chevron-down Created with Sketch Beta.

The Business Lawyer

Summer 2022 | Volume 77, Issue 3

The Three Faces of Control

Ann Meredith Lipton


  • Controlling shareholders are subject to distinct legal obligations under Delaware law, and thus Delaware courts are routinely called upon to distinguish “controlling shareholders” from other corporate actors. That is an easy enough task when a person or entity has more than 50 percent of the corporate vote, but when a putative controller has less than 50 percent of the vote—and is nonetheless alleged to exercise control over corporate operations via other means—the law is shot through with inconsistency.

    What is needed is a contextual approach that recognizes that the meaning of control may vary depending on the purpose of the inquiry. Under Delaware doctrine, the controlling shareholder label subjects that entity xv to unique legal treatment along three distinct dimensions. First, controlling shareholders—unlike minority shareholders—have fiduciary duties to the corporation. Second, interested transactions with controlling shareholders—unlike interested transactions with other fiduciaries— are subject to a unique cleansing regime in order to win business judgment deference from reviewing courts. Third, when certain transactions involving sales of control are challenged in court, they may be treated as direct rather than derivative actions, even when similar transactions that do not involve control sales would be treated as purely derivative.

    By teasing out these three aspects of the legal framework and analyzing them separately, courts can more closely attend to the reasons why control carries special significance, and ultimately develop a more rational and consistent set of definitions. Most critically, courts may properly designate someone a controlling shareholder for some purposes, but not others.
The Three Faces of Control

Jump to:

I. Introduction

Delaware has a controlling shareholder problem.

For many years, controlling shareholders have been subject to a distinct legal regime, whereby they are deemed to owe special duties to the corporation and their actions are subject to special scrutiny by the Delaware courts. But this alternative legal universe necessitates that controlling shareholders be consistently identified in order to distinguish them from other corporate actors. The usual formulation is that a controlling shareholder is one who “(1) ‘owns more than 50% of the company’s voting power’ or (2) ‘owns less than 50% of the voting power of the corporation but exercises control over the business affairs of the corporation.’” This definition makes controlling shareholders easy enough to spot in the former scenario, but in the latter, the law is shot through with inconsistency, allowing similar facts to result in divergent outcomes.

“Minority” controllers, with less than 50 percent of the vote, may be deemed so only with respect to a particular transaction in which they “actually dominated and controlled the corporation, its board or the deciding committee,” or they may be deemed to have overall control of corporate operations if they “actually dominated and controlled the majority of the board generally.” Courts must determine “whether the stockholder, as a practical matter, possesses a combination of stock voting power and managerial authority that enables him to control the corporation, if he so wishes.” In so doing, the court may consider any and all sources of influence, such as “(i) relationships with particular directors … , (ii) relationships with key managers or advisors … , (iii) the exercise of contractual rights to channel the corporation into a particular outcome … , and (iv) the existence of commercial relationships that provide the defendant with leverage over the corporation, such as status as a key customer or supplier,” in addition to “ownership of a significant equity stake … , the right to designate directors … , decisional rules in governing documents that enhance the power of a minority stockholder or board-level position, and the ability to exercise outsized influence in the board room, such as through high-status roles like CEO, Chairman, or founder.”

The most notable feature of the test for minority control is its elasticity: It includes so many factors and considerations that it allows for a great variety of actors to be designated as controllers. An additional notable feature of the test for minority control is that it is utterly untethered from its consequence: It does not matter why control is legally relevant in a particular scenario; the test remains the same.

These things are not unrelated. Under Delaware doctrine, a single label—controlling shareholder—carries an enormous amount of legal weight. Once the label is applied to a person or institution, that entity is immediately subject to unique legal treatment along three distinct dimensions. First, controlling shareholders—unlike minority shareholders—have fiduciary duties to the corporation. Second, interested transactions with controlling shareholders—unlike interested transactions with other fiduciaries—are subject to a unique cleansing regime in order to win business judgment deference from reviewing courts. Third, when certain transactions involving sales of control are challenged in court, they may be treated as direct rather than derivative actions, even when similar transactions that do not involve control sales would be treated as purely derivative. Given these multiple functions that the designation of control serves, it is no wonder that courts feel the need for flexibility in applying the label.

As I documented in a previous essay, recent factual and legal developments have given a particular urgency to the problem of identifying those persons or entities that are capable of controlling the corporation from a minority voting position. Today, startup businesses remain private for longer periods of time than in the past, typically conducting multiple rounds of financing from a variety of investors, including venture capital funds, sovereign wealth funds, family offices, and even traditional mutual funds. These different investors are often granted individualized rights, such as designated board seats and the ability to block various corporate actions. The result is that corporate control rights are increasingly allocated in unique and idiosyncratic ways, making a simple “50%” metric inadequate to assess controlling shareholder status. And these complex control arrangements have begun migrating to public companies. Recent initial public offerings have involved dual, triple, and even quadruple class stock structures, often granting several insiders high-vote shares that shift or sunset over time. Even in companies with single-class structures, shareholder agreements may replicate the special rights that have become common among preferred stockholders in private corporations, and disputes involving those companies wind up in the Delaware courts.

On the legal front, just in the past few years, Delaware’s corporate doctrine has evolved to draw even sharper distinctions than existed previously between transactions involving controlling shareholders and those that do not. In general, interested transactions that do not involve controlling shareholders receive business judgment deference from reviewing courts so long as they are blessed by either independent and disinterested directors or the vote of the disinterested shareholders. Those that do involve controlling shareholders, by contrast, must be negotiated and approved in strict compliance with a set of multiple cleansing measures before the same deference will be accorded—a procedural gauntlet that many putative controllers fail. The practical effect is that the presence or absence of a controlling shareholder frequently represents the difference between a shareholder claim that proceeds to discovery and one that is dismissed on the pleadings. Some courts, chafing under these rigid distinctions, have apparently found flexibility in the definition of control itself, allowing claims that smack of self-dealing to proceed by designating minority shareholders as potential controllers. The result is a confusing body of caselaw that leaves the definition of control shifting and uncertain.

