The negotiated resolution of securities class action lawsuits – and absent dismissal, there is rarely any other types of securities suit resolution – is always complicated and occasionally messy, and often involves inefficiencies and sometimes produces distortions and even excesses. Anyone who has ever been through a securities suit settlement negotiation likely will have had the thought that there has to be a better way for resolving the cases.
In an April 12, 2012 paper entitled “How Collective Settlements Camouflage the Costs of Shareholder Lawsuits” (here), Fordham Law School Professor Richard Squire catalogues the many shortcomings in the current securities class action settlement process and sets out his proposal to improve the process and to eliminate process inefficiencies and excesses. UPDATE: Please note that Professor Squire also completed a separate response to this blog post in a seprate guest post of his own. His guest post can be found here.
According to Squire, securities class action settlements suffer from a “collective action problem,” owing to the fact that current practices and law require a single case resolution that collectively binds the defendant and all of its D&O insurers – even though the D&O insurance itself is “segmented” in a tower of insurance with the insurers in the different layers having different settlement positions and differing perspectives and interests regarding the settlement.
Among other things, Squire notes that insurers in the primary layers and lower level excess layers are often compelled to contribute toward settlement when the settlement demand (or more accurately, the settlement opportunity) exceeds their layer. This compulsion, Squire notes, is often effectively given legal force through a rarely identified but nonetheless very real “duty to contribute.” These forces lead to a number of ills, including “plaintiff overcompensation at insurer expense”; overpriced liability insurance; and lawsuits of doubtful merit.
Identifying the requirement for collective settlements as the source of the problem, Squire proposes allowing “segmented settlements” – that is, allowing each defense-side party (and in particular each of the carriers in the D&O insurance tower) to “settle with the plaintiffs separately for its respective slice of the damages ranges.” Under this approach, trial would occur unless all slices settled and the plaintiff would collect at trial only those awarded damages (if any) that fell within the unsettled slice.
Squire postulates that this approach would eliminate the conflict of interest between the defense-side parties, “removing the cramdown dynamic that can lead to plaintiff overcompensation.” The elimination of this dynamic will, Squire contends, ultimately lead to more shareholder lawsuit settlements being paid by corporations rather than by insurers, an outcome Squire that further contends would “benefit shareholders, as it would improve the accuracy of a firm’s reported earnings as a measure of the contribution of that firm’s managers to overall shareholder wealth.”
I have set out below my thoughts about Squire’s proposal. I note at the outset that I approach commenting on academic papers with trepidation. Even though D&O insurance and D&O claims resolution are areas to which I have devoted my entire professional life, the world of academic analysis, even with respect to a topic within my area of expertise, seems unbound by constraints that operate in the world to which I am accustomed. My usual trepidation is even greater where, as here, I am already committed to commenting in person on the academic analysis in a public forum. Specifically, on May 8, 2012, I will be attending a conference at Fordham Law School at which I will be participating in panel in with my friends Tom Baker of U.Penn. Law School and Sean Fitzpatrick of Endurance Risk Solutions. The purpose of the panel is to discuss Professor Squire’s paper.
As a threshold matter, I will say that an important aspect of Squire’s analysis for which I give him high marks is his understanding of the central importance of D&O insurance in the securities class action settlement process. All too often, commentators under-appreciate the significance of the role that D&O insurance plays in the process. A particularly important insight Squire has with respect to the role of D&O insurance is that the different D&O insurers’ interests and positions in the settlement process differ based on where they are in the insurance tower.
Squire’s understanding of the role of D&O insurance is particularly accurate when he describes the “cramdown” effect – that is, the pressure that the insured company and the upper level excess carriers can bring to bear on the primary and lower level excess insurers to force the lower level insurers to throw in their limits. Squire perceptively describes what is too often unconsidered, which is the presumed “duty to contribute” that compels lower level insurers to tender their limits where there is pressure to settle a case at a number beyond their limit of liability.
Squire is also on target with his identification of the undesirable consequences this dynamic can produce. It can, as he notes, produce plaintiff overcompensation (which he also correctly notes, inures disproportionately to the benefit of the plaintiffs’ lawyers) and it can result in higher priced liability insurance. And the opportunity for further occasions of overcompensation undoubtedly attracts additional lawsuits.
Having identified these problems with the current system, the question is whether Squire’s proposed solution would in fact solve the problems in an appropriate and acceptable way, a question to which I turn below. However, having noted above the respects in which I agree with Squire’s understanding of the process and the dynamic, I must also as a preliminary matter identify the respects in which my understanding differs from Squire’s. These considerations may or may not alter the ultimate merits or demerits of Squire’s proposed solution to securities suit settlement; but to the extent these considerations might (and they well might, given that much of Squire’s analysis depends on these understandings and assumptions), it is worth my setting out here my differing understandings.
