Abstract
Arkes and Blumer show sunk-cost effects — people continue investing because resources were already spent, even when only future payoffs should govern the decision — across multiple experimental paradigms.
Arkes, H. R., & Blumer, C. (1985). Organizational Behavior and Human Decision Processes, 35(1), 124–140. DOI90049-4)
Start here
Read Reversibility and IKEA effect — sunk cost is the normative violation behind escalation and over-valuing your own build.
Key findings
- Irrelevant past investment — Money, time, or effort already spent biases stay/quit choices — only marginal future costs and benefits should count.
- Escalation pattern — Larger prior investment increases commitment to failing lines — not because future EV improved.
- IKEA effect — Effort invested in building raises valuation — cousin bias, not identical mechanism but same coaching week.
- Reversibility — Sunk cost fights clean quit — reversibility design reduces trapped capital.
Limits
Experimental economics and psychology paradigms — real projects carry reputation and relationship costs not in the lab. Use to interrupt escalation, not to forbid all persistence.
When to open this
Open when IKEA effect or Reversibility is live — separate sunk cost from genuine commitment before the next escalation.
Try this week: one project you continue because “we’ve already invested” — rewrite the decision using future-only payoffs.