The Virtuous Circle of Investor Attention in Social Media

In previous posts, we discussed how media and firms can succefully draw investor attention. In this post, we shift the focus to investors themselves. The 2016 Journal of Behavioral Finance paper “Investor Attention on the Social Web” by Li, Hendler, and Teall asks whether there is a virtuous circle of investor attention, i.e., whether investor attention is “contagious”.

The authors analyze investor message data on the social media platform StockTwits.com. Founded in 2008, StockTwits has 1.1 million users who share their trading ideas in tweet-like text messages.

Identifying the virtuous circle of investor attention

To measure whether investor attention draws more investor attention, Li et al. (2016) proceed as follows. First, they divide the investor universe into two groups, the stimulus and the affected crowd. Investors who tweet about a stock on a given day and at least once during the previous 3 days form the stimulus crowd. Investors who tweet about the stock on a given day but not during the previous three days form the affected crowd. In the second step, the authors run a linear regression. The size of the affected crowd is the dependent variable, and the size of the stimulus crowd is the main explanatory variable.

Actions based on peers’ social cues

The basic idea behind this type of analysis is the social cue effect: Our actions depend on cues we receive from our peers. To cleanly identify the impact of the social cue vs. other effects, the authors include a number of control variables (such as firm size, or whether the stock price jumped). The regression runs separately for groups with increasing volatility. Table 1 presents the main result.

Quantile Stimulus crowd size Dummy (jump) Firm size
1 0.363 0.138 0.135
2 0.421 0.099 0.121
3 0.437 0.111 0.085
4 0.479 0.146 0.054

Table 1: Impact of the size of the stimulus crowd on the size of the affected crowd. Information taken from Table 3 and 5 of Li et al. (2016). Bold font indicates statistical significance at the 5% level or higher.

Crowds draw crowds!

Table 1 clearly shows that the affected crowd is larger when the stimulus crowd is larger. This is true for all volatility groups, and the effect is even stronger for higher volatilities. Hence, investors rely on social cues more when there is more uncertainty. Also, the effect of the stimulus crowd size is larger than that of the control variables. So, social cues are a more important factor for investor attention than firm size or recent jumps in stock prices.

Take-away: Investor attention is a virtuous circle

In summary: Investors look for social cues when allocating their attention, and rely on their peers to provide such cues. The result is a virtuous circle of investor attention.

Author: Tapas Mohapatra