abstract. This Article reports the findings of the largest and most comprehensive study to date of the role qualified immunity plays in constitutional litigation. Qualified immunity shields government officials from constitutional claims for money damages so long as the officials did not violate clearly established law. The Supreme Court has described the doctrine as incredibly strong—protecting “all but the plainly incompetent or those who knowingly violate the law.” Legal scholars and commentators describe qualified immunity in equally stark terms, often criticizing the doctrine for closing the courthouse doors to plaintiffs whose rights have been violated. The Court has repeatedly explained that qualified immunity must be as powerful as it is to protect government officials from burdens associated with participating in discovery and trial. Yet the Supreme Court has relied on no empirical evidence to support its assertion that qualified immunity doctrine shields government officials from these assumed burdens.

This Article is the first to test this foundational assumption underlying the Supreme Court’s qualified immunity decisions. I reviewed the dockets of 1,183 Section 1983 cases filed against state and local law enforcement defendants in five federal court districts over a two-year period and measured the frequency with which qualified immunity motions were brought by defendants, granted by courts, and dispositive before discovery and trial. I found that qualified immunity rarely served its intended role as a shield from discovery and trial in these cases. Across the five districts in my study, just thirty-eight (3.9%) of the 979 cases in which qualified immunity could be raised were dismissed on qualified immunity grounds. And when one considers all the Section 1983 cases brought against law enforcement defendants—each of which could expose law enforcement officials to burdens associated with discovery and trial—just seven (0.6%) were dismissed at the motion to dismiss stage and thirty-one (2.6%) were dismissed at summary judgment on qualified immunity grounds. My findings enrich our understanding of qualified immunity’s role in constitutional litigation, belie expectations about the policy interests served by qualified immunity, and show that qualified immunity doctrine should be modified to reflect its actual role in constitutional litigation.

author. Professor of Law, UCLA School of Law. For helpful conversations and comments, thanks to Will Baude, Karen Blum, Alan Chen, Beth Colgan, Richard Fallon, William Hubbard, Aziz Huq, John Jeffries, Justin Murray, Doug NeJaime, James Pfander, John Rappaport, Alexander Reinert, Louis Michael Seidman, Stephen Yeazell, and participants in workshops at The University of Chicago Law School, Duke University School of Law, Harvard Law School, and UCLA School of Law. Special thanks to Benjamin Nyblade, who calculated the statistical significance of my findings. Thanks also to Michelle Cuozzo, David Koller, Rosemary McClure, David Schmutzer, and the expert research staff at UCLA’s Hugh & Hazel Darling Law Library for excellent research assistance, and thanks to R. Henry Weaver, Arjun Ramamurti, Kyle Edwards, Kyle Victor, Erin van Wesenbeeck, and the editors of the Yale Law Journal for excellent editorial assistance.

This Article's dataset is available at: https://datadryad.org/stash/dataset/doi:10.5061/dryad.sbcc2fr3w.

Introduction

The United States Supreme Court appears to be on a mission to curb civil rights lawsuits against law enforcement officers, and appears to believe qualified immunity is the means of achieving its goal. The Supreme Court has long described qualified immunity doctrine as robust—protecting “all but the plainly incompetent or those who knowingly violate the law.”1 And the Court’s most recent qualified immunity decisions have broadened the scope of the doctrine even further.2 The Court has also granted a rash of petitions for certiorari in cases in which lower courts denied qualified immunity to law enforcement officers, reversing or vacating every one.3 In these decisions, the Supreme Court has scolded lower courts for applying qualified immunity doctrine in a manner that is too favorable to plaintiffs and thus ignores the “importance of qualified immunity ‘to society as a whole.’”4 As Noah Feldman has observed, the Supreme Court’s recent qualified immunity decisions have sent a clear message to lower courts: “The Supreme Court wants fewer lawsuits against police to go
forward.”5 And the Court believes that qualified immunity doctrine is the way to keep the doors to the courthouse closed.

