«Abstract I show shareholder litigation rights are an important tool for mitigating agency conﬂicts. To empirically identify the eﬀects of ...»
Governance by Litigation∗
I show shareholder litigation rights are an important tool for mitigating agency
conﬂicts. To empirically identify the eﬀects of litigation, I use the staggered adoption
of universal demand (UD) laws in 23 states between 1989 and 2005. These laws
impose a signiﬁcant obstacle to lawsuits against directors and oﬃcers for breach of
ﬁduciary duty. UD laws are associated with increased use of governance provisions
(e.g., classiﬁed boards) that entrench managers or otherwise limit shareholder voice.
I also document fewer institutional blockholders, changes to ﬁnancial policies and CEO compensation, and impaired performance for ﬁrms subject to UD. Overall, my ﬁndings cast doubt on the traditional notion that shareholder lawsuits primarily beneﬁt attorneys rather than corporations or their shareholders.
JEL Classiﬁcation: G34, G38, K22, K41 ∗ I thank the members of my dissertation committee: Todd Gormley, Donald Keim, and Michael Roberts. I also thank Christine Dobridge, Erik Gilje, Itay Goldstein, Jessica Jeﬀers, Doron Levit, Hao Liang, Michael Lee, Jonathan Karpoﬀ, Michelle Lowry, Jillian Popadak, Sang Byung Seo, Eric Talley as well as seminar participants at Wharton, Financial Research Association, Conference on Empirical Legal Studies at Berkeley, Olin Corporate Finance Conference, Academic Conference on Corporate Governance at Drexel, and LBS Trans-Atlantic Doctoral Conference. I thank Moshe Cohen and Todd Gormley for sharing data used in this paper. Grace Chung and Jeyhun Park provided research assistance. I gratefully acknowledge ﬁnancial support from the Jacobs Levy Equity Management Center for Quantitative Financial Research. The title of this paper is borrowed from Chapter 10 of Corporate Governance: Promises Kept, Promises Broken by Jonathan Macey. All errors are, of course, my own.
† The Wharton School, University of Pennsylvania. firstname.lastname@example.org A rich literature in ﬁnancial economics studies ineﬃciencies arising from the separation of ownership and control. Factors that help to resolve agency problems include managerial labor markets (Fama (1980)), legal protections (La Porta, López de Silanes, Shleifer, and Vishny (1998)), and the market for corporate control (Grossman and Hart (1980)). However, shareholders generally exert little inﬂuence over these aspects of corporate governance. Rather, they are primarily conﬁned to three fundamental rights associated with equity ownership: voice, exit, and litigation. While the voice and exit have recently received considerable attention in the literature (Edmans (2014)), the roleof shareholder lawsuits in corporate governance remains unclear. In this paper, I examine whether litigation rights serve as a mechanism to mitigate agency conﬂicts.
I focus on a particular type of shareholder lawsuit, known as a derivative action, that alleges a breach of ﬁduciary duty by directors or oﬃcers.1 Corporate law in the U.S.
requires ﬁduciaries to exhibit prudent judgment (the duty of care) and refrain from selfserving conduct (the duty of loyalty). Derivative lawsuits serve as an ex post enforcement mechanism for these duties. Between 2000 and 2009, over 13% of ﬁrms in my sample faced at least one derivative lawsuit. Yet, there are several reasons why the eﬀects of these lawsuits on ﬁrms may be minimal, if not negative. First, shareholders may be able to adequately exert governance through alternative mechanisms (e.g., voice and exit), making litigation rights redundant. In addition, the “business judgment rule” largely shields managers from liability for corporate decisions. Even when not protected by this legal doctrine, directors and oﬃcers face a very small chance of personal liability due to the pervasive use of exculpatory charter provisions, indemniﬁcation contracts, and directors and oﬃcers (D&O) insurance (Black, Cheﬃns, and Klausner (2006)). Finally, ﬁnancial recoveries from derivative lawsuits tend to be low while legal fees are often high.
In fact, some critics maintain that the primary beneﬁciaries of litigation are lawyers rather than corporations or shareholders (Romano (1991)).
Others argue derivative lawsuits confer beneﬁts that potentially outweigh their costs.
