«JOURNAL OF LAW, ECONOMICS & POLICY VOLUME 10 SPRING 2014 NUMBER 2 EDITORIAL BOARD 2013-2014 Steve Dunn Editor-in-Chief Crystal Yi Meagan Dziura Sarah ...»
While at first blush it appears that the defendant simply ought to litigate all matters, it can be shown that under certain conditions a defendant’s optimal strategy is actually to settle with all plaintiffs—both legitimate and frivolous ones. As a result, previous literature on the topic of frivolous lawsuits suggests that in some cases defendants are unable to avoid paying off frivolous lawsuits. Settlement payoffs to frivolous plaintiffs are simply a fixed cost of engaging in a risky activity.
Contrary to the assertion that defendants are “defenseless” to frivolous lawsuits, it is apparent that potential injurers often can affect their incidence. Under standard economic models of torts, for instance, potential injurers cause accidents not only as a consequence of their choice of care, but also as a direct result of their activity level—how frequently they engage in the risky activity in question. Thus, in the extreme case, a particular (defendant) business could avoid frivolous lawsuits altogether by simply choosing not to engage in any activity at all, i.e., by shutting down. More generally, it presumably can reduce the threat of frivolous claims by reducing its activity level. For example, consider the hypothetical case of a supermarket owner who causes harm to a subset of customers who slip and fall on his premises.53 The owner will face legal claims from customers who are genuinely injured as well as those customers who are either uninjured or suffered injuries caused by a different source. In this case, the success of frivolous claims in securing a pretrial settlement depends on the existence of genuinely injured customers. Absent genuinely injured customers, frivolous plaintiffs would not likely succeed in obtaining settlement, and hence, they would rarely (if ever) initiate a lawsuit. This scenario reflects the idea that frivolous lawsuits piggyback on legitimate lawsuits filed against individuals or businesses whose ordinary activity results in 52 For the purposes of our theoretical model, we view the terms “frivolous” and “nuisance” in the context of lawsuits as synonymous. However, at least one commentator has discussed a technical distinction between these two terms. See Lance P. McMillian, The Nuisance Settlement “Problem”: The Elusive Truth and a Clarifying Proposal, 31 AM. J. TRIAL ADVOC. 221, 222–23 (2007) (arguing that the two terms are not interchangeable, since a nuisance lawsuit implies a filing is made in bad faith, while a frivolous lawsuit does not). This dissimilarity is immaterial to our analysis.
53 Katz also considers slip-and-fall accidents in his discussion of frivolous lawsuits. Katz, supra note 20, at 6.
some accidents.54 We will often refer to this hypothetical scenario to stress important aspects of our theoretical model.
Consistent with the preceding logic, we formally define piggyback
lawsuits to be those brought by the following:
(1) actual accident victims whose injuries were caused by someone other than the defendant (for example, negligence by another customer of the supermarket);
(2) actual accident victims whose injuries were caused by “nature” (for example, a fall caused by a sudden dizzy spell suffered by the victim);
or (3) uninjured plaintiffs (those feigning a fall).55 This taxonomy of frivolous-as-piggyback lawsuits provides a straightforward definition upon which we can evaluate the impact of frivolous lawsuits on deterrence. In addition, it is consistent with the current jurisprudence governing frivolous lawsuits. Under our taxonomy, a defendant injurer is never liable at trial to a piggyback victim,56 either because the defendant injurer did not factually cause the victim’s injuries or because the victim is truly uninjured. In other words, although the above categories consist of a mixture of genuinely injured and uninjured plaintiffs, as a matter of law they are all frivolous in the sense that, even under a rule of strict liability, the injurer would not be held liable for their damages in court.
This view is harmonious with how courts routinely interpret the many policies and procedures providing a remedy to defendants when fending off frivolous litigation.57 Under American jurisprudence, for a case to be frivolous it must be meritless—meaning, according to the United States Supreme Court, “groundless or without foundation, rather than simply that the plaintiff has ultimately lost his case.”58 Accordingly, courts are hesitant to provide a remedy for the filing of a frivolous lawsuit in cases that exhibit even an iota of merit. For instance, in Tancredi v. Metropolitan Life Insurance Co.,59 the Second Circuit held that a finding of frivolity, which would trigger an award for attorney’s fees under the Civil Rights Attorney’s Fees 54 Our treatment of frivolous lawsuits as piggyback lawsuits should not be confused with lawsuits that piggyback on governmental or regulatory investigations.
