Skip to main content

Bounded rationality and social concerns in decision processes: theory, experiments, and applications

Final Report Summary - BRSCDP-TEA (Bounded rationality and social concerns in decision processes: theory, experiments, and applications)

Individuals take daily economic decisions that deliver random outcomes. Financial decisions are a prominent example of this. In Frank Knight's view randomness is best understood if split in two different components: risk and uncertainty. Risk refers to a decision in which the likelihood of the outcomes is well understood and randomness is measurable. Following the terminology of Anscombe and Aumann, a bet on a fair roulette wheel is the simplest example of such randomness. On the other hand, uncertainty refers to a source of randomness which is not entirely understood by the decision maker (henceforth, DM) or, equivalently, which is not measurable. This translates into the absence of a well known probability law governing realizations. Following again Anscombe and Aumann, a bet on a horse race exemplifies such randomness.
The early writings in Economics focused on risk. Nevertheless, in the past twenty-five years a literature of choice under uncertainty blossomed. Models were developed and some of them found several fruitful applications in Finance and Macroeconomics.
One contribution of our research project is to provide a systematization of the existing literature and the existing models, by formally connecting them and by characterizing their common features. In particular, this systematization is most successfully developed for models built to capture a preference for diversification of the DM. This is done by proposing a rich and unique theory which is able to describe the DM’s uncertainty attitudes and to encompass a large part of the models developed in the last years.
In terms of social choice theory, we were able to provide a behavioral foundation of interdependent preferences, where the outcomes of others affect the welfare of the DM. An axiomatic foundation that could take into account social emotions (pride/envy) was established thus providing the link between observation of choices and a functional representation which is convenient, free of inconsistencies and can provide the basis for measurement.
In terms of bounded rationality, due to limited foresight, a model where bounded rationality and its extent emerge endogenously was developed. The model allows for predictions with respect to how changes in the terminal payoffs affect the depth of search and, consequently, choices.
The PI and his team studied some of the implications of the above decision theoretic models in Finance, Game Theory, General Equilibrium, Statistics, and Experimental Economics.
Finally, part of the previous research yielded some novel mathematical results.