How Investment Banks’ Disagreement over Valuation Contributes to the Winner’s Curse

The well-developed theory of the winner’s curse can potentially explain the poor performance of mergers and acquisitions (Roll, 1986). A key reason for the curse is the uncertainty concerning a deal’s value (e.g., Capen, Clapp, and Campbell, 1971; Bazerman and Samuelson, 1983; Varaiya, 1988).  The greater the disagreement over the target’s value, the more likely that a winning bid will fail to account for the uncertainty and lead to overpayment. The empirical relevance of the winner’s curse is central to takeover efficiency, because it teaches that the bidder who overpays the most, instead of the bidder who can create the highest synergy, will acquire the target. However, to date, such empirical evidence is limited.

A major challenge for testing the winner’s curse is to accurately measure divergence of opinion about target valuation among potential bidders. Using a unique setting where target firms hire multiple investment banks as advisers, we construct a novel measure of target valuation uncertainty based on investment banks’ disagreement. Our valuation disagreement measure overcomes several limitations shared by traditional measures, such as stock return volatility and analyst forecast dispersion. Specifically, investment bank valuation directly captures target firm value rather than a single accounting or stock market metric. Moreover, in addition to public information, investment banks’ valuation disagreement incorporates confidential information on the merging parties.  Furthermore, our measure captures investment banks’ valuation disagreement on both target firms’ stand-alone value and merger synergies.

We obtain the M&A data from the Securities Data Corporation (SDC) database from 1994 to 2020. Our final sample consists of 462 deals in which target firms hire two or more investment banks to provide valuation analyses. We manually collect valuation information from the fairness opinion sections of merger filings to construct our measure. We find a substantial disagreement on the deal value among banks even though these banks have the same access to the confidential information provided by the target management. For example, the average investment banks’ disagreement on the maximum value of the target firm is about 15 percent relative to the target firm’s pre-merger stock price.

We test two major hypotheses. First, with the winner’s curse, bidders fail to account for the high uncertainty about target value and thus overpay. Therefore, the winner’s curse hypothesis predicts that higher valuation disagreement leads to a higher acquisition premium paid by bidders. Second, a high acquisition premium in the presence of high valuation disagreement is likely caused by the winner’s curse and therefore leads to poor bidder returns. In contrast, a high acquisition premium without the presence of high valuation disagreement is unlikely to be driven by the winners’ curse or to cause poor bidder returns. Therefore, the second hypothesis predicts that, when there is high valuation disagreement, bidders have lower returns when they pay higher premiums to acquire target firms. We examine both bidders’ announcement returns and the long-term returns in the post-merger period.

We find that, consistent with the winner’s curse hypothesis, acquirers pay a significantly higher acquisition premium when there is more valuation disagreement. For example, a one standard deviation increase in the valuation disagreement about median target value is associated with a six percentage-point increase in acquisition premium. These results are robust after we control for commonly used investor disagreement measures in the regression analysis. Next, we investigate the interactive effect of valuation disagreement and acquisition premium on bidder performance. We first measure bidder performance by bidder announcement returns and find that, consistent with the winner’s curse, more valuation disagreement leads to a more negative relation between acquisition premium and bidder announcement return. When valuation disagreement rises to a very high level of two standard deviations above its mean, a one standard deviation increase in acquisition premium is associated with a decrease of bidder return of 1.4 percent, which is large given the average bidder announcement return of -0.2 percent. We also find similar results using the bidder’s long-term return in the one-year to three-year windows after merger completion.

While the above results provide evidence of the winner’s curse hypothesis and bidders’ overpayment, they do not necessarily suggest resource misallocation. Specifically, a bidder’s poor performance may reflect the relative gain between the target and the bidder (a division of merger synergy) rather than the level of merger synergy. We therefore follow the literature and measure merger synergies, using combined bidder and target announcement returns, and find that higher valuation disagreement also leads to a more negative relation between acquisition premium and merger synergies. These findings indicate that, when valuation disagreement is high, winners who pay a greater acquisition premium are actually the ones who create lower merger synergies. Therefore, our results suggest that the winner’s curse distorts resource allocation in M&As.

We conduct several tests using valuation disagreement to further analyze the mechanisms related to the winner’s curse. First, we investigate how valuation disagreement influences the dynamics of the private process before deal announcement. Literature suggests that the target firms have incentives to control the number of bidders (Hansen, 2001; Brown, Liu, and Mulherin, 2021). Under the winner’s curse hypothesis, valuation uncertainty increases the likelihood that some bidders make  high valuation errors. Therefore, a target firm with higher valuation uncertainty only needs to approach a small number of bidders to obtain a high acquisition premium.  We find that, consistent with this prediction, target firms with higher valuation disagreement, despite receiving a higher acquisition premium, contact fewer bidders during the private sale process and have fewer bidders enter into a confidential agreement. Second, we study the relationship between investment bank disagreement and deal completion. Under the winner’s curse hypothesis, valuation disagreement leads to  overpayment by the winning bidder, which in turn will increase the likelihood of target-shareholder approval. Consistent with this prediction, we find that deals are more likely to be completed when valuation disagreement is higher.

Our paper makes several contributions to the literature. First, we provide strong evidence that supports the winner’s curse hypothesis in the takeover market, using a direct measure of target valuation uncertainty. Second, we construct a novel measure of valuation disagreement in the M&A setting and add to the literature on the divergence of opinion. Third, we extend the literature on takeover efficiency by showing that the winner’s curse also contributes to resource misallocation in the takeover market. Finally, our paper sheds light on the debate about the informativeness of fairness opinions and suggests that disagreement among investment banks over valuation can also provide useful information to merger participants.


Bazerman, M.H. and Samuelson, W.F., 1983. I won the auction but don’t want the prize. Journal of Conflict Resolution27(4), pp.618-634.

Brown, W.O., Liu, T. and Mulherin, J.H., 2021. The development of the takeover auction process: The evolution of property rights in the modern wild west. Available at SSRN 3112724.

Capen, E.C., Clapp, R.V. and Campbell, W.M., 1971. Competitive bidding in high-risk situations. Journal of Petroleum Technology23(06), pp.641-653.

Hansen, R.G., 2001. Auctions of companies. Economic Inquiry39(1), pp.30-43.

Roll, R., 1986. The hubris hypothesis of corporate takeovers. Journal of Business, pp.197-216.

Varaiya, N.P., 1988. The ‘winner’s curse’ hypothesis and corporate takeovers. Managerial and Decision Economics9(3), pp.209-219.

 This post comes to us from professors Tingting Liu at Iowa State University, Tao Shu at the Chinese University of Hong Kong, Shenzhen, and Jasmine Wang at the University of Virginia. It is based on their recent paper, “Divergence of Investment Bank Valuation and the Winner’s Curse in Takeovers,” available here.