Publications
Tra i Leoni: Revealing the preferences behind a superstition
with Giovanna Invernizzi, Joshua B. Miller, Martin Dufwenberg and Luiz E. Oliveira, Journal of Economic Psychology 2021, 82.
We examine a superstition for which adherence is nearly universal among its target population. Using a combination of field interventions that involve unsuspecting participants and a lab-style value elicitation, we investigate the nature and strength of peoples’ underlying preferences. While a substantial minority of people are willing to incur a relatively high individual cost in order to adhere to the superstition, for many, adherence is contingent on the behavior of others. Our findings are consistent with the idea that it is the conforming nature of the majority that sustains the false beliefs of the minority.
Working Papers
Partial Identification of Welfare Effects in the Presence of Demand Frictions
The validity of traditional welfare analysis in economics, based on the revealed preference paradigm, can be undermined by demand frictions such as default options, cognitive limitations, and limited or distorted information. This paper develops a general framework for discrete choice welfare analysis given quasi-experimental interventions that remove such frictions, which relies on relatively minimal assumptions on individual heterogeneity and overcomes key limitations of existing methodologies. It does so by mapping this problem to the literature on identification of functionals of the joint distribution of two potential outcomes, and takes a partial identification approach. I illustrate the approach in the context of product demand with non-salient taxes.
Censorship and Experience Goods. Evidence from Online Wine Retail
with Igor Cerasa
This paper investigates the role of pre-purchase information in the market of experience goods, specifically in the context of online wine retail. We study how sellers influence the perceived quality of these goods by strategically censoring information. We create a unique dataset of expert ratings combining an independent source and a major online retailer. Our results document that the seller is likely to censor the lowest scores. To understand if consumers are in turn affected, we design and implement an online stated choice experiment, combined with an information treatment. We estimate the impact of expert scores on consumers' demand and test whether consumers are aware of censorship. Our results show that consumers care about expert ratings, and wines with higher scores are more likely to be selected than wines with lower. Our information treatment did not stimulate a change in consumers' behavior. We propose two alternative explanations: that consumers are naive, meaning that they always act as if they observe ratings from a complete source, alternatively, that consumers are sophisticated and are not surprised to learn about the seller's behavior.
Bayesian Persuasion with Private Types
I study an extension to the simple Judge/Prosecutor example in Kamenica & Gentzkow (2011), where the Judge has a private type. The distribution of types (G) and the one of states (F) are commonly known. I provide a simple method to determine the optimal persuasion mechanism, under the assumption that G is single-peaked. I then provide some partial results for general G’s, for the case where the state space is non-binary, and for situations where there are multiple receivers who care about each other’s actions.
Work in Progress
Optimal Transport and the Measurement of Inequality
with Vesa-Matti Heikkuri and Matthias Schief