image description

A Scotch Verdict on Regularity and Other Issues

by Thomas A. Dubbs
Wisconsin Law Review |
The circularity hypothesis is simply that: a hypothesis. Some have viewed the mere assertion of the hypothesis as proof of its correctness


I was asked to respond to the remarks of Professors James Cox and Randall Thomas relating to current issues in securities class-action study, and, in particular, to comment on what future research areas might be explored.' Among several issues, two potential ones stand out: the first deals with the so-called circularity theory or hypothesis, and the second is that of directors and officers (D&O) insurance.


Professor James Spindler has recently summarized the essence of the circularity hypothesis, a theory adopted in one form or another by several participants at the Institute for Law and Economic Policy conference:

The most fundamental criticism of the compensatory function of the fraud on the market mechanism is the argument that diversified investors do not benefit from securities liability on the premise that losses from fraud are diversifiable risk. The reasoning is that since a trader is just as likely to be on the winning side of a fraudulent transaction as the losing side, gains and losses ought to even out with a large number of trades. If that reasoning holds, then 10b-5 liability simply shifts money from one pocket of the investor to another, minus transaction costs.(2)

Indeed, Professor Joseph Grundfest has recently framed the issue more aggressively, characterizing the private-securities-litigation framework as a mechanism "for moving money around for the benefit of the people moving the money around."' He argues further that there is little deterrent effect produced by the current system, and, given the use of D&O insurance, he concludes there is no, or little, "individual responsibility" assigned to corporate managers.(4) He asserts that the. current system is "nothing more than a very elaborate and costly pocket shifting," as a result of which it is mathematically impossible for a fully diversified investor to come out ahead.(5)

There is little doubt that the circularity hypothesis has received significant attention in the academic literature. The purpose here is to outline areas of inquiry and open questions with respect to the theory, given the relatively uncritical acceptance that it has received in academic circles.(6) Close scrutiny of the hypothesis is warranted given that it has led to questioning the compensatory nature of securities class-action lawsuits in general. The need for closer examination is only heightened by the U.S.-Chamber-of-Commerce-funded research conducted by Professor Anjan Thakor, which purports to provide empirical evidence to support the circularity hypothesis.(7)

As noted, at its core, the circularity theory posits that diversified investors (or some of them) that trade actively are not harmed by securities fraud. The theory pivots on the legal rule that does not require a "claw back," or disgorgement, of gains that are obtained by "winning" investors who sold their stock in a firm before fraud was disclosed by the firm. It thus assumes that those gains are wrongfully "captured," and should ultimately be netted against instances where such investors "lost" because of fraud announced by another firm. Over time, the argument goes, such investors experience either a "wash" or are actually overcompensated by the recoveries they receive from participating in securities class-action lawsuits.(8)

Initially, it should be emphasized that a number of the proponents of circularity concede several important points that raise questions about the scope and application of the theory. Importantly, they concede the theory applies only with respect to highly diversified, active traders with a certain turnover rate that is presumably constant, or relatively constant, over the period being studied.' In fact, however, Professor Richard Booth and other observers sometimes argue that circularity applies to all such diversified institutional investors, which suggests that those who engage in index trading are also included.(10)

According to Booth, "most investors are diversified" and are thus "effectively protected against securities fraud."(11) Along those lines, Thakor states that the holdings of large institutional investors constitute "roughly half of the total market capitalization of U.S. common stocks." (12) However, these facts beg the question of what happens to the rights of the remaining nondiversified investors who Booth concedes "may suffer significant harm from securities fraud."(13)

In calling for drastic change of the current securities class-action framework, the implicit assumption by the advocates of the circularity hypothesis is that such defrauded investors-this "half" of the market-simply should not obtain any recovery. Whether this is because these investors are not "morally worthy" because they have not diversified sufficiently or otherwise is not clear. At the very least, by proposing that large numbers of investors be left without any recourse, this is-to borrow a phrase-"academic activism" that rewrites clearly manifested Congressional intent as to who can recover damages for securities fraud.