Uncertainty, as it turns out, begets uncertainty. The malleability of the controlling shareholder label—coupled with its increasing legal significance—has encouraged a certain creativity on the part of the plaintiffs’ bar when alleging control status meriting discovery and close judicial review. Thus, as of this writing, one lawsuit is pending, and another was recently tried to a verdict, both alleging that Elon Musk, with only a roughly 20 percent stake in Tesla, was a controlling shareholder, which would require heightened judicial scrutiny of both Tesla’s acquisition of a related company and Musk’s pay package as CEO. In another case, plaintiffs claimed that a company’s private equity sponsor operated as a corporate controller due to its contractual influence and relationships with company management, despite holding no equity in the company itself. Sponsors of special purpose acquisition companies (SPACs), with only 20 percent of the equity, have been alleged to be controllers who seek out unfavorable acquisitions to win personal benefits. Each of these claims, and the many more that will likely be forthcoming, tests the coherence of a legal regime that purports to make a single label—controlling shareholder—virtually outcome determinative across a wide range of scenarios.

The core problem here is that the definition of a controlling shareholder has come unmoored from the policies that inspired the divergent legal treatment in the first place.

When control is defined in terms of more than 50 percent of the corporation’s voting power, it may well be appropriate that all legal implications—(1) the imputation of fiduciary duties, (2) the application of enhanced judicial scrutiny, and (3) the assumption of enhanced financial value—follow. But when control is exercised from a minority position, it is far less clear that the full suite of legal consequences should be imposed, especially given the myriad mechanisms by which minority controllers exert their power.

The solution, then, is for courts to apply a more contextual approach, teasing out the three aspects of the legal framework governing control and analyzing them separately. In so doing, courts can more closely attend to the reasons why control carries special significance, and ultimately develop a more rational and consistent set of definitions. Most critically, as this article shows, courts might deem it appropriate to designate someone a controlling shareholder for some purposes, but not others.

II. Control as a Locus of Fiduciary Duty

Ordinarily, individual shareholders owe no fiduciary duties to the corporations in which they invest—they are only the object of such duties—and they are therefore free to act solely to further their own interests. That situation changes, however, for controlling shareholders: Controllers, uniquely, owe the corporation and the minority shareholders a fiduciary duty of loyalty.

The justification for the divergent treatment stems from the definition of a fiduciary: one who has discretionary control over the assets of others. Controlling shareholders own only a portion of the business, but, by definition, exercise control over the entirety; hence, fiduciary obligations are imposed to the extent they dispose of interests that are held in trust for the minority.

Correlatively, controlling shareholders are not considered fiduciaries with respect to all actions they take. “[L]ike other stockholders they may vote their stock, and take other actions with respect to the entity, in their own self-interest free of fiduciary strictures, so long as they do not employ the corporate machinery itself.” Thus, Delaware recognizes that controlling shareholders are free to cast their votes on corporate matters in their own self-interest, including director selection, whether to accept a merger offer, and the terms on which they will sell—or refuse to sell—their shares. In these scenarios, the controlling shareholder is dealing with assets that are indisputably his own; it is only when there is a threat, or a perceived threat, that the controlling shareholder’s actions drag along the minority that the fiduciary duties attach.

With these premises in hand, courts can identify the situations where entities who do not hold 50 percent of the corporation’s voting power should be deemed controllers for the specific purpose of imposing fiduciary duties: It should occur when those entities exercise discretionary control over assets that properly belong to the remaining shareholders.

This rule seems obvious—after all, it is nearly identical to the current definition of a minority controller—but starting from first principles helps provide guidance for certain types of difficult cases. For example, if control over another’s assets is the hallmark of a fiduciary, and if it is the controlling shareholder’s dominion over corporate assets that gives rise to fiduciary duties on its part, then it follows that no particular level of stock ownership should be necessary before one can be deemed a controller. Indeed, there is no reason that stock ownership, specifically, should be required at all. This is especially so because, in a world where control rights are often allocated not through equity interests directly, but through shareholder agreements, the distinction between control emanating from stock holdings and control emanating from contractual rights is difficult to parse.

That foundation sheds light on another knotty problem: when the exercise of contractual rights should be deemed to give rise to fiduciary duties on the part of the rights holder. A frequent maxim is that a contractual counterparty who merely exercises his rights—which may prompt a reaction by the corporation—is not a controller or a fiduciary with respect to the corporate action taken; however, the counterparty may be deemed to be a “controlling shareholder” if the contractual rights “are used to induce or to coerce the board of directors to approve (or refrain from approving) certain actions.” So far, Chancery courts have taken an “I know it when I see it” approach to drawing this distinction; in Superior Vision Services, Inc. v. ReliaStar Life Insurance Co., for example, the court held that a large investor was not a controller merely because it exercised a blocking right contained in a shareholder agreement to prevent a dividend payment, but, in Skye Mineral Investors, LLC v. DXS Capital (U.S.) Ltd., the court found that certain LLC members became controllers—with concomitant fiduciary duties—when they exercised contractual rights contained in the LLC agreement to block the LLC from obtaining new financing.

At least in some situations, though, it may be helpful to frame the inquiry in terms of distinguishing between one who exercises control with respect to her own assets versus one who exercises control over the assets belonging to another. A party who contracts for particular rights with respect to the corporation—such as a right to block financing or to approve certain corporate actions—does so for a purpose that was likely understood at the time of contracting. A lender, for example, might want to ensure repayment, while an equity investor might want to ensure its stake will not be diluted. When the contract right is exercised to further that purpose, the holder might be said to be exerting control over assets that, in a real sense, belong to her.

But suppose the contract right is used for a wholly different purpose, such as to choke off the corporation’s access to new capital in order to allow the contract holder to claim its assets at bargain prices. That is not the right that was bargained and paid for; it is not the right that the contract holder “owns,” and by using its contract right in this way, the holder may be viewed as controlling assets that belong to others.

