EV-0025
Commitment has a failure mode, and it is a character failure
Sleesman, D. J., Conlon, D. E., McNamara, G., & Miles, J. E. (2012). Academy of Management Journal, 55(3), 541–562 · Source
What this entitles us to say
A meta-analytic review of 35 years of escalation research finding support for 14 of 16 predicted determinants of persisting with a failing course of action — with sunk costs, personal responsibility for the original decision, and ego or reputational threat among the strongest.
What it found
A synthesis of 35 years of research on escalation of commitment — continuing to invest in a failing course of action — testing the competing theoretical accounts against each other. Support was found for 14 of 16 main-effect predictions, including:
- Sunk costs and time already invested - Personal responsibility for the initial decision - Self-efficacy and confidence, and personal experience or expertise - Ego threat — reputational exposure — reported among the strongest predictors - Anticipated regret
Self-justification — the need to validate a past decision by making it come good — emerged as a key mechanism.
Why it matters here
This is the counterweight for Commitment, and it is the sharpest one in the Achievement region because it implicates the trait's own virtue. The very things that make someone committed — ownership of the decision, investment already made, confidence — are the things that make them unable to stop.
The finding that matters most for a character program is ego threat. People escalate because backing down would embarrass them. That is not an analytical failure that better data fixes; it is the leader protecting their own image at the organization's expense, which is precisely the sort of thing this program claims to be about.
It gives the Commitment module its honest second half. The book teaches counting the cost and paying the price. The research adds the question the book does not ask: what do you do when the price turns out to be higher than the thing is worth, and everyone knows you chose it? That is the module's hardest scenario, and it is a Truth-seeking problem as much as a Commitment one — which is why this entry is tagged to both.
The limits
Much of the underlying escalation literature is laboratory work with student participants making hypothetical investment decisions, which is a long way from a chief executive killing a real program. Effect sizes vary considerably across determinants and this entry summarizes the pattern rather than quoting any of them. Verified to abstract level. Note also that persistence is not always wrong — the research is about persisting with a failing course, and telling the two apart in real time is the actual difficulty, which this literature does not solve.
What we do not say
"Committed leaders make worse decisions," or any suggestion that commitment is a liability. The research concerns persistence with a failing course, and distinguishing that from legitimate perseverance in real time is exactly what this literature does not resolve.