6. Summary and conclusion
Risk seeking in the loss domain is one ofthe most salient contributions of Prospect Theory. However,
the theory has evolved and is supported by experimental evidence, and the realism of the theory
remains in question, especially when the stakes are high.
In this paper, I provide the beginning of that evidence of realism: The prediction of risk seeking is
not an uncommon practice of decision making in the real world. While hundreds of cases of riskseeking
behavior could be identified from publicly available sources, I limited the evidence presented
in this paper to some of the publicly-known cases that cost at least $200 million of direct loss. Of the
selected 60 cases, the identities of risk seekers were publicly known with assurance for 40 of them, but
only the substantive issues of 14 cases are summarized briefly in this paper. The summaries show the
decision situations and manners in which decision-makers actively sought more risky prospects
following the realization of losses.
While the evidence is anecdotal, it is nevertheless compelling due to its frequency, recurrence in
different types of businesses and environments, and cost. It is possible, therefore, to conclude that
risk-seeking behavior in the manner predicted by Prospect Theory observable in the real world, but it is
not possible to use this evidence to argue that risk seeking is a characteristic of all decision-makers.
Accordingly, it would be a contribution to the literature if one were to identify the unique
characteristics of these decision-makers that separate them from the rest.
Additionally, the fact that all identified risk seekers are young males should be of research interest
to many social scientists as further evidence that men are more aggressive in risk taking than women.
Finally, these decision-makers were able to use the weaknesses in the accounting and administrative
systems in their organization to either camouflage their losses or even present them as gains. The
processes that enabled them to accomplish these tasks should be of interest to researchers in
accounting and management control systems.