55 Thomas J. Miceli & Michael P. Stone, “Piggyback” Lawsuits and Deterrence: Can Frivolous Litigation Improve Welfare?, 3 (Univ. of Conn. Dep’t of Econ. Working Paper Series No. 2013-16, July 2013), available at http://www.econ.uconn.edu/working/2013-16.pdf.
56 That is, the probability of victory at trial for a piggyback victim is zero.
57 See infra Part III for a discussion of the legal regimes that potentially affect the frequency of frivolous lawsuits.
58 Christiansburg Garment Co. v. Equal Emp’t Opportunity Comm’n, 434 U.S. 412, 421 (1978) (in the context of interpreting Title VII of the Civil Rights Act of 1964).
59 Tancredi v. Metro. Life Ins. Co., 378 F.3d 220 (2d Cir. 2004).
2014] THE IMPACT OF FRIVOLOUS LAWSUITS ON DETERRENCE 311 Awards Act,60 was improper when the original complaint was “very weak, but it was not completely without foundation.”61 Similarly, in Hughes v.
Rowe,62 the United States Supreme Court held, again in the context of the Civil Rights Attorney’s Fees Awards Act, that attorney’s fees should not be awarded to the prevailing party when “[a]llegations that, upon careful examination, prove legally insufficient to require a trial are not, for that reason alone, ‘groundless’ or ‘without foundation’.... The fact that a plaintiff may ultimately lose his case is not in itself a sufficient justification for the assessment of fees.”63 Because courts are hesitant to treat cases with some merit as frivolous in nature, there is symmetry between our taxonomy and current jurisprudence. Under our taxonomy, all piggyback cases are wholly without merit in the sense that the defendant is never liable to a piggyback plaintiff under a rule of strict liability.
In the next section, we provide an overview of our theoretical model, which partially follows the model originally set forth by Katz.64 However, our model extends Katz’s original analysis by grafting it onto a standard economic model of torts65 to determine how defendants’ optimal strategy for dealing with piggyback lawsuits affects their prior choices of care and activity. In this way, we focus on how frivolous-as-piggyback lawsuits affect deterrence. We illustrate that, under certain conditions, these suits create incentives for potential injurers to engage in more efficient accident avoidance. That is, under a particular set of assumptions, piggyback lawsuits may actually enhance deterrence in a socially valuable way.
II. THEORETICAL FRAMEWORK FOR EVALUATING THE EFFECT OF
PIGGYBACK LAWSUITS ON DETERRENCEIn this section, we set forth a theoretical model to illustrate how piggyback lawsuits affect accident avoidance. First, we provide a broad overview of the structure of the game, including its underlying assumptions.
Next, we analyze the resulting settlement–trial equilibria. It is shown that due to the existence of information asymmetry, defendants face a dilemma when attempting to formulate a pretrial settlement policy. In particular, when defendants try to settle all cases, they permit piggyback plaintiffs to extract positive settlements, which result in increased settlement costs.
However, when defendants litigate all cases, they avoid settling with piggyU.S.C. § 1988 (1976).
61 Tancredi, 378 F.3d at 230.
62 Huges v. Rowe, 449 U.S. 5 (1980).
63 Id. at 15–16.
64 See Katz, supra note 20, at 6–8 (enumerating the assumptions of the model).
65 See, e.g., STEVEN SHAVELL, ECONOMIC ANALYSIS OF ACCIDENT LAW (1987) (for an introduction to the economic analysis of tort law).
back plaintiffs, but they incur litigation costs which otherwise could have been avoided by settlement. This dilemma leads to the emergence of two distinct equilibria in the settlement–trial subgame. From these two equilibria, we then derive the prior care and activity choices of the defendant-injurer. We first show that piggyback lawsuits serve to enhance deterrence compared to a world without frivolous suits, and then go on to show that under certain conditions this enhanced deterrence is socially desirable.
