Sunk Cost Fallacy
Type: Decision — Commitment Also Known As: Concorde fallacy, escalation of commitment
Definition
Continuing an endeavor because of previously invested resources (time, money, effort) rather than based on current costs and benefits. Past investments are “sunk” and shouldn’t affect future decisions, but they do.
“We’ve already spent $2 million on this project — we can’t quit now.”
Form
- Resources are invested in a project/course of action
- Returns are below expectations or negative
- Rational analysis suggests abandoning the effort
- The prior investment is cited as reason to continue
- More resources are wasted attempting to “save” the sunk cost
Examples
Example 1: Bad Investments
An investor’s stock drops 50%. They refuse to sell, waiting to “break even” rather than accepting the loss and reinvesting in better opportunities.
Problem: The original purchase price is irrelevant to whether holding is wise now.
Example 2: Bad Relationships
Someone stays in an unhappy relationship because “we’ve been together five years.” The time invested becomes a prison.
Problem: Past time doesn’t obligate future time. Compatibility now matters.
Example 3: Failed Projects
A company pours more money into a failing product launch. The VP says “we’re too invested to stop now” while competitors innovate elsewhere.
Problem: The initial investment is gone. Continuing throws good money after bad.
Example 4: Entertainment
“The movie is terrible, but I paid $15 for the ticket, so I should stay.” Two more hours of suffering to “get your money’s worth.”
Problem: The $15 is gone whether you stay or leave. The question is: what serves you now?
Why It Happens
- Loss aversion — realizing sunk costs feels like admitting defeat
- Ego protection — quitting feels like failure
- Commitment consistency — we want to justify past decisions
- Waste aversion — we hate “wasting” resources
- Social pressure — others judge us by our commitment
How to Counter
- Ignore the past: Ask “if starting fresh today, what would I do?”
- Opportunity cost: What else could these resources achieve?
- Separate decision-maker: Have someone uninvolved evaluate
- Preset limits: Decide exit criteria before investing
- Reframe: Past spending was tuition, not obligation
When It’s NOT a Bias
Continuing despite costs IS valid when:
- There are genuine switching costs that make change expensive
- Reputation or signaling value requires persistence
- The learning/experience gained justifies the cost
- Abandoning creates cascade failures in other systems
Related Concepts
- Loss Aversion — Sunk costs feel like losses we want to avoid realizing
- Status Quo Bias — Continuing is the default path
- Escalation of Commitment — The organizational version
- Endowment Effect — We overvalue what we have (including investments)
References
- Arkes, H.R. & Blumer, C. (1985). The psychology of sunk cost
- Thaler, R. (1980). Toward a positive theory of consumer choice
- Staw, B.M. (1997). The escalation of commitment: An update and appraisal
Part of the Convergence Protocol — Clear thinking for complex times.
Note: Also see Sunk Cost for the logical fallacy perspective.