Among legal scholars and other commentators, there is a widespread belief that the Supreme Court is succeeding in its efforts. Scholars report that qualified immunity motions are raised frequently by defendants, are granted frequently by courts, and often result in the dismissal of cases.6 As Ninth Circuit Judge Stephen Reinhardt has written, the Supreme Court’s recent qualified immunity decisions have “created such powerful shields for law enforcement that people whose rights are violated, even in egregious ways, often lack any means of enforcing those rights.”7 Three of the foremost experts on Section 1983 litigation—Karen Blum, Erwin Chemerinsky, and Martin Schwartz—have concluded that recent developments in qualified immunity doctrine leave “not much Hope left for plaintiffs.”8

The widespread assumption that qualified immunity provides powerful protection for government officials belies how little we know about the role qualified immunity plays in the litigation of constitutional claims.9 The scant evidence available on this topic points in opposite directions. Studies of qualified immunity decisions have found that qualified immunity motions are infrequently denied, suggesting that the doctrine plays a controlling role in the resolution of many Section 1983 cases.10 But when Alexander Reinert studied the dockets in Bivens actions—constitutional cases brought against federal actors—he found that grants of qualified immunity led to just 2% of case dismissals over a three-year period.11 If qualified immunity protects all but the “plainly incompetent or those who knowingly violate the law,”12 and qualified immunity motions are infrequently denied, how can qualified immunity be the basis for dismissal of such a small percentage of cases?

More than descriptive accuracy is at stake in answering this question—it goes to a core justification for qualified immunity’s existence. Although the concept of qualified immunity was drawn from defenses existing in the common law at the time 42 U.S.C. § 1983 was enacted, the Court has made clear that the contours of qualified immunity’s protections are shaped not by the common law but instead by policy considerations.13 In particular, the Court seeks to balance “two important interests—the need to hold public officials accountable when they exercise power irresponsibly and the need to shield officials from harassment, distraction, and liability when they perform their duties reasonably.”14 Since the doctrine’s inception, the Court has repeatedly stated that financial liability is one of the burdens qualified immunity is intended to protect against.15 Yet, as I showed in a prior study, law enforcement defendants are almost always indemnified and thus rarely pay anything towards settlements and judgments entered against them.16 Near certain and universal indemnification drastically reduces the value of qualified immunity as a protection against the burden of financial liability.

In recent years, the Court has focused increasingly on a different justification for qualified immunity: the need to protect government officials from nonfinancial burdens associated with discovery and trial.17 This desire has arguably shaped qualified immunity more than any other policy justification for the doctrine.18 Yet we do not know to what extent discovery and trial burden government officials, or the extent to which qualified immunity doctrine protects against those assumed burdens. Although both questions demand critical investigation, this Article focuses on the latter. Assuming that discovery and trial do impose substantial burdens on government officials, and that shielding officials from discovery and trial is a legitimate aim of qualified immunity doctrine, to what extent does qualified immunity actually achieve its intended goal?

To answer these questions, I undertook the largest and most comprehensive study to date of the role qualified immunity plays in constitutional litigation. I reviewed the dockets of 1,183 lawsuits filed against state and local law enforcement defendants over a two-year period in five federal district courts—the Southern District of Texas, the Middle District of Florida, the Northern District of Ohio, the Eastern District of Pennsylvania, and the Northern District of California.19 I tracked several characteristics of these cases including the frequency with which qualified immunity was raised, the stage of the litigation at which qualified immunity was raised, the courts’ assessments of defendants’ qualified immunity motions, the frequency and outcome of interlocutory and final appeals of qualified immunity decisions, and the cases’ dispositions.