The principle might be usefully compared to the doctrine of unconstitutional conditions, which, in some circumstances, treats government action as coercive (and thus prohibited) when a benefit is conditioned on the waiver of a constitutional right. In that context, the permissibility of the condition often depends on its germaneness to the purposes for which the benefits are granted or withheld. Thus, in Nollan v. California Coastal Commission, the Supreme Court held that California could not condition a land use permit on the grant of a public easement, because the easement was unrelated to the reasons why a permit would otherwise be denied. A similar analysis might apply to some exercises (or threatened exercises) of contractual rights to induce board action: When the action sought is unconnected to the purposes for which the contractual right was procured, the right’s use becomes coercive, and thus a mechanism of control that, while not prohibited, potentially subjects the contract holder to fiduciary duties.

Vice Chancellor Glasscock recently deployed exactly this logic when he held that a shareholder vote in Sciabacucchi v. Liberty Broadband Corp. had been coerced. There, shareholders approved a new equity issuance to an existing blockholder who had ties to several board members. Vice Chancellor Glasscock concluded that the shareholder vote did not ratify the transaction because certain unrelated, beneficial corporate acquisitions were conditioned on shareholder approval of the sale. The only way the shareholders could get the benefit of the favorable acquisitions was by approving an unrelated deal with an insider, suggesting that the shareholders’ vote to approve the deal was not fully voluntary. As the Vice Chancellor put it, “the insider financing was not integral to, but was extrinsic to, the Acquisitions,” and it was that separability—the lack of germaneness—that potentially “strong-arm[ed] a favorable vote.” That same germaneness inquiry—whether the contract right was exercised to further the purposes for which it was originally granted—can inform when the use of contract rights crosses over into coercion of a corporate board, with attendant legal responsibilities.

This is not to say that it will always be easy to identify the purpose behind a contractual provision; in many cases it may not be. But framing the issue in terms of those rights that the contract holder has “purchased”—and thus may assert selfishly—can structure the inquiry into what counts as coercive conduct, while staying true to the rationale for imposing fiduciary duties in the first instance. Moreover, courts are frequently called upon to adjudicate the intent behind various contract terms; if anything, the test proposed here is not unlike the one used to determine whether there has been a violation of the covenant of good faith and fair dealing, where a court must “assess the parties’ reasonable expectations at the time of contracting.” When a contract right is used in a manner that falls outside those expectations, the contract holder is exerting control over assets that it does not own.

III. Control as a Basis for Heightened Scrutiny

As described above, controlling shareholders are treated differently from other shareholders in that they—uniquely—have fiduciary duties to the corporation. They are also treated differently from other fiduciaries in that their interested transactions with the corporation are viewed with more suspicion by reviewing courts and, consequently, receive less judicial deference. By examining why this is so, courts can create a more targeted test for identifying controllers specifically for the purpose of determining when heightened review is appropriate.

The system for reviewing fiduciary decisionmaking is a complex one, and it begins with the principle that corporate directors, like controlling shareholders, owe fiduciary duties to the corporation and its stockholders. However, directors’ decisionmaking is usually subject to an extreme level of deference by courts known as “business judgment review.” If challenged, courts will presume that the directors’ decision was made in good faith, on an informed basis, and in the best interests of the company, and will not look any further before concluding that the directors satisfied their fiduciary obligations. Business judgment review has been described as “something as close to non-review as [Delaware’s] law contemplates,” and challenges to transactions reviewed under the business judgment standard are generally dismissed on the pleadings.

That changes when the transaction is one in which the directors have a personal interest, such as payment of the directors’ salary or purchase of property from a director. For these “interested transactions,” the presumption of good faith is rebutted. Directors’ objectivity and faithfulness can no longer be presumed, and courts will instead examine the transaction to ensure that it was “entirely fair” to the corporation, meaning that it occurred at a fair price after a fair process. Complaints that allege fiduciary breaches arising out of transactions subject to “entire fairness” review will almost always survive a motion to dismiss due to its fact-intensive nature.

Because entire fairness review is onerous, consuming myriad corporate and judicial resources, corporations typically try to “sanitize” these interested-director transactions by seeking approval from an impartial decisionmaker. That approval can come either from the directors who have no financial interest in the transaction (disinterested directors) and who also have no personal ties to those who do (independent directors), or from a vote of the disinterested stockholders, upon full disclosure of the material facts. Disinterested, independent director approval adds another layer of potential complexity: If the board is majority-disinterested and independent, it is generally assumed it can sanitize the transaction; if not, approval can come from a duly appointed committee of disinterested and independent directors. Either way, if the appropriate disinterested body approves the transaction on a fully informed basis, business judgment review is restored.

Matters are slightly different when the transaction is an extraordinary one, namely, a sale of the company, either for cash, or in a manner that will transfer control from public stockholders to a new controlling entity, or in a manner that will result in a fundamental reorganization and break up. These are deemed end-stage transactions, when the company, as it is currently constituted, will cease to exist. In these circumstances, the imminent disappearance of the market mechanisms that ordinarily discipline managers create a heightened risk of self-dealing. Thus, there is a twist: Even if the sale is negotiated and approved by directors who have no obvious personal interest, their actions are still subject to heightened form of judicial review known as “enhanced” scrutiny. Enhanced scrutiny is somewhat more robust than the highly deferential business judgment standard, but not as exacting as entire fairness; for that reason, it is sometimes known as “intermediate” scrutiny. Still, as the Delaware Supreme Court recently made clear in Corwin v. KKR Financial Holdings LLC, whether the transaction is one that would ordinarily receive entire fairness review (e.g., because a majority of the board is interested), or one that would receive enhanced scrutiny because it constitutes a sale of the company, business judgment review can still be restored by a fully informed, uncoerced vote of the disinterested stockholders.

When the transaction is one in which the interested fiduciary is not a director, but is instead a controlling shareholder, the situation changes. There is no concept of “intermediate” scrutiny when it comes to controlling shareholder transactions; either the transaction is one in which the controller is interested because it will receive some benefit not shared by the minority shareholders, or it is not. If the controller has no special interest, the controller’s presence has no effect on the standard of review. When the controller’s interests differ from those of the minority, however, then—as with other interested transactions involving fiduciaries—the deal is presumptively examined for its fairness.