Likewise, the circularity hypothesis assumes that even as to large, diversified, actively trading investors, the turnover rate must be fairly constant, which may not be a reasonable assumption. For example, if we look at the Magellan Fund, when it was managed by Peter Lynch from 1977 to 1990, we find that its turnover rates varied.(14) Of course, even Lynch bought stocks relying on the market, the prices of which were artificially inflated by fraud and, when disclosed, lost value. Yet, according to many of the proponents of the circularity hypothesis, investors in such instances should not be entitled to any recovery regardless of the variation in turnover ratios."(15)

Along these lines, it is not clear what turnover rate produces the "netting out" that the proponents of circularity predict. What is clear, however, is that turnover rates among institutional investors vary widely. For example, in 1993, the turnover rate among institutional investors with more than $100 million under management ranged from less than 5 percent to nearly 400 percent.(16) During that year, 20 percent of such investors had turnover rates of 25 percent or less,(17) while more than 50 percent of institutional investors had turnover rates of 50 percent or less." Similarly, in 2004, approximately half of all stock funds had turnover rates of less than 60 percent, with nearly three-quarters of all such funds having turnover rates below the institutional average of 117 percent that year.(18)

An additional question is whether the circularity hypothesis includes index funds. This issue is relevant because, although index funds certainly satisfy the circularity theory's requirement of a large and diversified portfolio, it is not clear that such funds trade with sufficient frequency-or, in particular, engage in a sufficient number of purchases and sales-to create the posited "netting" effect, because these funds typically have turnover rates far below those of actively managed funds.(19) Professor Thakor, for example, bases his findings on section 13(f) filings,(20) but those filings almost certainly include a variety of index funds (or institutions that rely heavily on indexing), which he may not have screened out or otherwise adjusted for. If that is the case, then the circularity hypothesis, even if proven, applies to even less than the "roughly half' of the investors some assert.

Moreover, the proponents of circularity agree-but downplay-that it does not apply to initial public offerings, merger and acquisition deals, or other transactions that involve capital raising from a firm where the damages remedy is more akin to disgorgement. The impact of these cases,(21) which represent a significant number of securities class-action lawsuits, also needs to be examined.(22)

With respect to Thakor's study, the reaction of the academic community needs to be analyzed given the limits of his work. Indeed, it does not appear that the dataset on which Thakor relies, or the intricacies of his methodologies, have ever been fully disclosed or-perhaps more importantly-thoroughly peer reviewed. Along those lines, to reiterate the questions noted above, it is unclear how Thakor deals with index funds or accounts for potential inconsistencies in the trading patterns or turnover rates among the institutional investors he studied. Moreover, and what may be most troubling, is the seemingly uncritical acceptance by academics of Thakor's study, which was commissioned by the Chamber of Commerce's Institute for Legal Reform, an institution whose hostility toward securities class-action lawsuits is well known and cannot credibly be disputed.

Why are these important questions? If, as Professor Grundfest argues, securities class-action lawsuits (or at least "open market" cases) are effectively a con game, and there are only Thakor's findings to prove this hypothesis, the academic community should carefully examine such proof and require replication of that study if it is to be relied on as a predicate for reform.

Moreover, even if the circularity hypothesis was in fact proven, why have its adherents adopted a conclusion that is at odds with the logical endpoint of the analysis? That is, that, at most, a procedural remedy should _be developed to exclude certain diversified market participants from a certified class who have no "true" overall losses as posited by the circularity hypothesis while including other damaged investors to whom the theory does not apply. Rather, the adherents have concluded that the whole regime is infirm. This logical contradiction should be addressed, and, even if the theory is proven, attention should be focused on developing a tailored remedy.


Another area in which additional research should be conducted is the role of D&O insurance in securities class-action lawsuits. This topic has been discussed in a helpful trilogy of articles by Professors Tom Baker and Sean Griffith." Certainly the issue of D&O insurance should be explored with respect to cases deemed to be nonmeritorious based on their relatively small settlement amounts, which may, in reality, be the reflection of an absence of D&O insurance. These cases may or may not have additional problems on the merits. This question deserves greater exploration, which is especially true given the observation of Baker and Griffith that "the vast majority of securities claims settle within or just above the limits of the defendant corporations' D&O coverage. As a result, it is not a stretch to assert that the principal party at interest in most securities class actions is the D&O insurer."(25)

A point not answered by Baker and Griffith that deserves attention is the extent to which class-action settlements increase premiums paid by settling defendants. This requires a structural analysis of the D&O insurance market. This is especially true given the relatively small number of market participants coupled with what many believe to be its noncompetitive pricing practices that have included price setting designed to take short-term market share from a limited number of competitors. In short, the assumption of premium increases flowing from securities class-action settlements should not be assumed until empirically demonstrated. And even if premiums are reduced and costs saved, are those savings returned to shareholders via dividends, or are they large enough to lead to an increased share price that, in turn, actually benefits shareholders? If transaction costs are to be evaluated, and criticized as Professor Grundfest does, then all of the real benefits and costs should be analyzed.

