The Perils of Governance by Stockholder Agreements

The Delaware State Bar Association (“DSBA”) and the Council of the DSBA’s Corporation Law Section recently approved a proposal (“the proposal”) to make an amendment (“the proposed amendment” or “the proposed legislation”) to Section 122 of the Delaware General Corporation Law (“DGCL”). The proposed amendment would permit expansive use of stockholder agreements—agreements between a board and a stockholder (or group of stockholders)—to opt out of the governance arrangements set by the company’s charter.

The proposal was sent to the Delaware legislature, and it is contemplated that the legislature would adopt it in the current session of the legislature, which ends in several weeks. As explained below, however, the proposal raises serious concerns. Adopting it would have a wide range of detrimental consequences for institutional investors, and other public investors, as well as for the quality of governance in Delaware companies.

In a Hurry

The proposal’s supporters seem to be in a hurry to have the amendment adopted. As explained in the Synopsis accompanying the text of the proposed legislation (“the Synopsis”), the proposal is motivated by and intended to reverse the Chancery Court’s decision in Moelis (discussed on the forum here), which was issued earlier this year.

(A Morris Nichols post, available on the Forum here, describes the proposed amendment and the motivation for it. A post by Professors Rauterberg and Sanga, available here, is willing to accept the desirability of overturning Moelis, which I do not do, but opposes the proposed amendment for not doing so in a well-targeted fashion.)

There does not seem to be any urgency, or at least none noted in the Synopsis, for passing a legislative change to reverse the recent decision in Moelis so quickly after the ink has dried. This is all the more the case given that the Moelis decision is now on appeal and the Delaware Supreme Court is expected to decide the appeal this year.

If there were a meaningful chance that Moelis would be reversed on appeal, why get the legislature involved now? It seems that the amendment’s supporters expect, as I do, that the Supreme Court would likely affirm the Moelis decision and support its reasoning and conclusions. With such expectations, the proposal’s supporters might prefer not to wait for the Supreme Court to speak on the issue and rush instead to obtain a legislative reversal.

Changing How Governance Arrangements Are Set

While a substantial part of Delaware law is judge-made, the Delaware legislature adopts with some frequency amendments to the Delaware code. However, these amendments are often of limited consequence. (This is the case, for example, with other changes included in the package sent to the Delaware legislature.) Not so in this case. If adopted, the amendment would have a broad effect on how governance arrangements can be made and changed.

The Moelis case considered an agreement between the company’s board and a shareholder who was also the company’s founder, CEO, and chair. The agreement set up governance arrangements fundamentally different from those that, given the company’s charter and bylaws, would prevail if the agreement had not been made. Among other things, the agreement granted the founder various consequential powers not provided by the company’s charter and bylaws, including both a veto power over any decision by the board and power to control multiple board-level decisions.

Based on an extensive and careful analysis of Delaware case law spanning several decades, the Chancery Court concluded that the stockholder agreement violated the basic statutory provision of the DGCL (Section 141(a)) that, absent a contrary provision in a company’s charter, reserves certain decision-making to the board of directors. As the court explained, this statutory provision and the governance philosophy underlying it have shaped and been reflected in Delaware court decisions over many years and in many settings.

Supporters of the proposed amendment do not seek to permit just certain specified types of governance-changing stockholder agreements such as the one considered in Moelis. Instead, the amendment seeks to protect a vast universe of stockholder agreements from being declared facially invalid. In particular, the amendment would permit any corporation to enter into a contract with one or more stockholders that would:

(a) “restrict or prohibit future corporate actions specified in the contract (even if those actions require approval by the board of directors under [the statute]);”

(b) “require the approval or consent of one or more persons or bodies (including the board of directors or one or more current or future directors, stockholders or beneficial owners of stock) before the corporation may take actions specified in the contract;” and

(c) “covenant that the corporation or one or more persons or bodies (including the board of directors or one or more current or future directors, stockholders or beneficial owners of stock) will take, or refrain from taking, future actions specified in the contract.”

Bring Law in Line with Practice?

What reasons are offered by supporters? First, as a memo by one prominent Delaware law firm explains, a key motivation is that the Moelis decision and its corollaries are “not necessarily in line with market practice” and the amendment is “designed to bring existing law in line with such practice.” But corporate law should not necessarily be in line with what private parties choose to do in the marketplace.

In some situations, private players might have incentives to take actions that would adversely affect corporate and stockholder value rather than serve them. In such situations, the law should constrain rather than fall in line with the choices of market participants.

Furthermore, it is far from clear that market practices introducing stockholder agreements are as broad and common as supporters seem to believe. The leading empirical study on the subject by Rauterberg, which the Moelis decision cites, documents only that stockholder agreements are common in companies that recently went public, especially companies in which certain pre-IPO investors have holdings. This study, however, does not tell us anything about whether and how often governance-setting stockholder agreements are present in corporations that have been public for more than several years.

