Prospect Theory Goes to War: Loss-Aversion and the Duration of Military Combat

This paper contributes to the empirical foundation of prospect
theory in real-life international relations by testing two of its
major implications in the field of military conflict. Using
duration analysis for a data set of twentieth century battles, it
is shown how the experience of losses contributes positively to the
preparedness to continue fighting, up to a point where casualties
clearly outweigh any direct utility drawn from ordinary
expected-utility theory. Moreover, the empirical results also
indicate that the relative position compared to the opponent's is
clearly less important for the decision whether to stop a battle or
not than the change of one's own position compared to the beginning
of the fight.