Finally, another D&O-insurance-related. issue overlaps with the circularity hypothesis, namely the role that D&O insurance might play with respect to deterrence. This question has experienced increased scrutiny because, as alluded to above, some of the adherents to the circularity theory call into question the compensatory function of securities class-action lawsuits. Indeed, many proponents of circularity, including Professors Coffee and Grundfest, complain that the individual wrongdoers-that is, corporate executives who have manipulated earnings or engaged in other wrongful conduct-are not being hit in the pocketbook because of the very existence of D&O insurance.(26) There may be some specious truth in that, but the reasoning is a little more complex than that articulated.

Under controlling case law, a lead plaintiff is a fiduciary for absent-class members.(27) As a result, a lead plaintiff is often faced with a dilemma during settlement negotiations. Oftentimes during such negotiations, a D&O insurance carrier will put the policy on the table, which usually amounts to the policy proceeds minus legal fees and costs incurred, minus a "haircut" that counsel for the carrier extract to show they did a good job with respect to the settlement. Even if the full amount of the policy is presented without any haircuts, the D&O insurance carrier demands releases running to all of its insureds, which include the very corporate officers who have engaged in the wrongful conduct. In such instances, a fiduciary is faced with the dilemma of whether to reject the amount of the policy proceeds being offered, or instead pursue the litigation -in the hope that it can recover more from the insured plus those individual defendants, while at the same time knowing that such additional litigation will waste the policy.(28) Thus, a fiduciary intent on pursuing individual wrongdoers risks there being less, or possibly no, compensation for the class if it cannot recover from the individual wrongdoers and the D&O policy becomes depleted. Not surprisingly, very few fiduciaries are willing to take that risk.(29)

Those advancing reforms that lead to a regime targeting corporate officers in the name of deterrence while rejecting compensatory goals need to deal with this conundrum.


The circularity hypothesis is simply that: a hypothesis. Some have viewed the mere assertion of the hypothesis as proof of its correctness. However, a close reading and analysis of the literature by even its proponents concedes that the theory applies to only a restricted group of large, diversified financial institutions. Its adherents should concede and expand, where called for, the limitations and assumptions of the theory, including dealing with variable turnover rates, and explain why, even if proven, it does not lead to different, smaller classes rather than a total dismemberment of the current compensatory scheme. In sum, the circularity-hypothesis argument at most results in a Scotch verdict of "not proven," which should be acknowledged by the academic community.

Finally, given the importance and role of D&O insurance, more analysis needs to be done on how it affects the litigation process and whether securities class-action settlements axiomatically lead to meaningful premium increases.


1 This commentary reflects and updates comments made at a roundtable discussion at the 14th Annual Institute for Law and Economic Policy (ILEP) Conference on April 10, 2008, by Professors Cox and Thomas.