The difference comes in the context of cleansing. Unlike other interested transactions, when the transaction concerns a controlling shareholder, business judgment review cannot be restored by the approval of the disinterested and independent directors or the disinterested (minority) shareholders. Instead, in Kahn v. M & F Worldwide Corp. (“MFW”), the Delaware Supreme Court held that business judgment review can only be restored if the deal is negotiated with several specific procedural protections, including the requirement that it be conditioned from the outset on both independent director and disinterested stockholder approval; neither body may cleanse the transaction acting on its own. So far, the Delaware Supreme Court has only approved the use of MFW procedures as a mechanism for cleansing controller conflicts in the context of certain types of transformative transactions, such as freeze-outs. Chancery courts, however, have agreed that MFW procedures can be used to cleanse all interested transactions involving controlling shareholders.

As tortuous as this regime is, it still rests on certain discernible principles. These principles can be used to identify controlling shareholders specifically for the purpose of determining the appropriate cleansing regime.

As I’ve previously explained, the divergent cleansing mechanisms are purportedly necessitated by the inherently coercive nature of a controlling shareholder’s presence. Controllers, the argument goes, are capable of retaliating against the company generally, the minority shareholders specifically, or the directors personally, if their interests are thwarted. For example, controllers may withhold dividends, or propose a freeze-out merger at a lowball price, or replace the directors, or block favorable corporate actions. That implicit threat will taint the decisionmaking process of either the minority shareholders or the independent directors acting alone; it can only be neutralized “where the controller irrevocably and publicly disables itself from using its control to dictate the outcome” via strict adherence to the procedural protocols outlined in MFW.

But though this is the explanation offered in the caselaw, it does not fully justify the distinction. Interested directors—especially when they constitute a majority of the board—have almost as much retaliatory power as a controlling shareholder. They too can withhold dividends, or forego profitable business opportunities, if their interests are thwarted by an unfavorable shareholder vote. Moreover, just like controlling shareholders, they have bargaining advantages: They are better informed than the shareholders, and they can time a transaction to take advantage of any temporary inaccuracies in market pricing. Yet Delaware courts still permit interested director transactions to be cleansed by a shareholder vote alone.

The best explanation for the divergent treatment, then, stems from the fact that directors are subject to the market for corporate control, which necessarily constrains any retaliatory measures they might take. Controlling shareholders, by contrast, are entrenched; if they retaliate, there are no remedies for affected shareholders other than to simply exit—at a discounted price in a public company, impossible in an illiquid one.

If the entrenchment factor combined with retaliatory capabilities are what necessitate MFW’s onerous cleansing regime, then it is those factors specifically, rather than a free ranging inquiry into control over the company, that should determine the level of judicial scrutiny that a transaction with a putative controller receives. In short, transactions with an entrenched party who is capable of retaliating against the board, the corporation, or the remaining shareholders should be deemed to involve a controller, subject to entire fairness review, absent MFW cleansing.

Under this test, courts can look not only to large equity holdings, but also shareholder agreements that guarantee board representation or approval rights, both of which have been previously recognized as allowing the rights holder to exert “negative control” by blocking corporate action, unless their personal demands are satisfied. Courts might also consider whether certain founders or CEOs are so closely identified with the company that it would be nearly unthinkable to oust them, as well as general indicia of shareholder power (or its absence), such as the existence of staggered boards, proxy access, and the board’s history of responsiveness to precatory shareholder votes.

No doubt, such a test would not be an exact one, but then, the current test for control is hardly a model of precision. Importantly, though, the current test—which looks to whether the putative controller “actually dominated” the corporation or the transaction—when used in the context of cleansing specifically, is in fact a proxy for identifying inherent coercion. Rather than test for the proxy, which will necessarily yield indeterminate results because “control,” in the abstract, is not the object of interest, courts would do better to test for the actual factors that make precautions necessary. Control, of course, would almost certainly be relevant to a determination of inherent coercion, but by focusing on retaliatory capabilities and entrenchment as the guidepost, rather than control per se, courts would know to focus on particular kinds of control rights and have an end point for determining the sufficiency of such control.

There are additional implications. If, in the context of cleansing, courts are concerned that a putative controller might cast a long shadow over the deliberations of independent directors and minority shareholders alike, courts should focus their attention not, precisely, on the retaliatory capabilities of the controller, but on the perception of those capabilities by others.

For example, a frequent issue that arises in these cases concerns the treatment of minority shareholders who have agreed to cooperate with each other. The ordinary rule is that the several shareholders can constitute a “control group,” with the legal consequences that follow, if they are “connected in some legally significant way—such as by contract, common ownership, agreement, or other arrangement—to work together toward a shared goal.” Commonly, however, there is no explicit agreement—at least, not one that has been publicly disclosed—and courts are left to examine the totality of facts to determine whether an agreement of sorts was likely reached among the minority blockholders sub rosa.

That is reasonable enough if the question is whether these shareholders exert enough control to owe fiduciary duties to the corporation, but if the question is whether they are sufficiently intimidating to undermine ordinary cleansing mechanisms, a secret agreement has no relevance; what is far more important is whether they are likely perceived as having one, or at least perceived as exercising sufficient joint power to retaliate in the face of opposition.

Garfield v. BlackRock Mortgage Ventures, LLC allows for a useful demonstration of the principle. In that case, PennyMac was a publicly traded corporation with two classes of stock: Class A, which had 15 percent of the votes, and Class B, which had 85 percent of the votes. Included among the Class B stockholders were PennyMac’s founding sponsors, BlackRock and HC Partners, with 20 percent and 26 percent of PennyMac’s votes, respectively, as well as seven of PennyMac’s eleven directors, several of whom had close ties (employees, former employees, and consultants) to BlackRock and HC.

PennyMac proposed a reorganization that would eliminate the dual class stock structure for the benefit of the Class B holders. The transaction required the approval of all shareholders voting as a class, and, though the Class B shares alone had sufficient votes to control the outcome, 69 percent of the disinterested Class A shareholders voted to approve it as well. Subsequently, one Class A shareholder sued, alleging that the share exchange ratio was unfair to the Class A holders, and violated the fiduciary duties of the interested directors, as well as BlackRock and HC, who, it was alleged, joined together to form a control group with 46 percent of PennyMac’s votes.

