Highlights

This study offers new data on the frequency and the effects of punitive damages, based on a detailed review of more than 1000 lawsuits filed and concluded in San Francisco County Superior Court.

Punitive damages are demanded in 27 percent of all cases where they are conceivably recoverable.

Lawsuit filing data show that business and government defendants are four times as likely as an individual defendant to face a lawsuit that demands punitive damages.

Lawsuits that include punitive damage demands take one-third longer to resolve than suits without these demands. The average lawsuit in our 1000 case sample took 15 months to resolve; cases with punitive damage demands took an average of 21 months to resolve.

Punitive damage demands play a significant role in the out-of-court settlement process, where the vast majority of lawsuits are settled. Punitive damage demands tilt the playing field in favor of demanding parties, and increase out-of-court settlement amounts.

Studies of punitive damage jury verdicts have been interpreted to suggest that the risk of receiving an adverse punitive damage judgment is remote. Closer scrutiny of the data, we argue, will show that the probability of punitive damage awards is vastly understated by these studies, in part because the data have been improperly qualified.

Introduction

The controversy over punitive damages in civil litigation has centered around the number of punitive damage awards, and the dollar amounts of such awards. Comprehensive data on this issue are scarce because there is no complete statistical database of trial verdicts. However, we believe that the focus of research on the number of punitive damage verdicts is misplaced to some extent. Focusing only on trial verdicts understates the scope and nature of the problem because the overwhelming majority of all lawsuits are resolved out of court. According to surveys of lawsuits, less than 2 percent of all cases go to trial. Looking only at the 2 percent of cases that reach a jury is like looking only at the visible tip of a large iceberg: it ignores the larger unseen part below the water line that may do more harm. Thus, to argue that punitive damage awards are rare is to miss an obvious point: jury verdicts of any kind are rare. No one would say, however, that because jury verdicts are rare, lawsuits themselves are insignificant or costless. Yet this is the inference that has been drawn from various punitive damage studies that focus only on trial verdicts.(1)

The right question to ask about the civil litigation process is: what is occurring in the other 98 percent of cases that are resolved out-of-court? Because 98 percent of cases are resolved out of court, it is important for researchers and for public policy makers to understand what is going on in these cases, and how the legal rules, including the relative probability of punitive damages, affect the outcomes of the negotiation process for out-of-court settlements.

A large proportion of lawsuits today include punitive damage demands. Because 98 percent of lawsuits are resolved out-of-court, an important threshold question to answer is: Do punitive damage demands in lawsuit filings have a significant effect on the out-of-court settlement process? In an attempt to shed light on this question, the Pacific Research Institute conducted a detailed examination of more than 1000 lawsuit filings in San Francisco County Superior Court. We conclude that punitive damages are used as a weapon to generate more favorable out-of-court settlements, especially against business and government defendants.

What Lawsuit Filings Data Tell Us

Most punitive damage studies focus on verdicts in the handful of cases that proceed to trial. There are no empirical studies that examine how often, and against whom, punitive damage demands are employed.

In an attempt to shed light on these issues, we have scrutinized a month's worth of lawsuits filed between January 2, 1991 and February 1, 1991 in the Civil Division of the California Superior Court for the County of San Francisco. There were 1,024 lawsuits filed in this venue during this time period. Our analysis seeks to determine what patterns exist with respect to the distribution of claims for punitive damages and statutorily mandated multiple damages. We reviewed each case to determine the type of case, the principal cause of action claimed, whether punitive damages were demanded, how the case was resolved, and how long it took to resolve. We picked cases from 1991 because over 98 percent of cases from this time period had been resolved, either by trial, settlement, or dismissal, so most could be traced out to their conclusion. (A complete description of the methodology used in surveying and classifying these cases is available upon request.)

The highlights of the analysis of these cases include:

• 78 percent of all punitive damage demands were filed against a business defendant. (See Figure 1 below.)
• Government defendants face punitive demands in more than one-third of lawsuits filed against government agencies. (See Table 1 below.)
• Lawsuits that include punitive damage demands take about six months longer to resolve than lawsuits that do not include punitive damage demands.
• The probability of a punitive damage award if a case proceeds to trial is 14 percent or higher. For business defendants, the probability is more than 20 percent.

