JH
Jonathan Haber
behavioral economics

What is the sunk cost fallacy, and how do you stop letting past investments trap future decisions?

Why past investment traps future decisions — and the practices that escape the trap

Short answer

The sunk cost fallacy is the tendency to continue a losing course because of unrecoverable past investment rather than on the basis of future expected value. It is one of the most robustly documented biases in behavioral economics. The corrective is to evaluate forward-only: what will each path deliver from here, regardless of what has already been spent.

Money spent, time invested, emotion poured in — none of it comes back. Yet the human mind treats irrecoverable past costs as reasons to stay in bad situations: finishing a terrible meal because you paid for it, continuing a failing project because the team has been on it for a year, staying in a relationship that no longer works because of shared history. This is the sunk cost fallacy, and it is remarkably expensive. Understanding the mechanism and training a cleaner decision process can recover significant time, money, and wellbeing.

The practices (7)

Why it works

The sunk cost fallacy works by making past investment emotionally salient at the moment of forward decision. Deliberately zeroing it out in the analysis creates a clean forward-only frame. The prior investment is neither added to nor subtracted from the value of future paths — it simply does not appear in the calculation, because it cannot affect outcomes.

How to do it
  1. 1Before any continuation decision, write the past investment (time, money, effort) in a column labeled "irrelevant."
  2. 2Now list only: what will happen if I continue? What will happen if I stop?
  3. 3Evaluate only those forward outcomes against each other.
  4. 4Make the decision that produces the better forward outcome, regardless of what is in the "irrelevant" column.
Evidence
RCT / meta-analysis

Arkes and Blumer (1985) demonstrated in controlled experiments that prior investment systematically increases willingness to continue bad courses. Zeroing-out framing is the standard economic and decision-theoretic antidote.

Honest caveat: Knowing about the fallacy reduces but does not eliminate it; the emotional salience of past investment is difficult to fully override even with correct analysis.

  • — Arkes & Blumer (1985), the psychology of sunk cost, Organizational Behavior and Human Decision Processes
Common mistake: Writing down the past investment and then trying to "ignore" it mentally — it remains visible and continues to influence judgment. Better to physically move it out of the working analysis frame.
Go deeper on this practice →

Practice this with IX Coach

Reading about a practice changes nothing on its own. IX Coach turns these into a guided, adaptive routine — discerning where you are in real time and walking the practice with you, session after session.

Start with IX Coach →

Related practices