Kimberly

Input:#

• Hypothesis: Deals announced during Covid are more often done by overconfident CEO’s

  • Think about the data required to do this: what is the DV and what is the IV?

• Deals done during Covid generally have a higher book to market ratio than before Covid

  • Why would you exactly want to focus on a book-to-market ratio, rather than just stock price returns?
  • You would expect the stock price returns to be negative, in part because they are effectuated by overconfident CEO’s, and overconfident CEO’s overrate their private information.

Intro#

Particularly, the shareholders of target firms earn positive abnormal returns following the announcement, and the bidding firm realizes negative to zero abnormal returns (Andrade et al., 2001).

Theoretically it could also be otherwise, maybe you should explain that.

What effect does an acquisition announced by an overconfident CEO have on firm value during the Covid-19 pandemic?

I think you should rephrase the question, because you are not only investigating merger announcements of overconfident CEO’s, but also of non-overconfident CEO’s as a control group. The previous empirical literature established that CEO overconfidence is associated with a larger negative premium. You think that, as a result of Covid, this negative premium becomes less large, so that an overconfident CEO is less harmful for the acquirer company’s shareholders.

What is the effect of the Covid-pandemic on the difference in acquisition premium between overconfident and non-overconfident CEOs?

(Or something in this direction).

Small model#

Some theory to help you thinking in the right direction. Valuation of a to be acquired firm for a manager of an acquiring firm $i$ is equal to:

$$ V_i = a_i (1 + \delta)$$ where $a_i \sim N(a, \sigma^2)$, $a_i$ is the private opinion of the manager $i$ and $a$ is the firm’s valuation on the stock market. $\delta$ is the degree to which the manager is overconfident and overvalues their signal.

Takeover if: $V_i > a:$

$\mathbb{P}(V_i > a) = \mathbb{P}(a_i (1 + \delta) > a) = \mathbb{P}[\frac{a_i - a}{\sigma} > \frac{a - (1+\delta) a}{\sigma (1 + \delta)}] = \Phi (\frac{a \delta}{\sigma(1+\delta)})$

Conlusion 1: higher delta, higher likelihood of wanting to engage in takeovers (like you want).

Then, manager of acquiring firm encounters 1 manager of the target firm. The manager of the target firm holds all the bargaining power, so they can charge the maximum willingness-to-pay of the acquiring manager: $p = v_i = a_i (1 + \delta)$. The premium is then $a_i (1 + \delta) - a$ and the expected premium is the expected value of that: $a (1 + \delta) - a = a \delta$.

Conclusion 2: The expected takeover premium is $a \delta$, so more overconfident managers (with a higher $\delta$) pay a higher premium. Similarly, the expected takeover premium for non-overconfident managers is 0 (they pay exactly the right price on average).

Conclusion 3: Take a look at the expected premium. It is a function of firm value $a$. What if in coronavirus times, $a$ is lower?

Further comments#

I think your hypothesis 2 and 3 contradict each other. Can you explain this?

You don’t explain yet how you will measure CEO overconfidence.