Of the 1,024 cases we examined, nine (9) cases were designated by the court as "sealed" and were unavailable for study. This left a pool of 1,015 lawsuits for study. 537 of these 1,015 cases, or 52.9 percent, were based on one of three principal causes of action where punitive damages are recoverable: 1) common law tort;
2) statutory tort or other statutory civil action; or
3) contract.*

The remaining civil filings within this pool-478 cases-consisted of civil law categories where punitive damages are not available or only very rarely available. These categories include equity causes of action (such as requests for a temporary restraining order); family law matters (chiefly dissolution of marriage proceedings); and civil petitions (such as a name change, to confirm an arbitration award, or to compel an audit entry).

Punitive damage demands were included in 145 of the lawsuits. This represents only 14 percent of the total pool of lawsuits, but 27 percent of the suits in areas where punitive damages are generally available (145 out of 537). Of these 145 filings, 112 or 78 percent were filed against a business defendant. This breakdown is displayed in Figure 1 below.

Figure 1: Distribution of Punitive Damage Demands by Type of Defendant

Although punitive damages are not normally recoverable in contract lawsuits, many contract suits nowadays include secondary causes of action, such as fraud or "bad faith," which can suffice as a basis for punitive damages. For this reason, contract cases have been included in the pool of cases which can potentially involve a viable claim for punitive or statutory damages.

Of the 1,015 cases reviewed, only 22 went to trial, or 2.1 percent. This is comparable to the trial rate estimated in the Department of Justice study, and is consistent with most studies of lawsuits. Seventeen of these 22 cases were tort, statutory, or contract cases (in which punitive damages may be alleged). Of these 17 cases, seven were court trials (heard by a judge instead of a jury) and 10 were jury trials. Business entities were the primary defendant in 11 of the 17 trials; government entities and individuals were named as the primary defendants in three trials. Six of the 17 cases (two court trials, and four jury trials) included punitive damage demands. Two of the 17 cases settled during trial. Of the remaining 15 cases, plaintiffs won in seven of the trials, while defendants won eight. No punitive damages were awarded in these verdicts.

Table 1 categorizes cases according to the type of primary designated defendant (individual, business entity, government entity, and non-profit organization). Table 2 displays the same data according to type of case instead of type of defendant. The data in these tables demonstrate that punitive damages are overwhelmingly alleged against business entities. Businesses face punitive damage demands roughly four times as often as individuals.

Government Agencies Frequently Targeted

Another remarkable feature is apparent in Table 1: Government entities face punitive damage demands more than one-third of the time (35 percent). While defenders of punitive damages argue that punitive awards are necessary to prevent allegedly gross negligence, malice, or willful harm by businesses, such malicious behavior can hardly be just as frequently alleged in government. But government entities, like businesses, are perceived to have "deep pockets" and make attractive targets for punitive damage demands. Thus, taxpayers, and not just business interests, bear the direct cost of punitive damages. Many of the lawsuits brought against government, and defended at taxpayers' expense, are patently frivolous. For example, in one case from our research, Buford v. California Department of Real Estate, the plaintiff demanded $3 million in punitive damages because the agency had refused to grant him a real estate license. The Department of Real Estate had to answer the suit in court, pointing out that the plaintiff was a convicted felon.

Another notable suit was Shervin v. California State Police. Alleging that five state police officers had burst into his home without a warrant but on the direct orders of Governor Deukmejian, Shervin demanded punitive damages based on seven different allegations, including "organized conspiracy, racketeering and/or extortion, to subvert the court and to obstruct justice."

Additional allegations included "torture," and the accusation that the California State Police were engaged in "an extensive, continued and perpetual gigantic organized conspiracy" of an unspecified nature. This was the fourth such lawsuit Shervin had brought against various government agencies in recent years, all defended at taxpayer expense.

