'Attention Triggers and Retail Investors' Trading Behavior'
(COAUTHORS: M. ARNOLD AND M.G. SUBRAHMANYAM)
This project investigates the impact of individual attention on investor risk taking and short selling. We analyze a large, novel sample of trading records from a European brokerage service that allows its customers to trade contracts-for-differences (CFD), and sends standardized push messages on recent stock performance to its client investors. The advantage of this sample is that it allows us to isolate the "push" messages as individual attention triggers, which we can directly link to the same individuals' risk taking. A particular advantage of CFD trading is that it allows investors to make use of leverage, which provides us a pure measure of investors' willingness to take risks that is independent of the decision to purchase a particular stock, and short selling. Leverage is a major catalyst of speculative trading, as it increases the scope of extreme returns, and enables investors to take larger positions than what they can afford with their own capital. Short selling provides important flexibility for investors to adjust their portfolio to reflect their changing views. We show that investors execute attention-driven trades with higher leverage, compared to their other trades, as well as those of other investors who are not alerted by attention triggers. Attention traders also take short positions more frequently. Working Paper
In a second paper, we also investigate the impact of attention on portfolio diversification. We show that the effect of push messages on investors' portfolios critically depends on whether investors hold a stock in their portfolio at the time of the push message. Investors who do not hold the stock in their portfolio when receiving the message hold better diversified portfolios 24 hours after the push message, while investors who hold the stock in their portfolio at the time of the message decrease their diversification. Hence, we highlight two opposing channels for the impact of attention triggers on risk taking. Whereas (i) attention triggers induce investors to take more idiosyncratic risk, (ii) these triggers also stimulate portfolio diversification.
Amongst others, our project complements studies that explore the reason behind retail investors' mistakes. Hence, the results of our project have several practical implications for investment management. Specifically, it allows investors and investment advisors to understand the impact of attention on individuals' trading and portfolios. Using this knowledge, advisors can provide guidance to their customers to improve their trading outcomes. For example, our results suggest that purposefully sent push messages may help investors overcome the common mistake of holding under-diversified portfolios. As such, the use of push messages may have stronger effects on investors' portfolios than other types of financial education.
Financial support by the Fritz Thyssen Stiftung (Az. 50.19.0.020WW) is gratefully acknowledged.
- OR AG FIFI Workshop, January 2019.
- 5th European Retail Investment Conference (ERIC), April 2019.
- Fourth CEPR European Workshop on Household Finance, May 2019.
- The Future of Financial Information Conference, May 2019.
- FEBS 2019 International Conference, May 2019.
- Behavioral Finance Working Group Conference, June 2019.
- 81. Jahrestagung des Verbands der Hochschullehrer für Betriebswirtschaft (VHB), June 2019.
- European Financial Management Association Annual Meeting, June 2019.
- Academy of Behavioral Finance & Economics Conference, September 2019.
- American Economic Association 2020 Annual Meeting, January 2020.
'Social Interactions and (Financial) Decision-Making'
(COAUTHOR: D. LOSCHELDER)
This project studies the impact of social interactions on investors' financial decision-making and risk taking. In particular, we study preferences for dependencies between payoffs for own prospects in relation to payoffs of peers' prospects and the implications of such dependencies for investors' risk taking.
Financial support by the Deutsche Forschungsgemeinschaft (PE 2525/5-1) is gratefully acknowledged.