Elon Musk’s behavior for Tesla: a blessing or a curse?

Nowadays, many successful companies have leaders with strong personalities. Nearly everyone can associate the name Mark Zuckerberg with Facebook, Steve Jobs and now Tim Cook with Apple, and Jeff Bezos with Amazon. In recent years, the name Elon Musk has similarly become publicly known as the Chief Executive Officer (CEO) of Tesla and SpaceX. However, it can be questioned whether all of the above have a positive impact on the company. The aim of this thesis is to answer the following research question: “Is Elon Musk a blessing or a curse for Tesla?” The study is conducted around 14 events that are tied to the behavior of Elon Musk (good or bad) from April 2018 until May 2020. The research is conducted using an event study analysis. The events will be tested for abnormal returns with a two-tailed test to answer the following research sub question: “Does the behavior of Elon Musk trigger abnormal returns of Tesla’s stock price?” Moreover, the events will be assigned a sentiment (positive, negative or neutral) and then tested with a one-tailed test to answer the following to sub questions: “How many events triggered a positive effect on Tesla’s return?” and “How many events triggered a negative effect on Tesla’s return?” Finally, a new event study method, the threshold method, is applied to test whether it is possible to forecast, based on past events, if an event was tied to Elon Musk’s behavior or not. The analysis results in concluding that only 2 out of 14 events had a significant overall impact on the company because they triggered abnormal returns. However, when we test for significance with a one-tailed test, we find that 2 out of 5 positive events triggered abnormal positive returns (40%) and that 4 out of 8 negative events (50%) triggered abnormal negative returns. Based on the higher ratio of negative events triggering abnormal returns, we can conclude that Elon Musk’s behavior is rather a curse for Tesla than a blessing, because his actions with negative sentiment trigger more volatility than his actions with a positive sentiment. Moreover, based on the threshold method, we can also conclude that out of the 14 events, it was possible to forecast 12 events in the right direction (positive or negative). Indeed, one event was assigned a positive sentiment and yielded negative returns and one event was assigned a negative sentiment and yielded positive returns.