First and foremost, D&O insurance provides a contractual way for companies to manage their indemnification obligations, and to ensure that these indemnification obligations can be honored even if the company itself is unable to do so when the need arises. Because D&O insurance derives from the indemnification obligation, and because the company’s indemnification obligation includes both the obligation to provide for defense expense and for indemnity amounts, the D&O insurance policy has always provided coverage for both defense expense and for settlements and judgments.
Second, companies buy D&O insurance in a marketplace that has been “soft” for almost all of the last 25 years. During that time, dozens of new insurers have entered the marketplace, while at the same time there are many fewer public companies than there were even a short time ago. With an abundance of insurance capacity chasing a dwindling number of insurance buyers, the marketplace is heavily tilted in the favor of the buyers. Buyers expect and get very broad coverage for prices that remain advantageous for the buyer. Owing to the constraints of competition, insurers have a limited ability to impose defensive measures or to constrain coverage. The insurers’ options are two-fold: either to play ball or to sit out the game. Most decide to continue to play.
Third, as a result of the number of persons insured under D&O policies and the number of insurers that are involved in the typical D&O insurance tower, there are in connection with any securities class action settlement many participants, each with their own counsel. Often there are other parties (auditors, underwriters) who are named as defendants but who are strangers to the D&O policy. Not only are the insurers’ interests not aligned but often the insured persons’ interests are not aligned, and the insured persons’ interests often differ from the other named defendants. The carriers’ interests differ not only according to their attachment point in the insurance tower, but also based on situation specific factors such as how much defense expense has been expended and what the anticipated defense expense burn rate will be. The various carriers’ approach to settlement may also depend upon whether they believe they may have coverage defenses that potentially affect their payment obligations.
Not only are there a host of negotiating parties, there often are a host of proceedings involved, other than just the securities class action lawsuit. If a securities suit is serious enough to implicate the excess insurance, there will almost always be multiple other actions ongoing at the same time – usually a parallel derivative suit, often a pending SEC investigation or enforcement action, sometimes even a DoJ investigation or prosecution. The existence of these other proceedings can complicate the settlement dynamic in myriad ways. Serious securities class action settlement negotiations usually do not take place in a vacuum.
A final point about securities suits that should be emphasized is that securities class action lawsuits go to trial so rarely that the possibility of trial can safely be disregarded for most practical purposes. Everyone involved in the case knows it will never go to trial. There are very good reasons that these cases very rarely go to trial. The most important is that the theoretical damages almost always exceed the available insurance and in many cases outstrip the defendant company’s financial resources as well. A less dramatic constraint is that trials are costly, burdensome, and unpredictable and would constitute a huge distraction on senior company management. By the same token, the company could ill-afford a judicial determination that conduct that would preclude coverage has taken place. The plaintiffs have their own concerns; the expense of a complex jury trial could be enormous yet at the same time involves the risk of a defense verdict.
Together with these preliminary observations about D&O insurance and about the settlement process, I should also add the following observations on Professor Squire’s understandings and assumptions about D&O insurance.
First, Professor Squire assumes that corporate buyers consciously and deliberately structure their D&O insurance in multi-layered tiers because that creates “conflicts of interest among insurers that serve the interests of corporate managers,” as, he postulates, the conflicts “actually make insurance-covered settlements more likely.” There may be some buyers out there who are so canny that they manage their insurance buying decisions with these calculations in mind.
My own experience is that most insurance buyers would prefer to buy fewer, larger layers of insurance, because that would afford claims administrative simplicity, eliminate the headaches that always arise in the claims context when the insured company has to fight its way through multiple layers, and would simplify the insurance acquisition process.
The reason that D&O insurance programs are layered is because of the carriers’ preferences, not the buyers’ preferences. No D&O insurer could sustain the concentration of risk that would be involved with exposing outsized amounts of capital to any single large corporate exposure. Professor Squire assumes away this concern by saying that insurers could issue a single policy and simply “protect themselves from this risk through reinsurance.” However, there are a finite number reinsurers and they require a spread of risk every bit as much as the insurers do. The layering of D&O insurance is an inevitable by-product of an insurance marketplace where any given company’s insurance needs exceed the ability of any one insurer (or even one insurer and its reinsurers) to absorb all of the concentrated exposure associated with one risk.
Professor Squire also sees something amiss with the fact that D&O insurance policies cover both defense expenses and indemnity amounts. From my perspective, this arrangement is a natural reflection of the outgrowth of D&O insurance as a way for companies to contractually manage their indemnification obligations. In any event, the typical insurance buyer would consider it a strange notion indeed that their D&O liability insurance would not provide defense expense protection or at least that they would have to buy a defense cost policy separate from their liability policy.
