Academic literature on the effects of political news is still deficient and has not been conclusive yet. Although a couple of researchers have attempted to assess the role of political risk, there is no clear-cut empirical evidence on the effects. The first time, the relationship between an individual Twitter account, namely Tweets by U.S. president Donald Trump, and stock returns are examined. There are two main reasons to link them together.
- Every political move is documented in his Twitter account and is often ‘leaked’ on Twitter before official announcement. Even the White House recently clarified that his Tweets represent presidential statements, carrying the same imprimatur as a comment issued by his press office.
- Donald Trump is among the most controversial users of the platform. He uses Twitter extensively and captures the attention of a wide audience.
The methodology goes as follows.
Tweets and minutes timestamps from Donald Trump’s account @realDonaldTrump for the period September 2018 to May 2019 are collected, resulting in a total of approximately 3,200 Tweets. The sample is filtered using keywords such as “China”, “Trade war”, “Tariffs”, among others. This results in a final sample of 224 Tweets.
The results indicate an impact on returns and market uncertainty.
To get a first impression of the impact of his Tweets, data is split into two subsamples: ‘with tweet’ and ‘without tweet’. As the figure below shows, there is a significant difference between average return ‘with tweet’ (–0.46%) and ‘without tweet’ (0.06%) (p-value < 0.01) for industries with a high degree of trade intensity with China. Results indicate that Trump Tweets predict stock returns for multiple days. While his Tweets appear to adversely affect returns one day after release, the direction is reversed on day two, which results in positive returns. Consequently, the market reacts to Trump Tweets, but mean-reverses in subsequent periods. However, this does not fully offset the effect from the previous period. In addition, there is a significant difference between average VIX (CBOE Volatility Index) ’with tweet’ (–1.12%) and ’without tweet’ (2.44%) (p-value < 0.05), indicating that Tweets significantly predict an increase in market uncertainty.
