Research
Working Papers and Research in Progress
Managerial Announcements and Investor Attention: Strategic Release of Good News (JMP)
Conferences: Financial Management Association Doctoral Consortium (2024), Behavioral Finance Working Group (2025), Financial Management Association (2025), European Finance Association Travel Grant (2025)
Abstract: In the context of corporate news dissemination, managerial decision-making involves considering several important factors, one of which is investor attention. This study provides empirical evidence showing a notable degree of predictability in investor attention levels. I also find that investor attention is lower on Mondays and Fridays compared to other weekdays. My findings also suggest that the likelihood of releasing positive news is much higher when predicted investor attention is also higher. Additionally, the study shows that the market’s reaction to earnings announcements is stronger when managers release good news during times of higher predicted attention. This connection highlights the significant effect of news release strategies on financial market behavior. I also show that managers are more likely to sell shares after releasing good news when predicted attention is higher. Based on these results, my research suggests that managers may strategically benefit from aligning their news releases with investor attention trends to sell shares.
Whose Attention Matters? Evidence from the Return Predictability Between Economically Linked Firms
with Mark Kamstra (Schulich) and Joseph Zhou (Ontario Tech)
Conferences: Financial Management Association Doctoral Consortium (2020)*, Southern Finance Association (2021)*, Eastern Finance Association (2021)*
Abstract: This study investigates the impact of limited attention from different types of investors in the context of return predictability between firms with a customer-supplier relationship, a collection of firms accounting for roughly 58.4% of the market capitalization of the US equity market. We fail to find a significant impact from the attention of institutional investors after controlling for retail attention, in contrast to the consensus in the literature which identifies institutional attention as instrumental. This result suggests a diminished role for the attention of institutional investors, or at least casts doubt on available measures of institutional attention. Finally, we find that the impact from heightened retail investor attention on return predictability is associated with the firm’s local retail investors not nonlocal retail investors, suggesting that it is local, presumably informed retail attention which facilitates information incorporation into asset prices. These findings also indicate that retail investors’ active involvement in the financial market helps enhance market efficiency, in at least some contexts.
ESG Bias in Analyst Revisions: Favoring the Green in Overvalued Markets (Draft available upon request)
with Ming Dong (Schulich)
Abstract: We document a negative causal effect of corporate ESG performance on analysts’ earnings revisions among overpriced stocks. Specifically, analysts are less likely to issue appropriate downward revisions for firms with high ESG scores compared to those with low ESG scores. This reluctance appears to stem from two key risks: (1) ESG sentiment risk— high ESG firms are more likely to experience sentiment-driven positive price shocks during periods of increased public attention to ESG, and (2) ESG reputation risk- analysts may fear damage to their professional standing when publicly releasing negative revisions for firms perceived as socially responsible. This bias in analyst behavior has important implications for asset prices and return predictability.
Noisy Attention: Evidence of Instability in Google Search Data
with Mark Kamstra (Schulich) and Joseph Zhou (Ontario Tech)
Abstract: Investor attention plays a critical role in shaping the behavior of capital market participants. The Google Search Volume Index (SVI) for ticker searches has become a widely adopted proxy for retail investor attention in the finance literature. However, the extent of noise and instability in this measure remains largely unexplored. In this paper, we investigate the reliability of Google SVI by conducting multiple downloads for the same set of firms over an identical sample period. We find that the resulting data vary significantly across downloads—even for the same firm and date—revealing previously undocumented noise in this commonly used measure. Importantly, we show that this noise is systematically related to firm characteristics. We further demonstrate that failing to account for this noise can lead to materially different empirical results, such as those based on stock price reactions to earnings announcements. Finally, we show that averaging SVI across multiple downloads substantially reduces this measurement noise, offering a simple but effective correction for future research that relies on Google SVI as a proxy for investor attention.
* : Presented by coauthor