Research

This paper studies the impact of information acquisition methods on individuals’ attention when predicting an asset value. Using a behavioral Bayesian updating model, I study two important aspects of information acquisition methods, i.e., the impact of cost and the impact of giving individuals the option to select signal precision levels, labeled discretion. The predictions of the model are tested in a laboratory experiment in which the incentives are meant to be, to some extent, naturalistic. The experiment allows to document that when the signal precision level is exogenously provided and free of charge, an increase in signal precision levels leads to decreased attention. When the signal precision level is exogenously provided and costly, attention is unaffected by the varying signal precision levels. If the signal precision level is endogenously decided and costly, i.e., participants choose the costly signal precision level by themselves, the higher the signal precision level, the higher the attention. I provide evidence that both information acquisition cost and discretionary power positively impact attention. The effort effect is attributed to the effect of the effort’s product, but not the effect of the effort itself.

This paper studies in theory and in an experiment the two fundamental environments which are available to firms that seek investments, i.e., whether they face a fixed return that is known to the investor and the entrepreneur ex-ante, or a variable return, that is decided ex-post and by the entrepreneur only. The paper allows us to estimate the cost of potential betrayal in investorentrepreneur investment relationships and documents how the obligation of a return (or the lack thereof) impacts the investor’s investment decision and the entrepreneur’s resource allocation decision. Our theoretical work allows us to document that having a fixed return leads to both a higher transfer from the investor and a higher engage ratio from the entrepreneur into the project. As predicted by theory, the experimental results show that the transfer and engage ratio and amount are higher in the fixed-return treatment compared to the variable-return treatment. While the results of the transfer behavior may be partially explained by the higher expected return and the lower variance induced by the fixed return, the empirically-documented behavior on the engage behavior is not fully explained by rational expectations. We find that what is financially engaged in the project is moderated by the received amount, which means that the higher the amount received by the firm, the higher the amount engaged in the project. We also find that the engage ratio is less sensitive to the change of received amount in the fixed-return treatment.

This paper studies the impact of a reputation system that combines information related to truthtelling and information related to economic performance on firms’ and investors’ behavior in an investment game. Such a reputation system allows us to link the two important aspects of reputation - honesty and performance - which have been studied so far separately. We analyze firms’ and investors’ strategies using a cheap talk framework and use laboratory experiments to test the model’s predictions. Both in theory and in experiments, we observe that firms’ truthful behavior can be induced via this reputation mechanism, especially for bad-type firms and environments in which the success rate is low. Second, the reputation system helps investors distinguish between good and bad investment firms via the reputation score, which explains the investors’ investment difference between good-type and bad-type firms. The higher the reputation score, the higher the propensity of investors to invest. Third, we find that different types of investment firms treat reputation differently: good-type firms’ announcements are positively impacted by their probability of success and reputation score while bad-type firms’ announcements are only related to the probability of success. Thus, although bad-type firms do not seem to care about reputation, the fact that only good firms treat reputation as an asset is enough to allow investors to distinguish good-type firms from bad-type firms. Our findings help inform questions related to the design of reputation systems in investment markets in future studies.

Limited attention has been devoted in the literature to exploring the nexus between market power and the dispersion of analysts' earnings forecasts. This study introduces a nuanced examination of market power, encompassing both market concentration (measured by the Herfindahl-Hirschman Index) and pricing power (quantified through the Lerner Index). Employing a comprehensive dataset amalgamated from I/B/E/S, CRSP, Compustat-IQ, and SDC Platinum, the analysis delves into the dynamics of analysts' earnings forecast dispersion, particularly before and after publicly listed companies' mergers. Companies wielding higher market power exhibit a notable reduction in complexity, contributing to analysts' ability to forecast earnings with diminished dispersion. This phenomenon is attributed to analysts' enhanced prognostic capabilities concerning the future performance of such companies. Furthermore, when contextualizing this relationship within the broader landscape of micro- and macro-level uncertainties, the adverse impact of market power on forecast dispersions becomes even more pronounced. This study unveils a structural transformation in the analytical model, discernible across distinct observation windows preceding and succeeding the merger event. The findings underscore the intricate interplay between market power dimensions and analysts' earnings forecast accuracy, shedding light on the multifaceted implications of market power within the corporate landscape.