Freeze-Out Mergers

Elif Dalkir is Associate Professor of Economics at the University of New Brunswick; Mehmet Dalkir is Associate Professor of Economics at the University of New Brunswick; and Doron Levit is Assistant Professor of Finance at The Wharton School of the University of Pennsylvania. This post is based on their recent article, forthcoming in the Review of Financial Studies.

Do freeze-out mergers mitigate the free-rider problem of corporate takeovers? We revisit this fundamental question in our article Freeze-Out Mergers, which is forthcoming in the Review of Financial Studies.

The ability of the market for corporate control to efficiently allocate resources is much debated. A seminal paper by Grossman and Hart (1980) argued that there is a free-rider problem that prevents acquirers from successfully taking over companies that are widely held. The crucial assumption in this argument is that target shareholders do not view themselves as pivotal in the success of the takeover. Therefore, each shareholder refuses to tender his share whenever he expects the post-takeover value to be higher than what is being offered. If all shareholders behave that way, then a value-increasing acquirer cannot convince shareholders to tender their shares and at the same time make a profit on the purchased shares. Without any private benefits of control, free-riding precludes efficient takeovers, leading to an inefficient market for corporate control.

As a remedy to the free-rider problem, Grossman and Hart argued in favor of diluting non-tendering shareholders, thereby excluding them from any gains due to the takeover. Subsequently, the academic literature has identified several mechanisms that could potentially mitigate the free-rider problem, most notably, freeze-out mergers. A freeze-out merger is a transaction in which the controlling shareholder buys out the shares of the minority, delists the corporation, and then takes it private. The rules that govern freeze-out mergers vary considerably across jurisdictions, but they all limit the ability of shareholders to retain their shares in the target company as a minority after the acquirer gains control.

How can a freeze-out merger solve the free-rider problem? The idea is as follows. If shareholders believe that the freeze-out threshold will be met (i.e., the fraction of tendered shares will exceed the ownership fraction that is required for freezing out minority shareholders), then each shareholder is indifferent between tendering and retaining his shares, irrespective of the offer price. Therefore, there always exists an equilibrium in which all shareholders tender even at a price significantly lower than the post-takeover value. In this equilibrium, a value-increasing acquirer (“raider”) can profitably take over the target firm, thereby restoring economic efficiency. If shareholders are infinitesimal, this argument holds for any freeze-out threshold strictly smaller than 100%. Building on this line of arguments, it has been argued that freeze-out mergers are the perfect panacea for the free-rider problem (e.g., Amihud, Kahan, and Sundaram 2004).

In this article, we challenge this proposition. We show that unless the most extreme freeze-out clause is considered (i.e., a 50% freeze-out threshold; the simple majority), freeze-out mergers do not solve the free-rider problem as long as shareholders can be pivotal for the takeover, even if the probability of being pivotal is arbitrarily small.

Following Bagnoli and Lipman (1988), we analyze a tender offer model with finitely many shareholders. Such a model allows each individual shareholder to have an impact on the success of the tender offer. Bagnoli and Lipman (1988) showed that although the raider can have a strictly positive expected profit in equilibrium, this profit converges to zero as the number of shareholders gets arbitrarily large, in line with Grossman and Hart’s (1980) predictions. Drawing from the existing literature, one might conclude that this argument is invalid when freeze-out mergers are considered. To examine this proposition, we add freeze-out mergers to their setup. As one might expect, the ability to freeze out shareholders increases the raider’s expected profit, and this profit is higher for lower (i.e., stricter) freeze-out thresholds. Furthermore, the expected profit decreases as the firm becomes more widely held. In this respect, our analysis provides novel predictions on the effect of freeze-outs on the success probability of a takeover, the expected takeover premium, and the raider’s profit.

However, our main result shows that as the number of shareholders gets arbitrarily large, the raider’s expected profit in equilibrium converges to zero for any freeze-out clause with an ownership threshold that is strictly above simple majority. That is, freeze-out mergers do not solve the free-rider problem in widely held firms.

The intuition behind our result is the following. As long as the freeze-out threshold is strictly greater than 50%, shareholders can never rule out in equilibrium the possibility that the takeover will succeed but the number of tendered shares will be insufficient to trigger the freeze-out clause. In those events, minority shareholders receive the post-takeover value. Therefore, conditional on the success of the takeover, non-tendering shareholders can expect to get on average strictly more than the offer price. If the takeover is expected to succeed as the number of shareholders gets arbitrarily large, each shareholder believes that the impact of his individual decision is negligible, and the incentives to free-ride prevail just as in the Grossman and Hart (1980) model.

This argument, however, is incomplete. In equilibrium, shareholders are more likely to tender their shares when the offer price is higher. Therefore, if the price is sufficiently high but still below the post-takeover value, the probability that the number of tendered shares exceeds the freeze-out threshold must converge to one as the number of shareholders gets arbitrarily large. Seemingly, shareholders should have no incentive to free-ride, and without free-riding, the raider could take over the firm and make a strictly positive profit. But this argument is false. Indeed, shareholders are always indifferent between tendering and not tendering whenever the freeze-out clause is triggered; they will get the same offer price either way. Because of their indifference, each shareholder conditions his decision to tender only on events in which the freeze-out clause is not triggered. As was explained above, conditional on those events, shareholders are strictly better off not tendering their shares. Therefore, even if shareholders were to believe that the freeze-out clause will almost surely be triggered, their incentives to free-ride would not be affected. The raider can induce them to tender their shares only if the offer price converges to the post-takeover value, which would leave the raider with no profit in the limit. The free-rider problem remains unsolved.

Finally, if the freeze-out threshold is exactly 50% (or if the tender offer requires the number of tendered shares to exceed the freeze-out threshold), shareholders can never free-ride since by design the tender offer cannot succeed without triggering the freeze-out clause. However, we show that this result is not robust to small perturbations of the model, for example, the introduction of a tendering cost. In general, the free-rider problem remains unsolved even in those special cases.

The complete article is available for download here.

Both comments and trackbacks are currently closed.