Table 1: Civil Litigation Filings in Common Law Tort, Statutory Tort, and Contract Cases, and Punitive or Statutory Damage Claims Frequency by Category of Primary-Designated Defendant, 1991

Table 2: Distribution of Punitive and Statutory Damages Claims Among Civil Flilings by Category of Lawsuit, 1991

Duration: Punitive Cases Take Longer to Resolve

Another significant finding of this sample of lawsuit filings concerns the average duration of the cases. We assigned a duration value to each case in the sample. The purpose of this analysis was to gauge the length of time until each case was resolved, and to see whether lawsuits that included punitive damage demands were concluded more quickly or less quickly than lawsuits that did not demand punitive damages. Our analysis of the duration of cases showed that lawsuits that did not include a punitive damage demand were concluded in an average of 15 months, while punitive damage lawsuits required an average of 21 months to conclude-a six month difference.

It is difficult to know whether this distribution of case filings, punitive damage demands, and average duration to resolution holds constant in the case filings in other jurisdictions. But for purposes of illustration only, if this distribution were roughly similar in all the jurisdictions covered in the Department of Justice estimates, it would suggest that a business defendant facing a punitive damage demand would have about a 14 percent probability of receiving an adverse punitive damage judgment at trial-substantially higher than the 5.9 percent of trials that resulted in punitive damages in all of the trial verdicts considered as a whole in the DoJ study. It should be emphasized here that many of the 75 counties included in the DoJ study, such as Washington state counties, do not allow or severely restrict punitive damages, so the overall probability in areas allowing punitive damages is certainly much higher than 14 percent.

Asking the Right Questions: The Dynamics of Lawsuit Settlement

To appreciate fully the significance of the findings of our analysis of lawsuit filings, it is necessary to understand how punitive damage demands may affect the calculus of out-of-court settlement demands. A review of the scholarship about this subject will establish the following conclusions:

1. The unpredictability of a prospective punitive damage award contributes significantly to the uncertainty (and therefore the risk) of a court trial outcome.

2. Both the uncertainty posed by the prospect of unlimited punitive damages, combined with the relative probability of a punitive damage award if a case goes to jury trial, provide litigants who demand punitive damages with potent leverage against risk-averse defendants, and tip the balance in settlement bargains in favor of litigants with weak or even frivolous cases.

As mentioned previously, concentrating on trial verdicts overlooks "where the action is" in civil litigation: out-of-court settlements. We are not suggesting that verdicts are unimportant. To the contrary, punitive damage verdicts are like the tip of the proverbial iceberg. The small number of trials affect decisions in the vast majority of lawsuits that do not proceed to trial. Verdicts are "information signals" for litigants. Even Stephen Daniels and Joanne Martin, who are strong proponents of punitive damage awards, note that "jury verdicts in the minority of matters actually adjudicated play an important role in determining the worth, or settlement value, of civil matters filed but not tried."

To get a proper perspective on how this process works requires a consideration of the basic dynamics of a lawsuit. There is growing scholarly literature that offers several models of decision-making in the litigation process, especially in cases involving doubtful or even frivolous legal claims. "Situations involving litigation are a paradigmatic case of bargaining conflict," Kip Viscusi has written, and hence susceptible to illumination through game-theory and decision-tree models. As professors Robert Cooter and Daniel Rubinfeld of U.C. Berkeley have noted, "The attributes of litigation bargaining-rivalry, communication, side payments, interdependency, and uncertainty-characterize bargaining games as analyzed in microeconomics." This approach can help clarify the role of punitive damages in changing the calculus of settlement between litigating parties.

The first thing to understand about lawsuits under American law is that the plaintiff has the opening strategic advantage: even a plaintiff with a weak case places the defendant in the position of having to defend himself (and therefore incurring legal costs), or else the defendant will be liable for the full claim on a default judgment. Hence, even a defendant facing a suit without merit is often willing to pay an amount that is less than his prospective defense costs to settle the case and "make it go away." According to various studies, the cost of defense in an average tort lawsuit ranges from $6000 to $10,000, depending on the kind of suit. A litigant with even a mildly plausible basis for an average suit can often expect a nuisance settlement value within this range. Professors David Rosenberg and Stephen Shavell of Harvard Law School comment: "By filing a claim, any plaintiff, and thus the plaintiff with a weak case, places the defendant in a position where he will be held liable for the full judgment demanded unless he defends himself. Hence, the defendant should be willing to pay a positive amount in settlement to the plaintiff with the weak case-despite the defendant's knowledge that were he to defend himself, such a plaintiff would withdraw." University of Michigan economist Avery Katz adds that in many cases "the defendant is willing to pay a settlement up to the amount of his defense costs in order to avoid having to respond to the plaintiff's complaint."