First, most settlements include reform of corporate governance practices. In fact, such reforms are often the primary goal of litigation (Erickson (2010)). For example, a 2008 settlement from a derivative lawsuit against the directors of Schering-Plough implemented These lawsuits are called “derivative” because shareholders sue directors or oﬃcers on behalf of the corporation. Further institutional background is provided in the next section.
annual director elections and removed supermajority voting requirements, while a 2005 settlement involving OM Group featured the termination of the CEO and appointment of two shareholder-nominated directors. In addition to settlements, derivative lawsuits may confer the beneﬁt of future deterrence by imposing non-pecuniary costs (e.g., reputational penalties) on directors and oﬃcers.2 Such costs may, in turn, discourage certain behaviors by managers.
Quantitatively assessing the eﬀects of derivative litigation poses a signiﬁcant challenge.
One possible empirical strategy is to match ﬁrms facing litigation to a control group. This approach suﬀers from two main drawbacks. First, lawsuits are an equilibrium outcome and not randomly assigned. While a matching strategy minimizes ex ante observable diﬀerences between groups, breaches of ﬁduciary duties are also inextricably linked to managerial preferences that are unobservable to the econometrician (e.g., value of private beneﬁts, sensitivity to reputational risk, etc.). Second, and perhaps more importantly, the matching strategy limits analysis to the realization of a lawsuit and cannot account for variation attributable to the deterrence function of litigation.
In this paper, I empirically identify the eﬀects of shareholder lawsuits using variation generated by universal demand (UD) laws at the state of incorporation level. These laws present a signiﬁcant obstacle to derivative lawsuits. Speciﬁcally, UD laws require shareholders to seek board approval prior to initiating derivative litigation. The board rarely grants this approval, however, because lawsuits typically name the directors themselves as defendants. I use UD laws as the “treatment” in a diﬀerence-in-diﬀerences framework. I include ﬁrm, industry-year, and state-of-location-year ﬁxed eﬀects in the main speciﬁcation to control for time-invariant heterogeneity across ﬁrms and time-varying diﬀerences across industries (e.g., demand shocks) and headquarters locations (e.g., local economic conditions). This identiﬁcation strategy addresses the two main shortcomings of the matching strategy. First, because UD laws are adopted at the state of incorporation level, they are largely unrelated to the characteristics of individual ﬁrms. Second, UD laws decrease the threat of future litigation and therefore account for deterrence eﬀects.
I ﬁrst show UD laws aﬀect the incidence of derivative litigation. To do so, I assemble a database of over 900 derivative lawsuits involving public corporations using SEC ﬁlings Brochet and Srinivasan (2014), Ferris et al. (2007), Fich and Shivdasani (2007), Karpoﬀ, Lee, and Martin (2008) provide empirical evidence of indirect costs associated with litigation. Helland (2006) ﬁnds contrasting results.
and other sources. I ﬁnd UD laws are associated with a decrease in derivative litigation of approximately 0.7 percentage points, a drop of over one third relative to the sample mean. The magnitude of this eﬀect is consistent with anecdotal evidence on the eﬀect of UD laws discussed in the next section. I argue this estimate is, if anything, a lower bound on the change to the threat of future litigation.
Next, I turn attention to the eﬀect of UD laws on the governance structures of ﬁrms.
I use the entrenchment index proposed by Bebchuk, Cohen, and Ferrell (2009) as the primary governance outcome. I ﬁnd UD laws are associated with about a 10% increase in the entrenchment index, indicating increased use of antitakeover provisions relative to the control group. The change in governance structures is largely driven by increased use of poison pills, supermajority voting requirements, and classiﬁed boards. I obtain qualitatively similar results for the GIM index (Gompers, Ishii, and Metrick (2003)), which considers a wider array of governance provisions. UD laws are also associated with a drop in the presence of large institutional blockholders. Additional evidence points to this being a consequence of the increased use of antitakeover provisions.
I also consider the eﬀect of litigation on diﬀerent corporate policies that are potentially sensitive to agency conﬂicts. First, UD laws are associated with a shift in the composition of CEO pay that reduces sensitivity to ﬁrm performance. Speciﬁcally, the ratio of cash compensation to total compensation increases by approximately 6%. However, this is oﬀset by a decline in equity-linked compensation, leaving total pay unchanged. Second, I ﬁnd evidence of a decline in share repurchases following UD, but no change in cash dividends.
While this result may support the idea that managers retain cash to engage in “empire building,” there is no change to measures of ﬁrm size or investment. Rather, I ﬁnd lower debt issuance and book leverage, consistent with managers either attempting to reduce ﬁrm risk or otherwise lessen the disciplining eﬀects of debt.