2 James C. Spindler, Vicarious Liability for Bad Corporate Governance: Are We Wrong About 10b-5? 5 (University of S. Cal. Law Sch. Ctr. in Law, Econ. & Org. Research, Paper No. C08-3, 2008), available at This argument has been elaborated by any number of eminent academics. See, e.g., Janet Cooper Alexander, Rethinking Damages in Securities-Class Actions, 48 STAN. L. REV. 1487, 1502 (1996); John C. Coffee, Jr., Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation, 106 COLUM. L. REV. 1534 (2006) [hereinafter Coffee, Reforming the Securities Class Action]; Merritt B. Fox, Demystifying Causation in Fraud-on-the-Market Actions, 60 Bus. LAW. 507, 529 (2005); Donald C. Langevoort, Capping Damages for Open-Market Securities Fraud, 38 ARIZ. L. REV. 639, 646 (1996); Donald C. Langevoort, On Leaving Corporate Executives "Naked, Homeless and Without Wheels": Corporate Fraud, Equitable Remedies, and the Debate over Entity Versus Individual Liability, 42 WAKE FOREST L. REV. 627, 632 (2007); Richard A. Booth, Who Should Recover What in a Securities Fraud Class Action? (University of Md. Legal Studies, Research Paper No. 2005-32, 2005), available at http://papers.ssrn.comisol3/papers.cfm?abstract_id=683197; John C. Coffee, Jr., Causation by Presumption? Why the Supreme Court Should Reject Phantom Losses and Reverse Broudo (Columbia Law Sch. Ctr. for Law & Econ. Studies, Working Paper No. 264, 2005), available at http://ssrn.comlabstract=648624; Letter from Donald C. Langevoort et al. to Christopher Cox, Chairman, Securities and Exchange Commission (Aug. 2, 2007) (on file with author).

3 See Kevin LaCroix, Private Securities Litigation: Important Deterrent or Wasteful Chum?, DANDODIARY.COM, Oct. 26, 2008, http://www.dandodiary.coml 2008/ 10/articles/securities-litigation/private-securities-litigation-important-deterrent-or¬wasteful-churn (summarizing Prof. Grundfest's recent comments).

4 Id.

5 Id.; see also sources cited supra note 2 and accompanying text.

6 Indeed, even the Harvard Law Review recently referred to-as an uncontroverted proposition-"the limited ability of [securities class] actions to fulfill their intended goals of compensation and deterrence." S ecurities Litigation-Class Certification Fifth Circuit Holds that Plaintiffs Must Prove Loss Causation Before Being Certified as a Class-Oscar Private Equity Investments v. Allegiance Telecom, Inc., 487 F.3d 261 (5th Cir. 2007), 121 HARV. L. REV. 890, 896 (2008).

7 See generally ANJAN THAKOR, THE UNINTENDED CONSEQUENCES OF SECURITIES LITIGATION (2005), available at get_ilr_doc.php?id = 857.

8 Of course, it is worth noting that many institutional investors with potential damages do not submit claims in securities class-action lawsuits, resulting in these investors foregoing as much as $1 billion per year in potential recoveries. See Coffee, Reforming the Securities Class Action, supra note 2, at 1547.

9 Specifically, Professor Booth posits, "A diversified investor who owns 200 to 300 different stocks with a modest turnover of about 60 percent per year (as is the case with a typical mutual fund) is likely to see gains and losses that are roughly equal over the course of a few years." Booth, supra note 2, at 5.

10 Id. at 6-7; Richard A. Booth, The Paulson Report Reconsidered: How to Fix Securities Litigation by Converting Class Actions into Issuer Actions 2-3 & n.2 (Villanova University Sch. of Law, Working Paper No. 2008-01, 2008), available at For example, Professor Booth cites studies that indicate that only approximately one-quarter of trading volume is attributable to traditional stock picking, with the remaining three-quarters attributable to other investment strategies. Booth, supra, at 2-3 & n.2.

11 Booth, supra note 2, at 5.

12 ANJAN V. THAKOR, THE ECONOMIC REALITY OF SECURITIES CLASS ACTION LITIGATION 11 (2005), available at http://www.instituteforlegalreform.comlissues/ docload. cfin?docId = 855.

13 Booth, supra note 2, at 9. Indeed, Professor Booth concedes that "not all stockholders are diversified" and recognizes that "the average individual investor in the U.S. holds just four stocks." Booth, supra note 10, at 4 n.7.

14 For example, between 1984 and 1990, the Magellan Fund's turnover rate Varied from 82 percent to 126 percent, broken down as follows: 85 percent (1984), 126 percent (1985), 96 percent (1986), 96 percent (1987), 101 percent (1988), 87 percent (1989), and 82 percent (1990). ROBERT F. BRUNER, FIDELITY MAGELLAN FUND, 1995, at 15 (1995).

15 The Magellan Fund's turnover rate also varied significantly in more recent times. For example, between 2002 and 2006, the fund's turnover rate ranged from 6 percent to 74 percent, broken down as follows: 15 percent (2002), 21 percent (2003), 13 percent (2004), 6 percent (2005), and 74 percent (2006). SEC, FORM N-CSR, CERTIFIED SHAREHOLDER REPORT OF REGISTERED MANAGEMENT INVESTMENT COMPANIES: FIDELITY MAGELLAN FUND (Sept. 30, 2006).


