Freeze-outs have been subject to different standards of judicial review in Delaware since 2001, when the Delaware Chancery Court, in In re Siliconix Inc. Shareholders Litigation, Civ. A. No. 18700, 2001 WL 716787 (Del. June 19, 2001), introduced a distinction based on the form in which the transaction is executed. In particular, in Siliconix, the chancery court held that, unlike freeze-outs executed as a merger (which have been subject to “entire fairness review” since 1952), freeze-outs executed as a tender offer were exempted from that standard of review. According to the court, tender offers do not warrant entire fairness because, in these transactions, in contrast to a merger, minority shareholders are protected by the decision itself of tendering or not tendering. Moreover, one month after Siliconix, in Glassman v. Unocal Exploration Corporation, 777 A.2d. 242 (Del 2001), the Delaware Supreme Court held that a short-form merger is also excluded from entire fairness review. As a result of these two decisions, a controlling shareholder was allowed to completely avoid entire fairness by acquiring the remaining shares from minority shareholders through a tender offer followed by a short-form merger.
This regulatory dichotomy created by Siliconix, in turn, gave rise to a tension in the literature over the treatment that freeze-outs should receive. While some commentators argue for regulatory convergence in the standards of judicial review by subjecting tender offers to entire fairness, others defend the regime established by Siliconix, and still others propose an alternative variation of convergence based on how the negotiation process took place. This approach, in fact, was recommended in dicta by Vice-Chancellor Strine in In re Cox Communications systems, Inc. shareholders litigation, Civ. A. No. 613-N, 2005 WL 2001310 (Del. Ch. June 6, 2005), and some of its elements were subsequently adopted in In re CNX Gas Corporation Shareholders Litigation, Civ. A. No. 5733-VCL, 2010 WL 2291842 (Del. Ch. May 25, 2010).
Despite the arguments against Siliconix even in subsequent chancery court opinions, there is limited empirical evidence on the effect of that decision and, therefore, on the practical importance that convergence or divergence in standards of judicial review has for minority shareholders in a freeze-out. In particular, with a sample of post-Siliconix freeze-outs between 2001 and 2005, Guhan Subramaninan found that minority shareholders receive lower cumulative abnormal returns (CARs), on average, when a freeze-out is executed as a tender offer than when it is executed as a statutory merger. Based on this, Subramanian holds that the difference in outcomes might be the result of the dissimilar protections for minority shareholders in tender offers and mergers after Siliconix, and, consequently, he argues for doctrinal convergence. That article, however, does not formally examine whether Siliconix generated a structural change in relative CARs in these transactional forms and, therefore, whether the differences in outcomes are actually attributable to dissimilar standards of review. The purpose of this work is, therefore, to fill this gap in the literature.
To examine the effect of Siliconix, this work uses a difference-in-differences approach in which the treatment group is the set of tender offer freeze-outs of Delaware targets announced and completed between January of 1996 (after the Delaware Supreme Court decision in Solomon v. Pathe Communications Corp., 672 A.2d 35 (Del. 1996)) and June of 2005 (before the Delaware Chancery Court decision in In re Cox Communications systems, Inc. shareholders litigation, Civ. A. No. 613-N, 2005 WL 2001310 (Del. Ch. June 6, 2005)), and the control group is the set of merger freeze-outs announced in the same period. Like in Subramanian’s article, the outcome variable is the CARs that minority shareholders receive in the transaction, defined as the daily return for the target shares relative to the CRSP value-weighted index.
This work has at least one policy implication. If the results showed that Siliconix was indeed a significant factor in the creation of differences in CARs between tender offers and statutory mergers, the case for regulatory convergence made by Subramanian on empirical grounds would be reinforced. For the same reason, the move toward unification marked by CNX may be interpreted as justified on the welfare of minority shareholders. If, in contrast, the counterfactual were true (namely, that Siliconix is not a cause of the gap in relative CARs found by Subramanian), then there would be no clear justification for regulatory convergence specifically on the basis of differences in transactional outcomes. This implies, in turn, that a policy like the one adopted in CNX might create a friction in the market without practical offsetting benefits. The results presented in this work, however, are generally consistent with the hypothesis that Siliconix had at least some negative effect on tender offer CARs, and, therefore, those results also support the limitations to this decision adopted in CNX.
The full paper is available for download here.