Finally, I show UD laws are associated with weaker accounting performance. Specifically, ROA declines by approximately 0.8 percentage points for ﬁrms subject to UD.
Consistent with the idea that shareholder voice and exit may substitute for litigation rights, I ﬁnd the drop in proﬁtability is driven by ﬁrms with low institutional ownership.
The eﬀect is also stronger for small ﬁrms, which may have weaker external governance mechanisms (e.g., monitoring by regulators) and ﬁrms with high cash ﬂows, which may be more prone to agency conﬂicts (Jensen (1986)).
The interpretation of the results rests on the assumption that the adoption of UD laws is independent of unobserved variables that inﬂuence corporate policies and outcomes.
Political economy factors are of particular concern in this regard. For instance, it may be the case that ﬁrms incorporated in a particular state lobbied for the statutes in response to heightened risk of litigation. To address this concern, I restrict the sample of treated ﬁrms to Pennsylvania, where UD was implemented by the state supreme court in Cuker v. Mikalauskas (1997). The eﬀects of lobbying are likely muted for this sample of ﬁrms since UD was not enacted by legislators as a matter of public policy, but by the courts for the sake of consistency with judicial precedent. This test yields results similar to the main analysis.
This paper builds on the literature that studies the eﬀects of shareholder litigation.
The bulk of this literature focuses on class action lawsuits (e.g., DuCharme, Malatesta, and Sefcik (2004); Hanley and Hoberg (2012); Hopkins (2014); Arena and Julio (2014);
Lowry and Shu (2002)). Derivative litigation is the speciﬁc focus of Ferris et al. (2007).
A related line of literature considers the eﬀects of ﬁduciary duties and director liability on ﬁrm policies and stock returns (e.g., Becker and Strömberg (2012); Donelson and Yust (2014); Grinstein and Rossi (2014)). I contribute to this literature by oﬀering a novel identiﬁcation strategy to study litigation that accounts for its deterrence eﬀects.
My results suggest such eﬀects inﬂuence multiple dimensions of corporate behavior and ultimately impair performance.
I also oﬀer new insights into the broader literature on corporate governance. A voluminous literature studies the relation between governance indices and diﬀerent corporate policies and outcomes (e.g., Bebchuk, Cohen, and Ferrell (2009); Chava, Livdan, and Purnanandam (2009); Cremers and Nair (2005); Gompers, Ishii, and Metrick (2003)). Other papers study factors that lead to ﬁrms implementing anti-takeover provisions in the ﬁrst place (e.g., Schoar and Washington (2011); Field and Karpoﬀ (2002)). I complement these papers by showing that the threat of litigation can also shape the governance structures of ﬁrms by restraining the adoption of provisions that entrench managers or otherwise limit shareholder voice.
Finally, this paper is related to previous work on the eﬀects of antitakeover laws.
In their seminal paper, Bertrand and Mullainathan (2003) argue business combination (BC) laws reveal managerial preferences to “enjoy the quiet life.” Other papers have analyzed the eﬀect of BC laws on innovation (Atanassov (2013)), use of debt (Francis et al.
(2010); Garvey and Hanka (1999)), executive compensation (Bertrand and Mullainathan (1999)), diversifying acquisitions (Gormley and Matsa (2014b)), payout policy (Francis et al. (2011)), and diﬀerential eﬀects based on industry competition (Giroud and Mueller (2010)). This paper provides a new environment to study the nature of agency conﬂicts that arise from shocks to corporate governance. One advantage of UD laws is that their relatively recent enactment allows researchers to test the implications of diﬀerent agency theories using datasets that do not overlap with the period most BC laws were adopted.
1 Institutional BackgroundA. Derivative Lawsuits
From a legal perspective, a corporation is a creature of statute that exists independently of its shareholders. Thus, if directors or oﬃcers engage in behavior that harms the corporation, the corporate entity itself can initiate litigation. A derivative lawsuit entails shareholders suing directors and oﬃcers on behalf of the corporation to address a breach of ﬁduciary duty. By way of example, suppose a manager wastes corporate assets by overpaying for an acquisition after failing to perform adequate due diligence. The primary recipient of harm from this action is the corporation because it (not the shareholders) was the owner of the wasted assets. Shareholders may be injured as well (e.g., by a lower stock price), but this injury is indirect in nature since it results from damage to the corporation.