When Regret Affects Human Decisions
Decision problems are often too complex to be solved by a rational approach. Heuristics are often needed to simplify decisions. It is also true that people reason in a complex way, hence, economic objectives are not the only ones playing a crucial role. It is not difficult to imagine that the combination of complex problems with complex human reasoning might lead to emotions being an important part in a decision-making process. Some thirty years ago, in order to explain behavioral anomalies, scientists turned to non-expected utility models with regret theory being one of the most psychologically appealing.
Regret is what we feel after making a wrong choice, but it can also influence the decision ex-ante. This means that people behave in order to prevent the regret feeling. Here is an example. Let's say that one has to choose to invest in either stocks or bonds. Assume for simplicity that investing in bonds always brings a small gain. If one invests in stocks, he would end up with a large gain, if market goes up, or in a large loss in case the market goes down. To minimize regret, the investor would probably end up investing in bonds. In this way, he would get a small gain and would avoid the feeling of regret and the loss of money that could come by investing in stocks if the market goes down.
Alessandra Cillo, Assistant Professor in the Department of Decision Sciences, with Aurélien Baillon and Han Bleichrodt, from Erasmus School of Economics in Rotterdam, explore regret theory in A Tailor-Made Test of Intransitive Choice, forthcoming in Operations Research.
Their paper addresses the question: Do people violate transitivity? "Regret theory predicts that violations of transitivity are possible", explained Cillo. According to transitivity, if John prefers A to B, and B to C, he should then prefer A to C. John's intransitive preferences: he prefers A to B and B to C, but prefers C to A, could lead to a "money pump", i.e. an arbitrage opportunity. Cillo illustrates with a following example: If John has C, but prefers B to C, he will give up C and pay 1 euro in order to get B. Now he will own B with 1 euro less in his pocket. Since he prefers A to B, he will give up B and pay 1 euro to get A. That means he has A with 2 euro less. Since he prefers C to A, he will give up A and pay 1 euro for C. So, at the end, John is in the same situation where he started off (he owns C), but with 3 euro less in his pocket.
So, according to Cillo's paper do people violate transitivity? Previous experiments by other researchers in the field do not provide a clear evidence: according to some of them, violations do exist, according to others they are due to errors. Cillo's work, which differs from others because it is tailor-made on each person that took part in the experiment, says that people don't violate transitivity. Does this mean that the regret theory is out of the window? "No, answered Cillo, "it does not mean that the regret theory does not hold. Rather, it suggests that the theory should be fine-tuned in order not to allow for violations of transitivity".