In "The escalation of commitment to a failing course of action," Joel Brockner (1992) explores escalation literature to discover the reasons decision makers tend to persist with failing courses of action. The following is a precis of Brockner’s paper.

In virtually all escalation situations, decision makers invest money, time, self-identify, and other resources with the hope they will attain goals. However, after investing resources the decision makers receive negative feedback that suggests they are not progressing toward the goal. Rather than changing course, the decision makers invest additional resources.

Self Justification

Self-justification has provided the dominant theory of escalating commitment.

However, some scholars criticize the self-justification argument as incomplete, suggesting alternative or supplementary explanations. Brockner argues that self-justification explains a significant portion of escalation, but not all of it. Combining competing and supplementary perspectives can provide managers with a more complete framework through which to explain escalation.

From the perspective of self-justifying or rationalizing behavior, decision makers persist in a failing course simply because they do not want to admit that they were wrong.

Alternative perspectives

Prospect theory and decision-dilemma theory provide alternative arguments for explaining escalation.

Prospect theory

Prospect theory explains the individual risk-taking propensities under conditions of uncertainty, assuming that changes of wealth are salient to decision makers. Prospect theory predicts that an individual's risk preferences will change depending on whether he or she perceives gains or losses. When faced with a failing course of action, the decision maker can chose to cut their losses or continue to commit. Loss aversion encourages risk seeking. While intended as an alternative to the self-justification argument, Brockner argues that self-justification and prospect both suggest that escalation increases when personal responsibility for the initial allocation is high.

Decision dilemma theory

Decision dilemma theory argues that self-justification is irrelevant in explaining escalation, proposing proposes that escalation is caused instead by the following motives: economic considerations, curiosity, the need to make a greater effort to see if persistence will lead to goal attainment, and to learn about the phenomenon. Brockner argues that decision dilemma theory does not adequately explain escalation without self-justification.

Supplementary theories

By considering the replacement and supplementary theories, managers might have a more adequate framework through which to explain escalation.

Antecedents of escalation

Different factors affect escalating commitment at different stages of escalation. In the first stage, project considerations, like costs and revenues, affect the decision makers. In the second stage, psychological and social factors become increasingly important. Psychological factors might include self-justification, decision-dilemma theory, and prospect theory. The social variables influencing escalating commitment include whether:

Situational and individual variables

While not empirically tested, other situational and individual variables affecting escalation may include:

Other factors that affect escalation include: leader behavior, group cohesion, and modeling. Powerful leaders can create a false sense of progress toward the group's goal, leading to escalation. Group polarization may

Expectancy theory

From the perspective of expectancy theory, decision makers determine if additional investment will lead to goal attainment and if the ultimate goal will be worth the investment. Focusing on what they expect and what they value, decision makers ignore the prior investments, focusing on whether persisting makes sense.


Brockner, J. (1992. "The escalation of commitment to a failing course of action: toward theoretical progress. The Academic of Management Review, Vol. 17, Issue 1; pg. 29, 23 pgs. Academy of Management.