrecoverable commitment anchors, then extract ongoing concessions by leveraging the target's reluctance to "waste" what has already been spent. In negotiation or procurement contexts, a counterparty can manufacture escalating sunk costs through staged contracting to lock in continued engagement even as terms deteriorate. Political or organizational manipulators can publicly associate a leader's identity with a failing initiative, making withdrawal politically costly and forcing continued resource allocation to an objectively losing course of action.

Institutionalize forward-only cost–benefit decision checkpoints that explicitly strip prior expenditures from option valuations, using structured templates that require decision-makers to list only prospective costs and returns. Pre-commit to kill criteria before projects begin — defined thresholds at which continuation is automatically halted regardless of prior investment — so the anchor of past spending cannot override evaluation at review points. Train decision teams to externalize sunk cost identification as a standing agenda item during project reviews, transforming an implicit cognitive bias into an explicit, discussable variable.