When Players Are Conservative, the Economic System Becomes More Unpredictable

In A Note on Comparative Ambiguity Aversion and Justifiability (in Econometrica, Vol 84, Issue 5, 1903-1916, DOI: 10.3982/ECTA14429), Pierpaolo Battigalli, Simone Cerreia-Vioglio, Fabio Maccheroni and Massimo Marinacci (Bocconi's Department of Decision Sciences) go back to the analysis of the effects of ambiguity aversion on interactive decision-making processes, as they have recently done in another article in American Economic Review.

In rigorous terms, we face risk in a situation similar to a roulette, in which the distribution of an event probability is known (in roulette, for instance, every number from 0 to 36 has one chance to be hit out of 37), while we face ambiguity in situations similar to betting on horses, in which a player cannot reckon the probability distribution.

In their article in American Economic Review, the authors had combined the concept of self-confirming equilibrium developed by Battigalli, Fudenberg (Harvard University), and Levine (European University Institute) – an equilibrium players reach building on their experience and not through rational expectations - and the measures of ambiguity aversion made possible by Marinacci's studies, and concluded that greater ambiguity aversion means a larger set of possible equilibria.

In the Econometrica note they analyze, instead, a game theory situation, in which the players interact strategically (by considering, that is, the possible decisions of the other actors, given the knowledge they have of the latter). Each player has a set of justifiable actions (justifiable, that is, in the light of their subjective opinions and knowledge they have of the other actors) and the analysis shows that, when ambiguity aversion increases, the set of justifiable actions becomes larger. "From the point of view of an external observer", Battigalli clarifies, "this means that, when ambiguity aversion increases, the system becomes more unpredictable".

The analysis is also repeated considering risk aversion instead of ambiguity aversion, and the result does not change.

"This line of research could have interesting applications in the field of macroeconomic policies", Battigalli says, "and in fact we are working with Thomas Sargent (Nobel Prize for Economics in 2011, editor's note). We managed to get significant results using the concept of self-confirming equilibrium while, when trying to give a role to ambiguity aversion, we are facing economic models that are still too simplified. To make sure that different degrees of ambiguity aversion lead to different outcomes we will have to use more realistic and sophisticated models".