As we will see, in such ongoing company situations, allowing governance via stockholder agreements would likely be especially problematic. A substantial majority of Delaware companies, both in number and in aggregate market capitalization, went public more than just a few years ago. Thus, even if the proposal’s supporters hypothetically had a valid point with respect to post-IPO situations, which I question below, they should have paused before putting forward a proposal that would broadly affect all public companies, including the many large and important companies in which shares are mostly held by institutional investors, with few (if any) pre-IPO shareholders.

[The importance of preventing considerations that might be relevant for new companies coming out of an IPO from having long-lasting effects on companies that are years or decades after their IPO is stressed in my article with Kobi Kastiel on The Untenable Case for Perpetual Dual-Class Stock.]

Necessary to Facilitate Private Ordering?

Supporters of the proposed amendment might be motivated by a desire to facilitate private ordering by enabling private parties to opt into a governance arrangement most fitting for their situation. But for every case in which Delaware law permits parties to opt out of default arrangements, Delaware law also establishes procedures that enable opting out without using stockholder agreements. In particular, any permissible opt-out from Delaware law may already be accomplished via a charter provision. The Moelis decision did not conclude that the governance arrangement set up by the invalidated stockholder agreement was completely impermissible but only that this arrangement had to be adopted through a charter provision.

Thus, the proposed legislation is not necessary for making any permissible opt-out possible, but rather it would provide an alternative avenue for effecting it. The choice is thus not whether to permit such opt-outs, but whether there is a good reason to introduce such an alternative avenue. Supporters of private ordering should not favor private ordering by any means; they should only favor private ordering through appropriate processes. The question then is not whether private ordering is desirable. The critical question is whether, for any given opt-out that is now permissible via a charter provision, it would be desirable to add an additional avenue of adoption via stockholder agreements.

Private Ordering without Adequate Safeguards

To be sure, as Rauterberg observed in his article, stockholder agreements specifying governance arrangements have been motivated by certain perceived advantages: Whereas amending the corporate charter requires advance notice to stockholders and an approval vote by a majority of the outstanding shares, adopting a stockholder agreement and amending it does not require either advance notice to stockholders or an approval vote by a majority of outstanding shares. Although these “advantages” could well be desirable to the private parties introducing governance-setting stockholder agreements, they hardly imply that allowing opt-outs via stockholder agreement is desirable.

To the contrary, the “flexibility” of avoiding the need for advance notice and approval by a stockholder majority comes at the expense of removing important safeguards. There are good reasons for the long-standing corporate law rules that require advance notice and approval by a majority of outstanding shares for changes in the corporate charter. And these good reasons imply that it is undesirable to enable private parties to practically circumvent these safeguards. For any given opt-out corporations are allowed to effect via a charter change, I do not see any good reason, and supporters of the proposed legislation have not advanced such a reason, for facilitating a way around the notice and stockholder approval requirements.

The problem with allowing the alternative avenue of opting out via stockholder agreements can be stated using the language of externalities. The requirement that a charter amendment be approved by a majority of outstanding shares makes it unlikely that a governance arrangement making shareholders generally worse off would be adopted in this way. By contrast, if shareholder majority approval were no longer required, and governance arrangements introduced by an agreement between the board and a shareholder holding a minority position were to be generally permitted, then the door would open for governance-setting agreements adopted by incumbent directors and a stockholder with a minority position that would impose significant externalities on stockholder value.

Thus, the proposed legislation’s provision of a general way around the notice and shareholder approval requirements for a charter amendment should not be viewed as a plus. It is a considerable minus, and it would eliminate a highly beneficial safeguard designed to preclude governance changes that would make most stockholders worse off. For this reason, institutional investors and their beneficial investors would be adversely affected by the legislation and would do well to oppose it strongly.

[Some of my earlier work expressed the concern that boards’ veto power over proposals for charter amendments—i.e., shareholders’ lack of power to initiate a charter amendment and bring it to a vote—could well prevent some desirable charter amendment from taking place. (See The Case for Increasing Shareholder Power and Letting Shareholders Set the Rules.) But this work fully accepted that majority stockholder approval should be a necessary condition for changing governance via a charter amendment.]

Wide-ranging Consequences

In considering the severity of the above concerns, it is important to recognize the wide range of stockholder agreements in a wide array of settings that would likely be brought about by the proposed legislation, and which have received little or insufficient attention in the Synopsis.

