In the following post, D’Amore-McKim School of Business Professor Jeffery Born answers questions about his recent research that examines the impact that tweets from President Donald Trump have on a Semi-Strong Form (SSF) Efficient Market.
Q: Which tweets by Trump did you investigate?
A: We focused on tweets by President Donald Trump, which mentioned publicly traded firms (n=10) from the date of his election on November 8, 2016, to his inauguration on January 20, 2017. Fifteen tweets were separated by enough time for the stock market’s response to the information to be considered independent.
Q: Why did you look at these events?
A: In real time, there were many in the press who reported that the President-elect’s tweets were driving the firm’s stock price in a significant fashion. None of the press reports contrasted these movements against the same day movement in broad market averages, nor did the press report how risky the firms were (compared to the broad market). Failing to control for movements in the broad market and risk limits the conclusions one can draw the firm responses.
Q: How would you expect the stock market to respond?
A: In an Efficient Market, we would expect the market to quickly and completely incorporate the value of new information into share prices. The question here is whether the tweets by President-elect Trump contained new information. Our reading of the tweets suggests that there was little or no new information contained in the tweet. If our supposition is correct, any share price impacts should be small and short-lived (a couple of days).
Q: What were your findings?
A: After controlling for movements in the broad market and the riskiness of the firm’s stock, Trump tweets that conveyed a positive message (n=9) were associated with a statistically significant average abnormal price increase of about 1% on the day of the announcement (t=0), and another 1% the next trading day (but this movement was not statistically significant). Trump tweets that convened a negative message (n=6) were associated with an average abnormal price decrease of about 1% on the day of the announcement (but it wasn’t statistically significant), and another 1% decline the next trading day (t=+1) that was highly statistically significant.
Q: When did most of the price movement on the first trading day take place?
A: More than two-thirds of average abnormal price movements for both groups of firms (positive and negative) occurred from the opening bell to the close of trading that first trading day (t=0). Less than one-third of the average abnormal price movement was reflected in the opening price (when compared to the firm’s closing price from the day before the tweet.
Q: Did the price impacts ‘stick’?
A: From a statistical point of view, the answer is no. Within five days of the tweet, it was impossible to detect price impacts that could not be explained by movements in the broad market and riskiness of the firm’s shares.
Q: How did the tweets impact trading volume?
A: Compared with average daily trading volumes from earlier in 2016, trading was about 87 percent higher on the first trading day following the tweet and 56 percent higher on the day after the tweet was released.
Q: Who was doing the trading following the Trump tweet?
A: The identity of traders is not public, so we can’t answer that question unambiguously. However, when we look at the pattern of Google search activity for the firms in the week of the tweet, we found that search activity spiked dramatically. Consistent with other studies, we infer that the trading volume spike is probably due to small investor (so-called ‘noise traders’) activity.
Q: What has happened since Trump was inaugurated?
A: While President Trump has continued to tweet, we observe that the President has virtually stopped referring to publicly traded firms by name, so this experiment seems to have come to a natural conclusion.
Jeffery Born is a professor and group coordinator in the D’Amore-McKim School of Business Finance group. Born is on the editorial board of the Journal of African Business. His primary research interests focus on security market reactions to new information, particularly the response of share prices to changes in dividend policy.