With respect to Professor Squire’s analyses of the settlement dynamic, I note that in all of Professor Squires hypothetical settlement examples (and all of his examples are in fact hypothetical), he refers to the “expected trial liability” or “the actuarially fair settlement amount.” These are abstractions. There is no such thing in the securities context as “expected trial liability,” simply because there is effectively no data to describe such a thing. (For the same reason, talk of a party’s “bias to toward trial” or “aversion to trial” is equally inapposite.) By the same token, the idea that there is an objectively knowable amount equivalent to an “actuarially fair settlement amount’ is equally unwarranted.
What we actually have is a much rougher, much more approximate concept that usually is described as “what cases like this settle for.” Basically every single person in the settlement room will have their own version of this number, as well as their own concept of what features of the case make the comparables relevant. These points of reference are subjective, not objective; rarely the subject of agreement between the parties, at least at the outset; dependent on a host of assumptions; and allow for a range of possible outcomes that might constitute reasonable case resolutions depending on the myriad of factors.
In addition, the procedural posture of the case matters; the existence of related pending actions matter; the level of defense expenditure and the anticipated defense expense burn rate matters; the extent to which the defendants’ interests are aligned matters; whether some insurers believe they have unique coverage defenses matters; these and many other features matter and will influence the settlement dynamic and possibly affect the outcome of negotiations.
My point is that the settlement dynamic is complex and multifaceted. The very idea that settlement negotiations can be reduced to a mathematical formula with a few variables that explain outcomes in all cases – or even one case – is an interesting theory, as is the idea that settlement outcomes might be measured against a single, knowable measure that represents the “fair” outcome. The idea that settlement of the class action could be taken in isolation from the myriad other factors – for example, the existence of other related pending proceedings – is an interesting hypothesis.
Turning to the specifics of Professor Squire’s proposal that settlements should be segmented rather than collective, I have the following observations. It is possible that the alternative settlement arrangement Professor Squire has proposed could work to produce settlements that are fairer to all participants and that could eliminate the many ills that Professor Squire has identified. I suspect his proposal would be particularly attractive to the carriers that are most routinely in the primary and first level excess layers. It would less attractive to the insurers that are most often in the upper excess layers. And it would be extremely unattractive to policyholders.
First, with respect to the upper level excess participants, if all defense side participants have to negotiate their own settlements, there could be an unfortunate dynamic that forces the excess to the settlement table and forces them to pony up money that they otherwise shouldn’t have to pay and wouldn’t pay now. This effect, the direct result of reducing the protection to the excess carrier from the underlying layers, would increase the loss costs of the excess insurers. This would likely make excess insurance more expensive and partially or entirely offset any savings that might be available if primary and lower level excess carrier loss costs were reduced. In other words, to eliminate the cramdown effect, the segmented settlement process would substitute a “cram-around” effect, or maybe a “cram-up” effect.
Second, the possibility of a settlement holdout will not be eliminated. But the process when there is a holdout will change. Let’s consider the possibilities. Assume that all of the other participants have settled except one middle layer insurer. Sure, there would be a lot of pressure on the holdout. But there would be enormous pressure on the policyholder, too. Every additional dollar of defense expense incurred will reduce the sole remaining layer, forcing the policyholder toward a possible trial with ever-dwindling amounts of defense costs protection and little or no insurance remaining for any judgments that might result. For all the reasons outlined above, the policyholder could never run this trial risk, and so could be forced to contribute to settlement to avoid the range of unpalatable outcomes. Indeed, crafty insurance players aware of this possible dynamic might become strategic obstructionists, in order to compel the policyholder to absorb settlement costs and possibly allow the obstructing insurer to negotiate a discount settlement deal.
My concern is that one almost inevitable outcome of the segmented settlement process that Professor Squire has proposed is that policyholders would be compelled to contribute more frequently and in much greater quantities toward settlement. To the extent I read Professor Squire’s proposal correctly, he fully anticipates this process effect. For example, he says “If the settlements were segmented, more of them would be paid for by corporations rather than by their insurers.” Indeed, as I understand his analysis, he reckons this as one of the positive factors supporting his proposal, because it would force more companies to recognize in their securities litigation loss costs in their financial statements.
However, if I am right about the way that this settlement dynamic would work out, the introduction of a segmented settlement process would simply substitute a different cramdown dynamic for the existing one. The new dynamic would be particularly pernicious as it would threaten to put the insurers in the position where they were jockeying to force loss costs elsewhere, including in particular onto their own insured. I am at a loss to see how this could ever be viewed as an appropriate or desirable arrangement of affairs between an insurer and a policyholder. No insurer interested in maintaining its reputation as a good corporate citizen would ever propose such a thing. No well-informed policyholder would ever propose such a thing, either.