Ruling on a motion to dismiss, Vice Chancellor McCormick framed the critical question as whether the plaintiff had alleged, for pleading purposes, the existence of an actual agreement between BlackRock and HC with respect to the reorganization. If there was one, it was likely that, given their voting power and involvement with the board, they had control of PennyMac and thus fiduciary duties to Class A stockholders. Additionally, if they formed a control group, the stockholder vote in favor of the deal was not sufficient to cleanse any fiduciary breaches under MFW. If, however, BlackRock and HC did not form a control group, then the approval of the Class A stockholders was cleansing under Corwin. Vice Chancellor McCormick concluded that, given BlackRock’s and HC’s history in forming PennyMac and their involvement with the reorganization, there was an inference, for pleading purposes, that BlackRock and HC had an “‘actual agreement’ to work together in connection with the Reorganization,” and therefore the plaintiff ’s claims could proceed.

It is hard not to notice the artificiality of the court’s focus. According to the court, MFW cleansing, rather than Corwin cleansing, would have been necessary in the case to eliminate suspicion that the Class A shareholder vote had been coerced. But whether there was an actual agreement between BlackRock and HC was irrelevant to that question. Class A shareholders would have seen that an interested board, with close ties to founding shareholders, were proposing a transaction for the benefit of the Class B stockholders—and further, that they had not taken even the most basic steps of requiring either a majority-of-the-minority vote to approve the deal or fully empowering an independent special committee to negotiate it. Class A shareholders were never told that there was an “agreement” between BlackRock and HC, but it did not matter: They likely suspected the presence of one, and, whether or not the founders actually had an agreement, they could not help but notice that the close relationship between BlackRock and HC allowed them to run the company as they saw fit in the event the transaction was voted down. And it was exceedingly unlikely to have been voted down no matter what the Class A stockholders did, seeing as how the Class B stockholders controlled 85 percent of the vote, which hardly gave the Class A stockholders much incentive to do anything. This is precisely the scenario where, at the very least, the votes of the minority shareholders were unreliable, due to the overweening power of a very small group of interested parties who would have continuing control of corporate operations and would find it relatively easy to coordinate their behavior going forward.

Had the court approached matters from that perspective, it likely would have concluded that the shareholder vote did not cleanse the reorganization, and the transaction would have to be reviewed for entire fairness. Therefore, at the very least, the plaintiff would have stated a claim for breach of fiduciary duty against the seven PennyMac directors who benefitted personally from the deal.

This is not to argue that the seven directors should have been liable for monetary damages simply because BlackRock and HC may have been perceived as controllers; it is only to argue that the seven directors could not expect an unreliable and potentially coerced shareholder vote to cleanse any fiduciary breaches they may have committed. Instead, as interested parties to a transaction that they proposed and approved, they would have a choice between defending the fairness of that deal in court or, ex ante, negotiating in compliance with MFW protections.

Separately, there would have been the question of BlackRock’s and HC’s liability. Even if the deal was not cleansed, if there was no actual agreement between them, they may not have had control over PennyMac’s operations—and without control, they may not have had any fiduciary duties to breach. Thus, the court might have still reached the question whether the plaintiff adequately alleged the existence of an agreement, in order to determine whether a claim had been stated against these entities directly. Alternatively, though, given these entities’ involvement with negotiation of the transaction, and their ties to the board, the plaintiff may have been able to state a claim against them for aiding and abetting the board’s breaches, even without alleging the existence of an agreement. If that were the case, the court could either have avoided the question whether actual control had been alleged (with the question of the existence of an agreement, and thus fiduciary duties, to be determined on a more complete record), or at least could have articulated both potential paths to BlackRock and HC’s liability as a guide to the parties as discovery proceeded.

The advantage to this approach is that, first, it focuses on what actually matters—whether there was a reason to distrust the shareholder vote—rather than the side issue of the existence of an agreement which, in this context, had little relevance to that question. Critically, it also softens some of the bright-line distinctions that the law currently draws between transactions involving controlling shareholders, and those that do not, because board liability would not have turned on the existence of a BlackRock/HC agreement, or the level of control that BlackRock and HC actually exerted. That, in turn, would help alleviate some of the pressure courts may currently experience to hold that parties have control over the board when they believe that Corwin cleansing has not entirely eliminated deal process problems.

The Tesla trial, which recently concluded in Delaware Chancery, could have been approached similarly. The case concerned Tesla’s acquisition of SolarCity. Elon Musk headed both companies, and several Tesla board members either were interested in the deal themselves or had ties to Musk. Thus, the acquisition was an interested transaction, and Tesla shareholders sued, claiming that Musk and the board breached their fiduciary duties to Tesla by favoring SolarCity. However, Tesla shareholders voted to approve the deal, which ordinarily would cleanse any fiduciary breaches; so, the plaintiffs claimed that Musk—with a mere 22 percent Tesla stake at the time of the transaction—was a controlling shareholder, necessitating MFW cleansing. Much of the parties’ arguments focused on the level of control Musk actually exerted.

Ultimately, the court chose not to decide the issue because it deemed the transaction entirely fair regardless of Musk’s status, but consider how it could have been handled. Elon Musk was both the CEO and a director of Tesla; there is no question he had fiduciary duties to the company. Thus, the only legal question was whether the shareholder vote was capable of cleansing the deal—and that question, in turn, should have depended on whether the transaction was so inherently coercive that it could have tainted the shareholder vote. In other words, the standard of review should not have depended on Musk’s level of control over the board, but on his degree of entrenchment and his ability to retaliate against disobedient shareholders. Once again, a more targeted inquiry could have directed the court’s attention to the issues that mattered, and avoided the risk that the definition of control would become muddied in an attempt to avoid Corwin’s harshness.

Distinguishing between control for the purposes of fiduciary duties, versus control for cleansing purposes, may also help guide the analysis when transactional, rather than general, control is at issue. As described above, Delaware recognizes that controlling shareholders may either control the company generally or may control solely with respect to a particular transaction. But, depending on the circumstances, transactional controllers may only have a transitory influence over the board. If that is so, it may be appropriate to impose fiduciary obligations on them, but it may also make little sense to require MFW cleansing procedures; instead, if their lack of entrenchment renders the prospect of retaliation unlikely, shareholder approval may suffice.