The main determining factor of whether a filed lawsuit will yield a settlement to the plaintiff is the "threat credibility" of the suit, i.e., what is the probability of a verdict favorable to the plaintiff if the case goes to trial, and what is the likely amount of damages that the plaintiff could win? The scholarly models of the out-of-court negotiation process suggest that an increase in the prospective amount of a jury verdict increases the likelihood of a settlement offer by the defendant, and tends to increase the amount of such settlements. Professors Kathleen Engelmann and Bradford Cornell argue that "it is almost invariably the case that increasing the cost of litigation increases the probability of settlement." Professor Barry Nalebuff of Princeton University concurs, noting that "an increase in the court award . . . raises the probability of settlement."

This can be true even in frivolous or marginal lawsuits, or lawsuits with a doubtful chance of success at a trial. Professor Katz comments: "The main reason that frivolous suits are not always met with a blanket denial and refusal to negotiate, of course, is that the defendant rarely knows the merits of the claim with certainty. Since refusing to take a valid claim seriously can be quite costly, a frivolous plaintiff may be able to take advantage of the defendant's uncertainty regarding the claim's validity to extract a substantial settlement." Moreover, Katz adds, "higher trial costs raise the defendant's benefit from settling with valid claimants and makes him more willing to tolerate the cost of settling with strike suitors."

The point is: punitive damage demands will often tip the balance of power in bargaining to the plaintiff, even one with a weak or frivolous case. It does so in two ways: by increasing the size of a prospective jury award (by an unpredictable and potentially enormous amount) if the case is taken to trial, and by increasing the legal costs that a defendant will have to incur to fight the suit at trial. First, to use a hypothetical example: while a $50,000 lawsuit with arguable merit might have a settlement value of $20,000 or $30,000, a $50,000 lawsuit that also demands $200,000 in punitive damages is no longer a $50,000 lawsuit for purposes of settlement. The presence of a punitive damage demand provides leverage for the plaintiff to force a higher settlement value from the suit. Second, the presence of a punitive damage demand often requires a more extensive, more costly, and more time-consuming defense by the defendants. Most punitive damage demands are based on claims of intentional wrongdoing or "conscious disregard" of the rights of the litigant. Defending against such extraordinary claims usually requires a more expensive discovery process than ordinary damage claims. In addition to a discovery process about the basic facts of the injury or fraud involved in the tort allegation, determining the malicious intent of the defendant will involve more extensive, and therefore more expensive, document searches and depositions.

The key dynamic of the out-of-court settlement process is uncertainty. Obviously if the outcome of jury trials were highly predictable, few if any cases would ever go to trial. The parties would always settle. It is the uncertainty of trial outcomes that has led to a thriving market for jury verdict data services, which are intended to provide at least some guidance to litigating parties to help estimate the risks of trial and the parameters of a reasonable settlement. Punitive damage demands add dramatically to the uncertainty of out-of-court settlement deliberations. The inclusion of a punitive damage demand increases the potential amount of an adverse jury award by an unpredictable degree, since punitive damages are unlimited.

To judge how serious a factor this is, it is necessary to consider the probability of receiving a punitive damage verdict if a case is taken to trial. The plaintiff's leverage is only effective if the threat of extracting punitive damages from a trial is credible. What makes a punitive damage demand credible in the eyes of a defendant? The studies that minimize the number of punitive damage awards are highly misleading on this point.