Another practical concern is that requiring segmented settlements would prolong cases. Process participants jockeying for position would have every incentive to try to game the system, hoping that the eroding limits and the ongoing litigation burden will compel other participants, particularly the policyholder, to pony up to get rid of the case. Company management would be burdened and distracted by a complex multistage process that never seems to end and distracts them from the things they need to do to make their businesses successful. Litigation is bad enough as it is, but one settlement deal ends it. If multiple deals were required, it could go on and on and on… And all of these problems would be magnified enormously if there are other related procedural matters pending.
My most important objection to Professor Squire’s proposal is that it is commercially impractical. He simply asserts that “segmented settlements could easily be achieved contractually.” This statement lacks a connection to the insurance marketplace. The theoretical possibility that in the long run segmented settlements might lower insurance costs or reduce the number of lawsuits will have no meaning to an insurance buyer, when compared to the very real possibility that the segmented settlement arrangement would cause the buyer to have to absorb loss costs that would otherwise be covered. No company would ever agree to do it -- nor should any company ever agree to it.
All of that said, there unquestionably are serious ills with the current securities suit settlement process, many of which ills Professor Squire identifies in his paper. I am not sure I know how to remedy these ills. One possibility that Professor Squire considers and rejects is the possibility of a quota share arrangement, where the various carrier participants’ interests are arranged vertically, rather than horizontally. Under this arrangement, each carrier would share ratably in each dollar of loss costs, so the carriers’ interests in trying to save loss costs would be aligned in a way that would eliminate the conflict of interest problems Professor Squire identifies.
The shortcoming of the quota share approach is that it would be very difficult for all of the participating insurers to cede control to a single decision maker. In the absence of a single point of control, the claims process could be reduced to chaos. But on the plus side, there are multiple places in the insurance world now where quota share arrangements already are in place and functioning successfully. There is a lot to be said for an insurance arrangement that already actually exists as a possible solution for an insurance-related problem.
I would like to thank Professor Squire for allowing me the opportunity to read his interesting and thought-provoking paper and for allowing me to comment about his paper here. I enjoyed reading his paper and writing this comment – it kept from going crazy on a very long plane flight. I hope that readers of this blog will review the Professor’s paper on their own and will post their thoughts and comments about his paper here using this site’s comment feature.
And in Other News: With a long plane flight on Saturday, I not only had the leisure to type out this long blog post, but I also had the opportunity to read the Financial Times weekend edition at length and in full. With a more thorough reading than I am usually able to enjoy, I gleaned a couple of very interesting observations from articles in the paper.
First, in a April 14, 2012 article entitled “Unorthodox Behavior Rattles Russian Church” (here), the article’s author notes the following about Kirill I, the patriarch of the Russian Orthodox Church: “Last week bloggers discovered that a photograph of the patriarch on the Church’s website had been altered to remove a $30,000 Bruguet watch from the churchman’s wrist. While someone had used Photoshop to erase the offending object, they had forgotten to erase the watch’s reflection on a nearby mahogany table.”
And in an April 14, 2012 article entitled “Who Needs the Shard When You Have Shakespeare?” (here), commenting on the latest addition to the London Skyline (now under construction), the article’s author notes that
The skylines of European cities have traditionally told us something about what they believe in, and they still do. Until Harold Macmillan eased restrictions on buildings exceeding the height of St. Paul’s, London’s skyline consisted of two 100m-high buildings: St. Paul’s and the Big Ben clock tower. It was easy to see what London stood for: the Church and parliamentary sovereignty and not necessarily in that order. The skyline of modern London, with its dozen or so buildings more than 100m-tall, also sends a message. It is best summed up by the London Eye, which proclaims the city less a workplace than a tourist attraction. The rest of the buildings are about money.
On the Road Again: This upcoming week I will be a panelist at a session in Beijing co-sponsored by the American Bar Association Torts and Insurance Practice Section and by the China Council for the Promotion of International Trade. Information about the conference, which is entitled “Doing Business in the United States: What you Need to Know about Investing, Product Liability and Dispute Resolution,” can be found here. I will be on a panel entitled “Directors and Officers Liability, Securities Issues and Class Action Exposure” with my good friend Perry Granof, as well as Patrick Zeng of Zurich China, whom I am looking forward to meeting for the first time on this trip.
From Beijing, I will be going on to Professional Liability Underwriting Society (PLUS) Chapter events in Hong Kong and Singapore. Perry Granof will also be joining me educational sessions at these PLUS Chapter Events, as will my old friend Joe Monteleone. To those readers who may be attending the PLUS Chapter events, I am looking forward to seeing you. If we have not met before, I hope you will please take the opportunity to introduce yourself.
With the distances, time differences and commitments involved, I am not sure what sort of publication schedule will be possible over the next few days. The D&O Diary’s normal publication schedule should resume the week of April 30, 2012.