FGV EPGE professors win 2024 Haralambos Simeonidis Prize

EPGE Brazilian School of Economics and Finance (FGV EPGE) professors Leandro Gorno and Lucas Maestri, together with their co-authors Mehmet Ekmekci (Boston College), Jian Sun (Singapore Management University) and Dong Wei (UC Santa Cruz), received the 2024 Haralambos Simeonidis Award in the Article category for their work entitled “Learning from Manipulable Signals”. In all, only two articles were awarded by ANPEC this year. The result was announced during the 52nd National Meeting of Economics, promoted by the National Association of Graduate Centers in Economics (ANPEC).

The winning article addresses market dynamics and decision-making in scenarios of uncertainty, analyzing the impact of manipulable information on long-term economic relations, as is the case between investors and startups.

A venture capitalist (the principal) faces the decision of whether or not to continue funding a startup (the agent). The investor relies on performance reports provided by the startup, such as user growth or revenue data. However, these reports do not always reflect reality, as the startup can adopt strategies to “inflate” the figures.

The research models this interaction as a dynamic game, in which the investor gradually learns about the startup's potential based on these reports, while the startup evaluates whether (and to what extent) it is worth manipulating the data in order to maintain funding.

Among the main results of the article:

• Signal manipulation before disruptions: The study reveals that market terminations or crises are usually preceded by an increase in the intensity of manipulation and expected performance. This pattern, known as “scramble-to-rescue”, reflects an effort by the startup to avoid termination.
• Impact of transparency: Counterintuitively, the authors show that greater transparency is not always beneficial. In some cases, clearer signals can lead to more manipulation, making it more difficult for the investor to make a decision.
• Limits to learning: When the investor and the startup are willing to maintain the relationship for a long time, the manipulation of reports can intensify to the point of rendering the information received useless. This highlights the difficulty of learning in scenarios where there is a lack of long-term commitment between the parties.

The results help explain real-world situations, such as the cases of startups that performed impressively before collapsing, like Theranos and WeWork.

The study was published in the American Economic Review (AER), one of the most prestigious journals in the field of economics. 

The award

Created in 1982 by ANPEC, the Haralambos Simeonidis Prize annually recognizes the best academic works in the categories Articles, Books and Doctoral Theses, stimulating reflection activities and research in Economics in Brazil.

Translated with DeepL.com (free version)