SPACs provide a relevant example here. SPACs are companies that go public without any business operations, in the expectation that they will merge with a private company, taking it public (known as a “de-SPAC” transaction). By law and custom, SPAC shares are sold in the initial offering for $10 each; if the shareholders do not like the merger proposal, shareholders may redeem their shares for that $10, plus interest. If the SPAC does not complete a merger within a particular time frame, usually eighteen months, it is obligated to liquidate and return the same $10 plus interest to the shareholders.

In a recent case, SPAC shareholders sued the sponsor of the SPAC, alleging that the de-SPAC transaction was a bad deal for the SPAC shareholders, and that the sponsor had only pursued it because it personally would benefit more from a bad merger than from liquidation. But the SPAC shareholders had voted in favor of the merger, which would have cleansed any breach of fiduciary duty by the SPAC’s managers. Thus, among other allegations, the plaintiffs claimed that the SPAC sponsor, with its roughly 20 percent stake, was a controlling shareholder; if true, the shareholder vote was not a sufficient cleansing mechanism, and the deal was therefore subject to “entire fairness” review.

Under current doctrine, a reviewing court would have to evaluate the plaintiffs’ allegations by looking to the sponsor’s power over the board and the business operations of the SPAC, which may, or may not, yield a finding of a control. But suppose the court instead looked to the sponsor’s degree of entrenchment, and its ability to retaliate in the event of an unfavorable vote. If that were the test, the sponsor would seem to have very little retaliatory power. If the merger were rejected by shareholders, the SPAC would be forced to liquidate, and if the merger went through, shareholders would still be able to redeem their shares for the same amount they would receive in liquidation. Either way, the ephemeral nature of the SPAC itself affords very few opportunities for punitive action by the SPAC sponsor, and that fact alone should weigh against a finding of control for the purposes of determining whether the sponsor’s coercive power tainted the shareholder vote.

Emphasizing entrenchment and retaliatory power also solves a particular conundrum that arises in the context of transactional control. As I’ve previously written, very often an existing fiduciary (such as a CEO, director, and/or founder) is alleged to be a controller with respect to a particular transaction in which they have an interest. In this context, courts struggle to evaluate negotiations that involve some, but not all, of the protections of MFW (such as, for example, use of an empowered special committee of independent directors). These may either be interpreted as failed efforts by a minority controller to cleanse a transaction or they may be interpreted as sufficient precautions to negate the inference of control in the first place. But once the test for control in the context of cleansing is distinguished from the test for control with respect to imputation of fiduciary duties—which, unquestionably, already exist in these cases—the problem becomes more manageable (if it does not disappear entirely). This is because some procedural protections may well minimize the fiduciary’s direct influence over the board, but have a lesser impact on the fiduciary’s level of entrenchment and punitive capabilities going forward, which is the relevant issue for cleansing.

As with any of these inquiries, there will certainly be complexities. For example, in Patel v. Duncan, two minority blockholders together held more than 50 percent of the shares. They had a shareholder agreement that allowed each to select two board members, with a fifth to be selected jointly, and to also put the CEO on the board. Because the board had ten members total, the shareholder agreement functionally gave the blockholders control over the board, and thus arguably made them a control group, with attendant fiduciary duties. The company subsequently engaged in an interested transaction with only one of the two blockholders, and the deal was approved by the directors who were independent of that blockholder alone. Was there “inherent coercion” in this scenario, undermining the independent directors’ approval? It is not clear; the single blockholder alone may not have had the power to retaliate, and because the other blockholder was not interested in the transaction—and the shareholder agreement concerned only director elections—that blockholder may not have had an interest in avenging its compatriot. Because the case was not argued in this manner, however, the court never had a chance to engage the facts from this perspective.

In their article, Optimizing the World’s Leading Corporate Law: A 20-Year Retrospective and Look Ahead, Lawrence Hamermesh, Jack Jacobs, and Leo Strine argue that the concept of “inherent coercion”—whereby the presence of a controller is presumed to undermine either shareholder approval or independent director approval—should be abandoned. They claim that institutional shareholders, independent directors, and SEC disclosure requirements render MFW’s extra protections unnecessary in most cases, and that controller transactions should be cleansed in the same manner as any other interested fiduciary transaction. They would therefore reserve MFW solely for going private deals or for those transactions that can only be legally effectuated via shareholder vote.

If adopted, this proposal would effect a dramatic shift in Delaware doctrine. Though it took a somewhat winding path to get there, by the time of the MFW decision, a significant, if not entirely consistent, body of caselaw both at the Delaware Supreme Court and Chancery levels had affirmed that interested transactions with controlling shareholders would always receive entire fairness review, regardless of the cleansing mechanism used. MFW loosened those standards by providing controllers with a narrow path to business judgment deference. The Hamermesh, Jacobs, and Strine proposal, then, would grant controlling shareholders (including those with more than 50 percent of the corporate voting power) still more leeway, by permitting even very substantial transactions—such as asset sales between the controller and the controlled company—to receive business judgment review on the basis of a single cleansing mechanism.

One problem with that approach is that these are the transactions, such as Tesla’s acquisition of SolarCity, that may be especially vulnerable to manipulation, because they can only be challenged via derivative action, where the demand requirement will introduce an additional barrier to relief. More importantly, though, the premise of the business judgment rule—that courts should not interfere with the decisions of directors whom the shareholders have selected to represent their interests—does not hold when shareholders are functionally disenfranchised due to the presence of a controller who is immune to market discipline. To simply assume away the problem by mythologizing the fortitude of independent directors would only further the distance between legal presumption and practical effect that has troubled Delaware courts ever since Corwin.

It is important to remember how we got to this point in the first place: After Corwin, the doctrinal precept that disinterested shareholders or independent directors could protect against unfavorable deals collided with the reality of exploitative transactions. Removing many of the remaining judicial backstops—without accounting for real-world problems, such as structural bias and the inherently constrained shareholder role—risks creating a body of caselaw that, by rubber-stamping even obviously flawed deals, undermines the credibility of Delaware law. More likely, though, Delaware judges would find an alternative way to address problematic transactions, such as by expanding the definition of dependence, the definition of materiality (so as to deem shareholder votes uninformed), or the definition of coercion, thus reintroducing the problem in a new form. A more straightforward solution would be to give courts the freedom to directly confront the factors that create the appearance of unfairness.