Under the proposed legislation, a stockholder agreement would require the company to receive consideration, but the board of directors would be allowed to accept a “minimum consideration as determined [by the board and] may include inducing stockholders or beneficial owners of stock to take, or refrain from taking, one or more actions.” This broad authorization would enable the adoption of stockholder agreements in a wide range of common situations. These include scenarios in which a board of directors makes an agreement with a private party and:

(1) The board views the party’s investing in the company as beneficial and thus seeks to induce such investment;

(2) The board views the company’s doing an IPO or some other transaction as beneficial and thus seeks to induce the private party to enable or cooperate with the IPO or transaction;

(3) The board views the company’s acquiring a seller’s assets in return for company shares as beneficial and therefore seeks to induce seller stockholders to agree to the sale;

(4) The board views ending a proxy fight or engagement by an activist hedge fund as beneficial and therefore seeks to induce the activist hedge fund to agree to end the engagement; or

(5) The board views a certain outcome of a pending stockholder vote as beneficial and therefore seeks to induce a certain stockholder to vote for this outcome.

In the wide range of situations in which such circumstances exist, the proposed legislation would enable governance-altering stockholder agreements that, under long-standing case law, have been viewed as facially invalid. Under the proposed legislation, for example, in such circumstances it clearly would not be per se impermissible for the board to considerably limit its decision-making power by granting a stockholder (including one holding a small minority of stock):

(i) an effective power to push the company into a sale if the stockholder so chooses by, say, granting the stockholder veto power over the adoption of any poison pill;

(ii) a veto power over entering any merger agreement (even one favored by the directors and a majority of the shareholders);

(iii) an effective power to push out the CEO when the CEO’s current employment agreement ends by, for instance, granting the stockholder a veto power over any subsequent employment agreement with the CEO;

(iv) a critical influence on the choice of any successor CEO by, say, granting the stockholder a veto power over the appointment of a CEO regardless of the breadth of support a CEO candidate enjoys on the board; or

(v) a long-standing or even indefinite power to shape the composition of the board by granting the shareholder veto power over the inclusion of any individual on the company‘s directors slate.

Under the proposed legislation, any such stockholder agreement could no longer be invalidated by a court as facially impermissible. Such an invalidation could take place only, if at all, following a judicial “facts and circumstances inquiry,” which I will discuss below.

A corporate world in which such agreements are permissible would operate very differently from what we have today. Many key dimensions of governance could well be affected, and some of the effects could well be detrimental. These potential effects warrant a careful and substantial analysis, which does not seem to have been done by supporters of the proposal, before adopting a legislative change.

Fiduciary Duties to the Rescue?

The Synopsis seems to suggest that any problems resulting from the expansive permissibility of stockholder agreements would be addressed by fiduciary duties. The Synopsis stresses that the amendment would “not relieve any directors, officers or stockholders of any fiduciary duties they owe to the corporation or its stockholders, including with respect to deciding to cause the corporation to enter into a [stockholder agreement] and with respect to deciding whether to cause the corporation to perform, or to breach, the [stockholder agreement.]” According to the Synopsis, the amendment would eliminate (i) relief with respect to stockholder agreements that would be based on “grounds that the contract is facially invalid,” but not (ii) relief that would be based on “equitable principles relating to fiduciary duties.”

My concern is that in situations such as cases (1)-(5) above, in which a business purpose is not patently absent, the statutory instruction to avoid blanket invalidation would lead courts to commonly accept adopted stockholder agreements as long as the board complied with the process requirements for reaching an informed and considered decision. Thus, the amendment would often preclude invalidation at the outset (i.e., at the time of entering the contract) of stockholder agreements that would be regarded as facially invalid under existing statutory language and case law.

But the Synopsis stresses that judicial intervention grounded in fiduciary duties might take place ex post. Depending on the facts and circumstances ex post, a court might require the board not to perform an agreement. For example, suppose that an initially valid stockholder agreement granted a stockholder the power to veto action of type X, but when the time comes, given the facts and circumstances, it is likely or at least possible that the company would be served best by doing X. In this case, a fiduciary duty analysis might lead the court to order the board to breach or at least to refuse to grant specific performance if the board chooses to breach. Even in this case, however, the stockholder agreement would have an effect.

When a stockholder agreement was not invalid as violating fiduciary duties at the outset, the amendment specifies that “the corporation shall be subject to the remedies available under the law governing the contract, including for any failure to perform or comply with its agreement under such contract.” Thus, even if fiduciary duties would preclude specific performance, a breach would entitle the stockholder that is party to the stockholder agreement to obtain damages, and the prospect of paying damage (which would not exist but for the stockholder agreement) might make doing X not worthwhile. In any event, years of litigation and many cases would be necessary to clarify the circumstances and facts that would warrant judicial intervention with respect to stockholder agreements that would become facially valid under the proposed legislation. The legislation thus would hardly introduce certainty to this area of the law.

Conclusion

The above discussion raises serious concerns about the proposed amendment. At a minimum, it is critical not to proceed to legislation without an adequate consideration of the wide array of potential effects of the proposed amendment.

In this process, it would be important to provide sufficient time and opportunities to get input not only from corporate lawyers but also from institutional investors, scholars, and all others who have an interest in effective stockholder protections and good corporate governance in Delaware companies.