Escalation of Commitment (Staw's Model)

Barry Staw's research on why decision-makers increase investment in failing courses of action rather than cutting losses. Distinct from simple sunk cost fallacy, Staw's model identifies four reinforcing drivers: (1) project factors—ambiguous feedback that allows optimistic interpretation, (2) psychological factors—self-justification, loss aversion, and completion desire, (3) social factors—need to save face, maintain consistency, demonstrate toughness, (4) organizational factors—political consequences of admitting failure, institutional momentum, and embedded stakeholders. The Concorde supersonic jet is the canonical case (hence 'Concorde fallacy'): both British and French governments continued funding despite knowing it would never be profitable, because stopping meant admitting decades of waste. Vietnam War escalation, corporate IT projects that balloon 500% over budget, and relationships where both parties know it's over but neither will say so all follow this pattern. The deepest insight: escalation feels like optimism and perseverance—virtues—which is why it's so resistant to rational intervention.

When to use it

When a project, relationship, or investment is underperforming but you keep investing. When 'we've come too far to stop now' is the primary argument for continuing. When you'd never start this if you weren't already in it. During any review of struggling initiatives.

How it can help

Provides a multi-factor diagnostic that goes beyond 'don't fall for sunk costs.' For each failing commitment, assess: (1) is the feedback genuinely ambiguous or are you selecting optimistic signals? (2) would you make this investment today if you were starting fresh? (3) are social/face-saving pressures influencing the decision? (4) who benefits from continuation regardless of outcome? Prevention requires: pre-committing to kill criteria before starting (if X metric isn't achieved by Y date, we stop), separating the decision to continue from the people who made the original decision (to remove self-justification), and conducting regular 'zero-base' reviews where continuation must be argued as if it were a new investment.