Takeover Rumors: Returns and Pricing of the Rumored Targets

Abstract:     
Rumors can be classified into two types according to whether they can credibly predict impending events. Our analysis of takeover rumors of publicly traded US companies shows that public information on a rumored takeover target, particularly its historical Cumulative Abnormal Return (CAR) before initial rumor’s publication, is indicative of whether the pending takeover will be materialized. To our knowledge, this predictive power of historical CAR for the takeover rumor credibility has not been documented in the prior literature. However, abnormal returns on the day of and the day after the rumor’s publication are statistically indistinguishable between the two groups, although the groups can be distinguished by their historical CARs. This suggests that the market overreacts to takeover rumors that will not be materialized. Furthermore, trading on rumors is profitable if an investor invests selectively according to the historical CARs of rumored targets. The profitability of this trading strategy is in contrast to what has been documented in the prior literature. It is also found that rumors have substantial impacts on stock prices in takeover processes, leading to a relationship between takeover premium and price runup that differs from those revealed by previous studies. For the rumored targets, the projection of takeover premium on the stock price runup tends to be strictly greater than one, suggesting that bidders pay takeover targets “twice” because of revising the takeover offer upwards more than the runup. The difference between our finding and those of previous studies, such as the markup pricing or substitution hypothesis, suggests that the very existence of takeover rumors can have some material impact on bidders’ pricing strategies and the final realization of takeover premiums.

Number of Pages in PDF File: 55

Keywords: Takeover rumor, merger and acquisition, runup, markup, takeover premium

JEL Classification: G14, G34

working papers series