I found that, contrary to judicial and scholarly assumptions, qualified immunity is rarely the formal reason that civil rights damages actions against law enforcement end. Qualified immunity is raised infrequently before discovery begins: across the districts in my study, defendants raised qualified immunity in motions to dismiss in 13.9% of the cases in which they could raise the defense.20 These motions were less frequently granted than one might expect: courts granted motions to dismiss in whole or part on qualified immunity grounds 13.6% of the time.21 Qualified immunity was raised more often by defendants at summary judgment and was more often granted by courts at that stage. But even when courts granted motions to dismiss and summary judgment motions on qualified immunity grounds, those grants did not always result in the dismissal of the cases—additional claims or defendants regularly remained and continued to expose government officials to the possibility of discovery and trial. Across the five districts in my study, just 3.9% of the cases in which qualified immunity could be raised were dismissed on qualified immunity grounds.22 And when one considers all the Section 1983 cases brought against law enforcement defendants—each of which could expose law enforcement officials to whatever burdens are associated with discovery and trial—just 0.6% of cases were dismissed at the motion to dismiss stage and 2.6% were dismissed at summary judgment on qualified immunity grounds.23

Although courts rarely dismiss Section 1983 suits against law enforcement on qualified immunity grounds, there is every reason to believe that qualified immunity doctrine influences the litigation of Section 1983 claims in other ways. The threat of a qualified immunity motion may cause a person never to file suit, or to settle or withdraw her claims before discovery or trial.24 Qualified immunity motion practice and interlocutory appeals of qualified immunity denials may increase the costs and delays associated with Section 1983 litigation. The challenges of qualified immunity doctrine may cause plaintiffs’ attorneys to include claims in their cases that cannot be dismissed on qualified immunity grounds—claims against municipalities, claims seeking injunctive relief, and state law claims. Qualified immunity likely influences the litigation of cases against law enforcement in each of these ways. But, as my study makes clear, qualified immunity does not affect constitutional litigation against law enforcement in the way the Court expects and intends.

One should not conclude based on my findings that the Supreme Court simply needs to make qualified immunity stronger. As a preliminary matter, qualified immunity may not be well suited to weed out only insubstantial cases.25 Moreover, my data suggest that qualified immunity is often fundamentally ill suited to dismiss filed cases, regardless of their underlying merits.26 Although district courts recognize that they should dispose of cases as early as possible on qualified immunity grounds, plaintiffs can often plausibly plead clearly established constitutional violations and thus defeat motions to dismiss. Factual disputes regularly prevent dismissal at summary judgment. And even when courts grant qualified immunity motions, additional defendants or claims often remain that continue to expose government officials to the burdens of litigation. My data also suggest that qualified immunity is less essential than has been assumed to serve its intended protective function. The Supreme Court suggests in its opinions that qualified immunity is the only barrier standing between government officials and the burdens of discovery and trial. Instead, my study shows that litigants and courts have a wide range of tools at their disposal to resolve Section 1983 cases.

One also should not conclude based on my findings that qualified immunity is more benign than has been assumed. My findings do show that Section 1983 claims against the police are infrequently dismissed on qualified immunity grounds. But qualified immunity doctrine has been roundly criticized as incoherent, illogical, and overly protective of government officials who act unconstitutionally and in bad faith.27 The fact that few cases are dismissed on qualified immunity grounds does not fundamentally undermine these critiques.

Qualified immunity doctrine is intended by the Court to balance “the need to hold government officials accountable when they exercise power irresponsibly and the need to shield officials from harassment, distraction, and liability when they perform their duties responsibly.”28 Were qualified immunity reliably insulating government officials from the burdens of litigation in insubstantial cases, one could argue that the doctrine’s incoherence, illogic, and overprotection of government officials were unfortunate but necessary to further government interests. Yet available evidence suggests that qualified immunity is not achieving its policy objectives; the doctrine is unnecessary to protect government officials from financial liability and ill suited to shield government officials from discovery and trial in most filed cases. Qualified immunity may, in fact, increase the costs and delays associated with constitutional litigation. Qualified immunity might benefit the government in other ways, and further research is necessary to explore this possibility.29 But the evidence now available weakens the Court’s current justifications for the doctrine’s structure and highly restrictive standards. The Supreme Court has written that evidence undermining its assumptions about the realities of constitutional litigation might “justify reconsideration of the balance struck” in its qualified immunity decisions.30 Given my findings, it is high time for the Supreme Court to reconsider that balance.

