In The Curious Incident of the Falling Win Rate, Alexandra Lahav and Peter Siegelman highlight a remarkable—but heretofore overlooked—fact: Between 1985 and 1995, the plaintiff win rate in adjudicated civil cases in federal courts fell precipitously, from 70 percent to 30 percent. In subsequent decades, although the plaintiff win rate has fluctuated, it has generally hovered at or below 40 percent, significantly off its 1985 peak.1 (P. 1371.) From there, Lahav and Siegelman put their discovery in context and explore potential explanations for the observed trend. In this Jot, I’ll endeavor to explain why their Essay—on the face of it, not about the tort system—in fact, supplies an important piece of the tort-decline puzzle. Then, I’ll offer a fuller account of the Essay itself. Finally, I’ll share a few questions the paper stimulates.
First, how is this Essay about the civil justice system in general—and the federal civil justice system, no less—a tort law piece? It’s a fair question since only around 2 percent of tort claims are litigated in federal court. But, the tort system has long been the most controversial corner of the civil justice system, and it’s still one where myths and misinformation abound.
Americans continue to believe that the system is “choked” with tort claims and that there’s a pot of gold waiting, even for undeserving plaintiffs. Yet, over the past two decades, the evidence has accumulated to paint a very different portrait. We know that tort filings are way down. According to the Wall Street Journal, in fact, we know that, in the state system, there is a “nationwide ebb in lawsuits.” Whereas, in 1993, about ten in 1,000 Americans filed tort lawsuits annually, as of 2015, that number was fewer than two in 1,000. “Tort cases declined from 16% of civil filings in state courts in 1993 to about 4% in 2015, a difference of more than 1.7 million cases nationwide.”2 Supplying information from the federal system, here, Lahav and Siegelman add to our understanding. Comparing federal tort suits initiated in 1985 and 2016, their data reveal a stunning drop: There were 29,655 such cases filed in 1985 but only 12,114 such cases filed in 2016, a decline of 69 percent.3
There’s some evidence (at least within the state system), that, when plaintiffs recover, their recoveries are lower than in prior years. According to a Bureau of Justice Statistics study, the median jury award in state court tort cases was $33,000 in 2005, whereas the median jury award in tort cases was $71,000 in 1992, in inflation-adjusted dollars—a drop of 53.5 percent.4
And, there is no question: Trials are also down sharply. The federal civil trial rate is 0.7 percent.5 One way to look at that: Federal courts conducted half as many civil trials in 2019 as they did in 1962, even as they disposed of over five times as many civil cases. A similar trend is apparent in the states. According to data maintained by the National Center for State Courts, in 2015, the percentage of civil cases resolved by jury trial ranged from .05 percent to .5 percent in the twenty-one jurisdictions studied.6
The final piece of the puzzle, of course, is plaintiff win rates. While limited to the federal system, Lahav and Siegelman help to bridge that gap by showing that, just as filings are down, damages are down, and trials are down, plaintiff win rates are dropping, too. As noted above, they show that, generally, plaintiff win rates in civil cases plummeted starting in 1985. And, zeroing in on tort cases in particular, their data reveal a sharp fall in plaintiff win rates from 1985 through 2016. Of the federal tort lawsuits initiated in 1985, the plaintiff win rate was 38.1 percent. Of those initiated in 2016, the rate was 11.9 percent. Even though fewer lawsuits were initiated over that three-decade period (suggesting, possibly, that lawyers had become more selective in their filings), the tort-specific federal court plaintiff win rate was slashed more than in half.
Even if the piece did nothing more than highlight those core findings, it would make a very large contribution, to tort scholars and beyond. Yet, the authors do more than merely surface this surprising trend. They also seek to explain it. Seeking to address why the general plaintiff win rate has dropped so sharply, they suggest that most of the observed decline (perhaps 60 percent) is attributable to shifting inputs, to “changes in the makeup of the federal caseload” during the period. (P. 1374.) Over time, they find, cases in subject-matter areas where plaintiffs tend to fare poorly (e.g., prisoner-rights litigation) became more prevalent, even while categories of suits in which plaintiffs tend to fare well (e.g., suits to enforce student loan obligations) became less prevalent or were eliminated altogether. But, even if that’s right, as Lahav and Siegelman note, that still leaves the remaining 40 percent of the drop unexplained.
This 40 percent may be attributable to any number of factors, which the authors proceed to explore, and for the most part reject, as either belied by the available data or inconsistent with rational litigant behavior. For example, one might hypothesize that a chunk of the reduction is attributable to selection effects, including plaintiffs filing a higher proportion of junk cases, plaintiffs litigating (rather than settling) a higher proportion of junk cases, and/or “selective settlement of winning cases that would previously have been litigated.” (P. 1387.) But, as Lahav and Siegelman observe, none of these hypotheses are clearly supported by available evidence.
So, too, it could be that the composition of plaintiffs, themselves, or defendants, themselves, has changed—which the authors can’t entirely rule out—but which, again, is hardly self-evident. Finally, drawing on the work of Steve Burbank, Sean Farhang, and others, the authors consider—but once again, mostly scrap—the possibility that “judicial attitudes towards plaintiffs” may have changed during the period, “making judges more skeptical of plaintiffs’ claims.” (P. 1408.)
