Merger Negotiations with Stock Market Feedback

The following post comes to us from Sandra Betton of the Department of Finance at Concordia University; B. Espen Eckbo, Professor of Finance at Dartmouth College; Rex Thompson, Professor of Finance at Southern Methodist University; and Karin Thorburn, Professor of Finance at the Norwegian School of Economics.

In our paper, Merger Negotiations with Stock Market Feedback, forthcoming in the Journal of Finance, we investigate whether pre-bid target stock price runups increase bidder takeover costs—an issue of first-order importance for the efficiency of the takeover mechanism. We base our predictions on a simple model with rational market participants and synergistic takeovers. Takeover signals (rumors) received by the market cause market anticipation of deal synergies that drive stock price runups. The model delivers the equilibrium pricing relation between the runup and the subsequent offer price markup (the surprise effect of the bid announcement) that should exist in a sample of observed bids.

A key model innovation is to allow the takeover signal to inform investors about potential takeover synergies, so that a stronger signal implies higher expected synergies and deal values. While this type of signal seems reasonable, it creates a runup-markup relation that is more complex than previously thought. Specifically, in settings where offer prices do not respond to runups, runups are conventionally thought to be offset dollar-for-dollar by a decrease in the markup. We show that this prediction presumes that the takeover signal is uninformative about deal synergies. The runup-markup relation with our more informative takeover signal is always greater than minus one-for-one and is inherently nonlinear, which our large-sample evidence supports.

We next add a costly feedback loop from runups to offer premiums to the model. In this loop, merger negotiations cause the bidder to transfer the runup to the target although the runup is driven by market anticipation of target deal synergies—the bidder effectively pays twice for anticipated target deal synergies embedded in the runup. Importantly, the costly feedback loop implies a strictly positive relation between runups and markups, a prediction that our empirical analysis rejects.

This empirical conclusion is robust to the method used to estimate runups and markups. In particular, adjusting target runups for an unbiased estimate of target stand-alone value changes over the runup period only strengthens the rejection. Our evidence therefore presents a strong rejection of the notion that runups increase bidder takeover costs: specifically, bidders do not pay twice for anticipated deal synergies embedded in run ups.

A further perspective on the presence of deal anticipation in runups is provided by studying bidder returns. In our model, bidders act rationally and share in the takeover synergies. This implies that, when the market anticipates a synergistic takeover and target runups do not increase the offer premium, bidder gains will be increasing in both the target runup and total target gains. This prediction is a direct implication of our assumption that the takeover signal informs investors about conditional deal values in addition to takeover probabilities. The empirical evidence strongly supports this prediction, which in and of itself contributes to the ongoing debate over the existence of bidder takeover gains but also lends support to the basic information structure underpinning our theoretical analysis.

Finally, we examine direct effects on the offer premium of exogenous shocks to the target stock price during the runup period. We show that block trades in the target, including bidder toehold purchases, tend to fuel target runups without increasing offer premiums (if anything, premiums are lower in bids where those trades occur). Moreover, the market return over the runup period is found to flow through to target shareholders by an almost dollar-for-dollar increase in the offer premium. This type of flow-through is reasonable as it does not increase bidder takeover costs.

What if, contrary to our model, the market were to systematically ignore the information in takeover signals (a type of market inefficiency)? This alternative hypothesis implies that markups will be independent of runups. Since our empirical tests strongly reject a zero linear slope in the projection of markups on run ups, and confirm that the projection is nonlinear, this alternative market inefficiency hypothesis is also rejected.

The full paper is available for download here.

 

Both comments and trackbacks are currently closed.