Working Papers
Working Papers
The Salience Effect and Investor Behavior - Job Market Paper
Presented at: AEA* 2026 (poster), FMA Annual Meeting, Brownbag Seminar at the University of Mannheim, 9th HEC Paris Finance PhD Workshop, 18th International Behavioural Finance Conference, AFFI PhD Workshop 2025, DGF 2024, Behavioral Finance PhD Workshop 2024
Abstract: This paper provides direct behavioral evidence linking individual trading behavior to the market-level salience effect. Salience captures how much a stock’s return stands out relative to the market average, both on the upside and the downside. Salience theory predicts that investors buy upside-salient stocks and sell downside-salient stocks. This trading pattern causes misvaluation which is subsequently reversed, resulting in a negative relation between salience and subsequent returns. I test this prediction using investor-level trading records and find that individual investors instead buy both upside- and downside-salient stocks. However, only the overvaluation of upside-salient stocks translates into market-level mispricing, particularly during boom periods. Experimental evidence further supports this mechanism: participants invest more in risky assets after observing upside-salient outcomes compared to their non-salient counterparts. Together, my findings show that individual investors do not fully drive the salience effect; rather, the asymmetric market impact of trading on upside-salient stocks underlies salience-based mispricing in financial markets.
*Scheduled
Social (Non)Interaction and Trade Decisions of Mutual Fund Managers - joint with Erik Theissen
Presented at: FMA European 2024, 17th International Behavioural Finance Conference, DGF 2023
Abstract: We investigate how social interaction among portfolio managers affects their trading decisions and portfolio composition. We leverage the stay-at-home orders during the COVID-19 pandemic. They imply a negative shock to in-person communication between fund managers located in the same city but should have little effect on communication between managers located in different cities. We implement a difference-in-differences analysis at the fund-pair level, with fund pairs being managed by managers located in the same city [in indifferent cities] constituting the treatment [control] group. We find a significant decline in the similarity of buy decisions and portfolio holdings among same-city managers, relative to the control group, during the work-from-home period. These results provide evidence that investment-relevant information is transmitted through in-person social interactions.
Salience Bias in Belief Formation - joint with Martin Weber
Presented at: M-BEES/M-BEPS 2024, AEA 2024 (poster), EFA 2023 (poster), EEA-ESEM 2023, Boulder Summer Conference on Consumer Financial Decision Making 2023 (poster), Experimental Finance Conference 2022
Abstract: We propose an experimental framework to study how salience shapes belief formation. In a controlled experiment, participants receive sequential signals, with some signals made more salient by design. Bayesian updating predicts that signals should be weighted according to their informativeness, yet we find that upside-salient signals are overweighted, even when this reduces overall accuracy. This bias is driven by a subset of individuals, salient thinkers, whose behavior is consistent with salience theory. They overweight upside signals even more when such signals are rare and are more confident than other participants. Our findings provide direct evidence that salience distorts belief updating, highlighting attention as an important driver of deviations from rational expectations.
A Credit Design for Pessimistic Entrepreneurs - joint with Umut Keskin
Abstract: We propose a credit mechanism designed to encourage pessimistic entrepreneurs to engage in economic activity. In our model, entrepreneurs are categorized into two types: pessimistic and rational. In a simple setting with three possible states of nature, pessimistic entrepreneurs overestimate the severity of the worst state. As a result, they refrain from undertaking certain projects, even when these projects have a positive expected return. The primary objective of our proposal is to establish a contract between the government and pessimistic entrepreneurs that balances the subsidy provided in loss states with the profits shared in gain states. By implementing this mechanism, the government can mitigate rising pessimism among potential entrepreneurs during economic downturns, thereby stimulating the economy without distorting the behavior of rational entrepreneurs.