From there, Lahav and Siegelman set forth the data that would be needed in order to better unravel the mystery. They survey the relevant theoretical and empirical literature (with particular attention to the classic work by George Priest and Benjamin Klein).7 And, in the paper’s final part, they draw out a normative claim: “[T]he justice system,” they assert “is obligated to explain suspicious developments in the administration of justice at the systemic level.” (P. 1375.)
I find Lahav and Siegelman’s discussion learned, persuasive, and extraordinarily thought-provoking. In terms of thoughts that spring to my mind, for starters (though, probably unrealistically), I find myself hungry for more data, and, in particular, data stretching back further, so we can identify trends prior to 1985. (Could it be that the early 1980s were themselves anomalous and what we see is merely a reversion to the mean?)8
Further, when it comes to probing potential hypotheses, I can think of a few more possibilities that might be addressed and, if possible, tested. Regarding inputs, for example, when the authors discuss the possibility that some of the observed reduction might be explained by plaintiffs opting to file lower-quality cases, it would have been helpful to draw out the fact that the boundary between federal and state litigation is porous. Could it be that, in the mid-1980s, federal courts started to get a bad rap, so plaintiffs started to filter—to file more viable cases in state court, leaving longer shots for the federal system?
So, too, while the authors draw on Burbank’s and Farhang’s important work to discuss the political leanings of trial court judges, we might also interrogate the law handed down from on-high. We know that, during this period, Congress enacted numerous statutes that made life harder for plaintiffs. (The Private Securities Litigation Reform Act and the Prison Litigation Reform Act come to mind.) We know SCOTUS handed down a raft of defendant-friendly decisions—doing everything from tightening standing requirements, to limiting expert testimony, to eliminating aider and abettor liability, to expanding the preemptive effect of certain regulatory activity, to liberalizing Rule 56.9
And, last but not least, we know that significant tort reform measures—including noneconomic damage caps, contingency fee restrictions, alterations to the collateral source rule, and extensions of statutes of repose—were enacted in many states.10 Might these myriad changes explain some of the observed trends?
Finally, while I am fully persuaded by the authors’ plea for more and better data, I’m not wholly convinced of the authors’ more insistent claim that, whenever “system-level consequences . . . are inconsistent with the understandings of how the civil justice system should work,” (P. 1430) “the system” must offer a “justification that is consistent with rule of law values.” (P. 1370.) The justice system, of course, is a “they” not an “it,” and it’s not clear to me who, within the messy and fragmented system, should bear such responsibility. Furthermore, when judging whether outputs are inconsistent with observers’ “understandings,” it’s not clear to me whose understandings should prevail.
Quibbling aside, the fact remains: Lahav and Siegelman document and highlight a phenomenon that had been hiding in plain sight. And they do so in a paper that is nuanced, technically sophisticated, and richly generative. Scholars—from tort and beyond—will be drawing on Lahav and Siegelman’s “curious” and critically important discovery for years to come.
- The authors rely on the Administrative Office (AO) of the U.S. Courts Civil Terminations dataset. They examine those adjudicated cases where the AO has coded the judgment as a win for either the plaintiff or defendant. An “adjudication” refers to any decision rendered by a court that terminates a case.
- Joe Palazzolo, We Won’t See You in Court: The Era of Tort Lawsuits Is Waning (July 24, 2017).
- Lahav and Siegelman generously ran these numbers for me, for purposes of this Jot. Those “tort” cases evaluated include those coded: tort product liability, motor vehicle personal injury, other personal injury, medical malpractice, personal injury product liability, and asbestos product liability.
- Lynn Langton & Thomas H. Cohen, Bureau of Justice Statistics, Civil Bench and Jury Trials in State Courts, 2005, at 10 (2009). These numbers come from trials conducted in the nation’s seventy-five most populous counties. Contract awards, by contrast, increased 19.5 percent during this period.
- Table C-4, U.S. District Courts—Civil Cases Terminated, by Nature of Suit and Action During the 12-Month Period Ending September 30, 2019.
- 2015 Civil Caseloads—Trial Courts, Nat’l Ctr. for State Courts (select “2015” under toolbar labeled “Select the Data Year,” then select “Civil Jury Trials and Rates” from the toolbar labeled “Select Chart/Table”).
- George L. Priest & Benjamin Klein, The Selection of Disputes for Litigation, 13 J. Legal Stud. 1, 4-5 (1984) (suggesting that there should be a tendency towards a plaintiff win rate of 50 percent).
- Notice here: The authors focus on the period from 1985 through 1995, but, based on their Figure 1, it appears that the period from 1982 through 1985 featured a rising plaintiff win rate, rather than a declining one.
- Lahav and Siegelman do highlight the 1986 “Celotex Trilogy,” 52 U.C. Davis L. Rev. at 1411 n. 79, but they downplay, and might too quickly dismiss, the trilogy’s significance. For the current role of summary judgment in the federal system, see Nora Freeman Engstrom, The Lessons of Lone Pine, 129 Yale L.J. 2, 68 & n. 297 (2019).
- For a catalog of various reforms enacted in 1985 through 1987 alone, see W. Kip Viscusi, The Effect of 1980s Tort Reform Legislation on General Liability and Medical Malpractice Insurance, 6 J. Risk & Uncertainty 165, 169-71 (1993). Many of these reforms were “substantive” and would therefore apply in federal courts sitting in diversity.