For example, the recent Department of Justice study's estimates on civil lawsuits seems to suggest that because so few cases result in punitive damages (364 out of 762,000 cases filed, or .0004 percent), the threat of a punitive damage demand in a lawsuit pleading is not very credible. But this is to miss something rather obvious in the DoJ statistics: because only 1.5 percent of the lawsuits actually went to trial, the relevant question to ask is: what is the risk of receiving a punitive damage judgment if a case is brought to trial? The 364 punitive damage verdicts in the DoJ estimates take on a new significance if pondered in this fashion: they amount to 3 percent of the 12,000 cases tried, or 5.9 percent of verdicts in which the plaintiff was the winner. Three percent, or even 5.9 percent, still may not seem very substantial to outside observers who do not bear the risk themselves, but it is crucial to remember, however, that many if not a majority of these cases that went to trial did not include a punitive damage demand as a part of their pleading. Secondly, the DoJ statistics do not tell how many of the suits that included punitive damage demands, or how many of the verdicts that included punitive damages, were suits brought against businesses as opposed to individuals. (The Department of Justice has the data to make this breakdown, but chose not to report it in its study.)

The disaggregated figures in Table 3 below, taken from the Department of Justice estimates, tell the story more clearly. Although the DoJ estimates are severely limited because they do not tell us whether the defendant in these verdicts is an individual or a business, and does not tell us what proportion of the cases that went to trial sought punitive damages, they do confirm that the probability of receiving a punitive damage award is significant. For example, in employment law cases (nearly all of which were brought against business defendants), we see that 26.8 percent of all verdicts included punitive damages, with a hefty median punitive award of $179,000.

Table 3: Punitive Damage Awards for Plaintiff Winners in Civil Jury Cases in State Courts in the Nation's 75 Largest Counties, 1992

Another point should be made from the DoJ estimates. While defenders of unlimited punitive damages prefer to use median punitive award figures because a few large awards can skew average award figures, this point can be turned on its head. The large disparity between median punitive award amounts and average award amounts ($50,000 and $735,000 respectively in the DoJ estimates) highlights the unpredictability of punitive awards. As our previous report on punitive damages in California showed, there was a huge range in punitive damages awarded between 1990 and 1994, demonstrating that punitive damages are unpredictable and arbitrary. In California cases during this period, the range of punitive awards runs from 710 times compensatory damages to .0001 times compensatory damages. (In one case, a defendant who was not assessed any compensatory damages was nevertheless hit with $62,000 in punitive damages.) It is precisely this uncertainty that provides the plaintiff with additional leverage in the settlement process.

Conclusions

The California statutes governing punitive damages use exceptionally strong language to prescribe when punitive damages are appropriate. Punitive damages should be awarded where there is "clear and convincing evidence" that a defendant has behaved with "malice," or has engaged in "despicable conduct which is carried on by the defendant with a willful and conscious disregard of the rights or safety of others." Other descriptions include "oppression, intentional misrepresentation, deceit, or concealment of a material fact. . ." Unless it is implausibly assumed that such extraordinary behavior is rampant and pervasive in California, the frequent appearance of punitive damage demands in lawsuit filings is evidence that they have become simply a regular litigation tactic.

It is not persuasive to say that the frequency of punitive damage demands have little or no effect on the cost and outcomes of litigation. The uncertainty and risk posed by potential punitive damage awards magnifies the leverage of such demands in out-of-court settlements. The prospect of "runaway juries" is far from fanciful. Even judges have felt compelled to speak out about this phenomena. In a recent California trial involving an employment dispute that resulted in an $80 million punitive damage award, the trial judge set aside the verdict, noting:

"This award is so disproportionate to the injuries, damages and conduct, and so unsupported by the evidence it shocks the conscience of this court to the point that the court cannot countenance such a result and feels compelled, despite its respect for the jury process, to grant a new trial . . . Punitive damages award is excessive and clearly motivated by passion and prejudice [of the jury]. The award does not bear a reasonable relationship to the nature of defendant's action and the extent of plaintiffs' injuries."

Despite the admonitions of California statutes that there be "clear and convincing evidence" of extraordinarily deliberate malicious behavior, it is clear that new guidelines and limitations on punitive damages are needed.

--By Steven Hayward, vice president, research for the Pacific Research Institute for Public Policy. William S. Loughman, an attorney and senior fellow in legal studies for the Pacific Research Institute, conducted the research into lawsuit filings.