IV. Control as a Valuable Right

The final arena in which controlling shareholders receive distinct treatment concerns litigation procedure, namely, the availability of a direct action by shareholders against a controller.

Ordinarily, when a shareholder alleges that corporate directors caused the company to sell equity for a lowball price, that allegation is considered a waste of corporate assets and thus a harm to the corporation itself. The claim must therefore be brought as a derivative action, which requires demand excusal, and for which remedies—usually—would be paid to the corporation, rather than directly to shareholders.

In Gentile v. Rossette, the Delaware Supreme Court carved out an exception to this rule, holding that, if the lowball sale was to a controlling shareholder, that claim could be brought both directly and derivatively, on the theory that the dilution of voting power was a harm to the minority shareholders distinct from any harms to the corporate entity. As a result, such claims would be easier to bring than other kinds of claims for waste of assets; demand would not be necessary, and, because remedies would be paid to affected shareholders, litigation might be more lucrative. The rule proved difficult to administer, however, as questions were raised about the kinds of transactions to which it would apply, and fifteen years later, in Brookfield Asset Management, Inc. v. Rosson, the Delaware Supreme Court declared the Gentile exception dead.

But when the courts close a door, some way they open a window, and in Brookfield, that window came in the form of an alternative rule that the Delaware Supreme Court announced:

To the extent the corporation’s issuance of equity does not result in a shift in control from a diversified group of public equity holders to a controlling interest (a circumstance where our law, e.g., Revlon, already provides for a direct claim), holding Plaintiffs’ claims to be exclusively derivative … is logical and re-establishes a consistent rule that equity overpayment/dilution claims, absent more, are exclusively derivative.

This appears to have been a rather backhanded way of holding that, if an equity issuance did result in a shift from an uncontrolled status to a controlled one, shareholders would be permitted to bring claims for breach of fiduciary duty directly, rather than derivatively.

To be sure, this is not the only reading of Brookfield’s admittedly Delphic pronouncement. It is possible that the citation to Revlon, a merger case, indicated that the court only intended to refer to sales of control in the context of mergers, where the law already provides for direct claims. But the court specified an equity issuance that shifted control—not an unheard-of scenario—and an equity issuance was not involved in Revlon.

Additionally, the Brookfield court also seemed to have endorsed the notion that equity issuances might give rise to direct claims even if they did not result in the creation of a new controlling shareholder, so long as they ended up redistributing specific control rights away from the public shareholders. The plaintiffs in Brookfield claimed that the company’s charter required a two-thirds vote for certain changes and that the board of directors breached their fiduciary duties by selling enough equity to a 51 percent controller to bring it to that threshold. Though the Delaware Supreme Court found that the plaintiffs had not alleged that controller ever obtained enough equity to amend the charter, and plaintiffs never, in fact, lost the supermajority voting provision, it implied that, if the plaintiffs had done so, they could have sued directly.

The matter is far from certain; after all, one of the reasons the Brookfield court gave for overruling Gentile is that allowing both direct and derivative claims would create questions as to how to allocate remedies, and permitting direct claims for equity issuances that transfer control may recreate the same problem. Nonetheless, at least one plausible reading of Brookfield is that Gentile has not been overruled so much as modified: Dilutive issuances will not always give rise to direct claims but will do so when the company goes from uncontrolled status to controlled status, or even when particular control rights have been shifted away from the shareholders generally to be concentrated in a new entity. In fact, since Brookfield, Delaware Chancery has permitted at least one direct claim to proceed based on allegations that an equity issuance transferred control to a new entity.

Prior to Brookfield, there was little precedent for these ideas outside of the Gentile frame, though they had gotten a small amount of play in the Chancery courts. In In re Loral Space & Communications, Inc., then-Vice Chancellor Strine—who presumably had Gentile in mind, even though he did not cite Gentile explicitly—allowed direct as well as derivative claims to proceed when a board issued additional equity to an existing blockholder with practical, but not explicit, control, in order to grant it special voting rights that amounted to near complete control over the company. In In re Coty Inc. Stockholder Litigation, the plaintiffs alleged various breaches of fiduciary duty when a 40 percent stockholder completed a partial tender offer for the public shares that ultimately brought its holdings to over 60 percent. The court allowed the nontendering shareholders, who collectively held a 40 percent stake, to bring their claims directly, due to the blockholder’s increased power. And in Louisiana Municipal Police Employees’ Retirement System v. Fertitta, the court permitted direct claims to proceed against a board that failed to deploy a poison pill to prevent a company’s chair and 46 percent stockholder from gradually accumulating hard control through open-market purchases.

Assuming this interpretation of Brookfield holds, when courts are confronted with these situations going forward and are forced to determine controller status before and after the sales, the critical point will—again—be to focus on why the question is being asked. If the relevant issue is whether the buyer owed fiduciary duties to the stockholders at the time of the sale, courts should inquire whether the buyer exerted control over the company generally or the transaction specifically. If the relevant issue is whether the transaction was cleansed, then control should be defined in terms of the purchaser’s coercive power. And if the relevant issue is whether the plaintiffs have stated a direct, rather than derivative, claim, courts should return to the policy rationale behind allowing direct claims in the first place.

That policy, apparently, is because the shareholders’ control rights have an independent value that requires compensation. The courts in Coty and Loral both sustained direct claims on the theory that the transfers of control robbed the minority shareholders of the premiums they might otherwise have received for their shares in a subsequent sale, such as through a merger. And the Brookfield court made a similar point in a footnote when it suggested that shareholders would have direct claims for dilutive equity issuances that transferred control:

As we explained in Paramount Communications, Inc. v. QVC Network, Inc., “[i]n the absence of devices protecting the minority stockholders, stockholder votes are likely to become mere formalities where there is a majority stockholder … . The acquisition of majority status and the consequent privilege of exerting the powers of majority ownership come at a price,” and that price “is usually a control premium which recognizes not only the value of a control block of shares, but also compensates the minority stockholders for their resulting loss of voting power.”