Finally, we provide a numerical example to illustrate the key conclusions of the analysis. Interested readers can find the formal details of the model in the Appendix.66
A. Structure of the Game and its Underlying Assumptions
Our model is a sequential move game67 consisting of two types of players: (i) an injurer who engages in a risky activity, and (ii) potential victims who are either legitimately injured (one such victim being labeled as a “genuine victim”), or are of the piggyback variety (one being a “piggyback victim”). There are four total periods. In the first period, the injurer chooses its activity level. In the context of our hypothetical scenario, the injurer’s activity level reflects the number of individual stores the supermarket owner wishes to operate. If the injurer’s activity level is positive—that is, if the owner operates at least one store—then the injurer also chooses its monetary expenditure in care per unit of activity. For instance, the supermarket owner takes steps to ensure that the aisles are clear of hazards and the floor is not slippery.
The injurer’s monetary expenditure on care directly influences its expected tort liability per unit of activity. Specifically, the probability of an accident is decreasing in care, meaning that as the injurer’s expenditure on care increases, the probability of an accident decreases. However, this does not imply that the injurer will necessarily find it desirable to invest in care to the point where the probability of an accident is zero. Care is costly and the rate by which it reduces the probability of an accident decreases as care increases.68 Thus, an injurer will only increase its expenditure on care as long as the marginal benefit of care exceeds the marginal cost. Under strict liability, the injurer’s marginal benefit of care is the marginal reduction in expected tort liability per unit of activity. Thus, the injurer will continue to invest in care until the marginal reduction in liability equals the cost of the last unit of care (the marginal cost of care). In a simple model without litiThe appendix is based on the analysis in Miceli & Stone, supra note 55.
67 Such a game permits sequential movement by players, i.e., one player moves first, then another chooses an action based on the first player’s observed action.
68 This is a standard assumption under economic tort models. We particularly assume that the probability of an accident decreases as care increases, but at a decreasing rate.
2014] THE IMPACT OF FRIVOLOUS LAWSUITS ON DETERRENCE 313 gation costs, this level of care coincides with the socially optimal level of care. Under a more realistic scenario incorporating costly litigation, however, injurers will generally underinvest in care for the reasons noted in the Introduction. It is this underdeterrence that potentially makes piggyback suits socially desirable from an accident-cost perspective.
In addition to the number of genuine victims, which depends on the injurer’s care level, we assume that there is an exogenous “supply” of piggyback victims per unit of the injurer’s activity. The total number of potential plaintiffs facing the defendant, per unit of activity, is the sum of these two quantities.
Would-be victims—be they genuine or piggyback—move in the second period. Victims choose whether or not to file suit, which is costly. If a victim files suit, he becomes the plaintiff in a cause of action against the injurer, which is now the defendant. We assume that it is always in the interest of genuine plaintiffs to file suit, whether or not they expect subsequent bargaining with the defendant to result in a pretrial settlement. The decision of piggyback plaintiffs, however, depends on whether or not they expect the defendant to offer a positive settlement amount.
Once a victim (now a plaintiff) files suit, the defendant and plaintiff engage in a subgame involving the decision of whether to settle the dispute or proceed to trial. Thus, in the third period, the defendant makes a take-itor-leave-it settlement offer (which could be zero) to the plaintiff, and then in the fourth period, the plaintiff either accepts or rejects this offer. If the plaintiff accepts, the parties settle and the game ends. However, if the plaintiff rejects the settlement offer, then the outcome of the game depends on whether the plaintiff is a genuine or a piggyback plaintiff. For simplicity, we assume that the prevailing legal rule is strict liability with pure compensatory damages, meaning the injurer will always be liable for a genuine victim’s harm, since by assumption a genuine victim can prove causation in court. Thus, if the plaintiff is genuine, it will win at trial with certainty.69 In contrast, a piggyback plaintiff cannot prove causation and will therefore lose at trial with certainty. Accordingly, only genuine plaintiffs ever proceed to trial, while piggyback plaintiffs who reject the defendant’s offer (which only happens when the offer is zero) withdraw their claims.
The equilibrium of the preceding game is derived by backward induction,70 which is the proper solution procedure for games in which players move sequentially. The Appendix provides a formal analysis of the model.
69 Assuming that judges and juries do not commit errors.