17 See id. at 163 fig. 6-1.

18 BRIAN REID & KIMBERLEE MILLAR, MUTUAL FUNDS AND PORTFOLIO TURNOVER 2 (2004), available at http:l/www.ici.orglstatementslreslrc_vin2.pdf.

19 For example, the turnover rate for index funds that represent the S&P 500 is approximately 5 percent. Robert B. Wolf, Total Return Trusts: Meeting Human Needs and Investment Goals Through Modern Trust Design, in SOPHISTICATED ESTATE PLANNING TECHNIQUES 343, 408 (2002). The S&P 500 itself had a turnover rate of 5.21 percent in 2007, while the rate for the S&P Composite 1500 was only 6.9 percent. Richard A. Booth, T he Buzzard Was Their Friend- Hedge Funds and the Problem of Overvalued Equity, 10 U. PA. J. Bus. & EMP. L. 879, 897 (2008).

20 The SEC requires institutions with assets greater than $100 million to disclose their aggregate stock positions each quarter. 17 C.F.R. § 240.13f-1 (2008).

21 See Coffee, Reforming the Securities Class Action, supra note 2, at 1556; LaCroix, supra note 3; Booth, supra note 10, at 2; Booth, supra note 2, at 7, 21.

According to Cornerstone Research, about 12 percent of securities class-action lawsuits in 2006 involved initial offerings. See CORNERSTONE RESEARCH, SECURITIES CLASS ACTION FILINGS, 2006: A YEAR IN REVIEW 20 exh. 14 (2007), available at This 12 percent does not include private placements to which the theory also does not apply, nor does it include mergers-and-acquisitions-based actions.

23 My observation on the importance of D&O insurance as a potential topic that should be incorporated into any analysis of securities class-action lawsuits was shared by my cocommentator at the ILEP Roundtable, Irwin H. Warren, of the firm Weil Gotshal & Manges LLP.

24 Tom Baker & Sean J. Griffith , Predicting Corporate Governance Risks: Evidence from the Directors' & Officers' Liability Insurance Market, 74 U. Ctn. L. REV. 487 (2007) ; Sean J. Griffith, Uncovering a Gatekeeper: Why the SEC Should Mandate Disclosure of Details Concerning Directors' and Officers' Liability Insurance Policies, 154 U. PA. L. REV. 1147(2006 ); Tom Baker & Sean J. Griffith, How the Merits Matter: D&O Insurance and Securities Settlements (University Pa. Inst. for Law & Econ. Research, Working Paper No. 08-22, 2009) [hereinafter Baker & Griffith, How Merits Matter].

25 Baker & Griffith, How Merits Matter, supra note 24, at 107.

26 See Coffee, Reforming the Securities Class Action, supra note 2; LaCroix, supra note 3 (summarizing Professor Grundfest's comments in a discussion on Oct. 23, 2008).

27 See Deposit Guar. Nat'l Bank v. Roper, 445 U.S. 326, 331 (1980); Martens v. Thomann, 273 F.3d 159, 173-74 n.10 (2d Cir. 2001); Crawford v. Equifax Payment Servs., Inc., 201 F.3d 877, 880 (7th Cir. 2000).

28 D&O policies are known as "wasting policies" because their coverage amount is reduced by the amount of legal costs and expenses that are incurred in defending the litigation. Accordingly, the amount available to a plaintiff is usually diminished by continuing to litigate the case.

29 It is unclear whether D&O insurance carriers must reserve amounts sufficient to cover all of their insureds, including insureds for whom it has denied coverage because of, for example, admitted criminal conduct as evidenced by guilty pleas. Whether or not a court would, in dealing with an extreme case of a convicted felon, permit the D&O insurance carrier to pay the class on the policy and insulate it from a later collateral attack by the insured, notwithstanding the denial of coverage based upon the conviction has, to my knowledge, never been litigated. It would take this sort of action to insulate a carrier before any carrier would, in all likelihood, pay the policy proceeds while not reserving some part of the policy for one of its insureds.