This was also the rationale used in Fertitta: that the board’s inaction allowed the chair to obtain control without paying a premium.

All of which implies that, when the direct-versus-derivative issue is on the table, the ultimate question courts must ask is not whether a buyer “actually dominated and controlled the majority of the board generally” before or after the sale; indeed, the answers to those questions may be entirely meaningless in a case where (as the Brookfield plaintiffs argued in the alternative) the claim is that an alteration of control was sold too cheaply. Instead, courts must assess whether, and to what extent, the board’s actions cost shareholders the independent value of their control rights—and, when it comes to the merits, one important issue will be whether the board recognized the value of those rights and negotiated for compensation.

In a roundabout way, Brookfield potentially sheds light on an unsettled issue in Delaware law, namely, the extent to which objections to takeover defenses must be litigated directly or derivatively outside the context of an active proxy contest. Some takeover defenses amount to a grant of control rights to a third party, such as a dead hand proxy put, which allows lenders to call the loan in the event of a change in control. Brookfield suggests that challenges to these kinds of takeover defenses can be litigated directly, at least to the extent that the control rights have some monetary value (which, messily, will collapse the merits issues with the procedural ones). Carrying that logic through to its conclusion would suggest that, when directors arrogate control rights to themselves via entrenchment devices (rather than place them with a third party), that too can be litigated directly.

More generally, the case law now offers intriguing incentives for plaintiffs. On the one hand, a plaintiff will usually want to argue that, prior to a challenged transaction, the purchasing entity was not a controller, and after the transaction, it was a controller, in order to be able to bring claims directly under Brookfield. On the other hand, in many cases, the plaintiff will want to argue that ordinary cleansing mechanisms were not sufficient to bring a transaction within the purview of the business judgment rule, which will require that the purchasing entity had been a controller all along. That sounds like a contradiction, but in fact, the lesson here is that, because control should mean something different for cleansing versus for drawing the direct/derivative distinction, the two positions are perfectly compatible; it would be entirely legitimate for a court to, for example, conclude that a purchaser bought control rights in an inherently coercive transaction, giving rise to a direct claim for fiduciary breach that could only be cleansed via MFW. The critical point for the court will be to determine whether the hard control conveyed in the deal had monetary value, in light of buyer’s existing levels of influence. This precise point was recognized by the Coty court when denying a motion to dismiss.

To bring this all back to where the discussion began, consider the case of In re Sirius XM Shareholder Litigation. There, the board of Sirius XM Radio negotiated for an equity investment from Liberty Media that, among other things, required the board not to deploy a poison pill to prevent Liberty from gaining a majority position by purchasing additional shares at the market price. Later, after Liberty took advantage of the contract and gained control of the company, shareholders alleged that the board and Liberty breached their fiduciary duties by allowing Liberty to avoid paying a control premium. The court dismissed the claim on statute of limitations grounds and never engaged the direct-versus-derivative question, but to the extent the shareholders alleged that the board sold control rights too cheaply, the logic here would suggest that the claim was a direct one.

Then-Chancellor Strine also shed light on the question of when the exercise of contractual rights gives rise to fiduciary obligations, holding that no claims could be brought against Liberty. Liberty had, in a sense, negotiated for the right to “control” the company by blocking the board from adopting antitakeover measures. At the time, however, Liberty was a third party with no obligations to Sirius, and the purpose of the contract was understood from its inception, namely, to allow Liberty to obtain a majority position without paying a premium. Therefore, Liberty owed no fiduciary duties to the company by eventually exercising that right. As then-Chancellor Strine explained, when contracts grant rights to a counterparty, “[t]he use of such rights to obtain control in the situations specifically contemplated by those contracts does not constitute a fiduciary breach.” Liberty was exercising the very right that it “owned” and had paid for in an arm’s length transaction, in accord with the parties’ expectations at the time of contracting. Because it was not controlling the assets of others, it was permitted to act selfishly. Any fiduciary breach—if there had been one—was by the board, when it authorized the contract in the first place.

V. Conclusion

Delaware’s controlling shareholder problem arises from the fact that a single label does far too much work. The importance of the category ends up placing great pressure on its boundaries, as courts try to accommodate a wide range of power arrangements and legal disputes. By disaggregating the scenarios in which control has legal relevance, and by adjusting definitions to fit the context, courts can relieve that pressure while ensuring that the test for control remains tethered to the policies that originally inspired the special legal treatment.

That said, as an alternative to adjusting its definition of control, when it comes to cleansing specifically—which is the context in which many of these disputes arise—Delaware could also relieve the pressure by scrutinizing even noncontroller deals more closely. Though, doctrinally, the divergent treatment between controller deals and noncontroller ones may be traceable to controllers’ degree of entrenchment (and thus their enhanced capacity for retaliation), that explanation may fail to justify the dramatic differences in how courts approach the two types of cases. The reality is, at least when it comes to major transactions like buyouts and restructurings, the advantages that all insiders enjoy with respect to control over process, information, and timing may be far more important than the risks posed specifically by a spiteful controller.

To a limited extent, the proposal in this article addresses that problem: At least in some cases, insiders will be sufficiently entrenched to count as controllers for cleansing purposes, even if they do not “dominate” the board in the absolute sense. But a wiser approach may simply be to reconsider Corwin and its ilk. It is one thing to assume that shareholders can ratify an ordinary-course interested transaction, such as payment of a director’s compensation, and quite another to assume that they can ratify far more complex decisions, like mergers, where no amount of disclosure can level the playing field between insiders and public shareholders, which is exactly the argument that has been made by numerous scholars criticizing Corwin. If Delaware courts were to restore heightened scrutiny to transformative transactions, at least when they present risks of self-dealing, the controlling shareholder label would fade in significance.

This paper greatly benefitted from discussions with, and comments from, Onnig Dombalagian, Joel Fleming, Da Lin, Brian Quinn, and the members of the Delaware Court of Chancery. Errors are very much my fault and not theirs.