The remainder of the Article proceeds as follows. Part I describes the Supreme Court’s assumptions about the burdens of discovery and trial for government officials, and the ways in which these assumptions have shaped qualified immunity doctrine. In Part II, I describe the methodology of my study. In Part III, I set forth my findings about the frequency with which law enforcement defendants raise qualified immunity, the frequency with which courts grant qualified immunity, the frequency and outcome of qualified immunity interlocutory and final appeals, and the frequency with which qualified immunity disposes of plaintiffs’ cases. In Part IV, I consider the implications of my findings for descriptive accounts of qualified immunity’s role in constitutional litigation and expectations about the policy interests served by qualified immunity doctrine. I also suggest adjustments to qualified immunity that would create more consistency between the doctrine and its actual role in constitutional litigation.

I. qualified immunity’s expected role in constitutional litigation

The Supreme Court has long viewed qualified immunity as a means of protecting government officials from burdens associated with participating in discovery and trial in insubstantial cases. Indeed, the Supreme Court has justified several major developments in qualified immunity doctrine over the past thirty-five years as means of protecting government officials from these assumed burdens. In this Part, I describe the Court’s stated assumptions about the purposes served by qualified immunity, the ways in which those assumptions have shaped qualified immunity doctrine, and the lack of evidence supporting the Court’s concerns and interventions.

A. The Court’s Concerns About the Burdens of Litigation

The Supreme Court has made clear that its qualified immunity jurisprudence reflects the Court’s view about how best to balance “the importance of a damages remedy to protect the rights of citizens” against “the need to protect officials who are required to exercise their discretion and the related public interest in encouraging the vigorous exercise of official authority.”31 Yet the Court’s descriptions of the ways in which qualified immunity protects government officials have shifted over time.

The Supreme Court’s original rationale for qualified immunity was to shield officials from financial liability. The Court first announced that law enforcement officials were entitled to a qualified immunity from suits in the 1967 case of Pierson v. Ray.32 That decision justified qualified immunity as a means of protecting government defendants from financial burdens when acting in good faith in legally murky areas.33 Qualified immunity was necessary, according to the Court, because “[a] policeman’s lot is not so unhappy that he must choose between being charged with dereliction of duty if he does not arrest when he had probable cause, and being mulcted in damages if he does.”34 The scope of the qualified immunity defense is in many ways consistent with an interest in protecting government officials from financial liability. For example, qualified immunity does not attach in claims against municipalities, claims against some private actors, and claims for injunctive or declaratory relief.35 Indeed, the Court has been clear that municipalities and private prison guards are not entitled to qualified immunity in part because neither type of defendant is threatened by personal financial liability.36

The Supreme Court’s decision in Harlow v. Fitzgerald, fifteen years after Pierson, expanded the policy goals animating qualified immunity. The Court explained in Harlow that qualified immunity was necessary not only to protect government officials from financial liability, but also to protect against “the diversion of official energy from pressing public issues,” “the deterrence of able citizens from acceptance of public office,” and “the danger that fear of being sued will ‘dampen the ardor of all but the most resolute, or the most irresponsible [public officials], in the unflinching discharge of their duties.’”37

In subsequent cases, the Court has focused increasingly on the need to protect government officials from burdens associated with discovery and trial, with the expectation that qualified immunity can protect government officials from those burdens. In Mitchell v. Forsyth, the Court reaffirmed the Harlow Court’s conclusion that qualified immunity was necessary to protect against the burdens associated with both trial and pretrial matters, like discovery, because “‘[i]nquiries of this kind can be peculiarly disruptive of effective government.’”38 In Ashcroft v. Iqbal, the Court again emphasized the value of qualified immunity in curtailing the time-intensive discovery process. As the Court explained:

The basic thrust of the qualified-immunity doctrine is to free officials from the concerns of litigation, including “avoidance of disruptive discovery.” There are serious and legitimate reasons for this. If a Government official is to devote time to his or her duties, and to the formulation of sound and responsible policies, it is counterproductive to require the substantial diversion that is attendant to participating in litigation and making informed decisions as to how it should proceed. Litigation, though necessary to ensure that officials comply with the law, exacts heavy costs in terms of efficiency and expenditure of valuable time and resources that might otherwise be directed to the proper execution of the work of the Government.39

In recent years, the interest in shielding government officials from the burdens of discovery and trial has taken center stage in the Court’s qualified immunity calculations. In 1997, the Supreme Court made clear that “the risk of ‘distraction’ alone cannot be sufficient grounds for an immunity.”40 Twelve years later, in 2009, the Court described protecting government officials from burdens associated with discovery and trial as the “‘driving force’ behind [the] creation of the qualified immunity doctrine.”41

The Court’s interest in protecting government officers from burdens associated with discovery and trial extends not only to defendants but to other government officials who may be required to testify, respond to discovery, or otherwise participate in litigation. In Filarsky v. Delia, the Court held that a private actor retained by the government to carry out its work was entitled to qualified immunity in part because the “distraction of lawsuits . . . will also often affect any public employees with whom they work by embroiling those employees in litigation.”42

B. Doctrinal Impact of the Court’s Desire To Protect Defendants from Discovery and Trial

Over the past thirty-five years, the Court’s interest in protecting government officials from discovery and trial has shaped qualified immunity in several important ways. Granted, some aspects of qualified immunity doctrine are inconsistent with the Court’s interest in protecting government officials from discovery and trial. After all, government officials must participate in discovery and trial in claims against municipalities—as witnesses, if not as defendants. In addition, government officials must participate in discovery and trial in claims for declaratory and injunctive relief. Yet, in the years since Pierson, the Court’s concerns about the burdens of discovery and trial have led the Court to remove the subjective prong of the qualified immunity defense, adjust the process by which lower courts assess qualified immunity motions, and allow interlocutory appeals of qualified immunity denials.

1. Defendants’ State of Mind

The Court’s interest in shielding government defendants from discovery and trial underlay its decision to eliminate the subjective prong of the qualified immunity defense. From 1967, when qualified immunity was first announced by the Supreme Court, until 1982, when Harlow was decided, a defendant seeking qualified immunity had to show both that his conduct was objectively reasonable and that he had a “good-faith” belief that his conduct was proper.43 In Harlow, the Supreme Court concluded that the subjective prong of the defense was “incompatible” with the goals of qualified immunity because an official’s subjective intent often could not be resolved before trial.44 Moreover, during discovery, gathering evidence of an official’s subjective motivation “may entail broad-ranging discovery and the deposing of numerous persons, including an official’s professional colleagues.”45 By eliminating the subjective prong of the qualified immunity analysis, the Court believed it could “avoid ‘subject[ing] government officials either to the costs of trial or to the burdens of broad-reaching discovery’ in cases where the legal norms the officials are alleged to have violated were not clearly established at the time.”46

2. The Order of Battle

The Court’s concerns about burdens associated with litigation also influenced its decisions regarding the manner in which courts should analyze qualified immunity. The Supreme Court believes that lower courts deciding qualified immunity motions are faced with two questions—whether a constitutional right was violated, and whether that right was clearly established. But the Court has wavered in its view regarding the order in which these questions must be answered—what is often referred to as “the order of battle.” In 2001, the Supreme Court held in Saucier v. Katz that a court engaging in a qualified immunity analysis must first decide whether the defendant violated the plaintiff’s constitutional rights and then decide whether the constitutional right was clearly established.47 The Court insisted on this sequence because it would allow “the law’s elaboration from case to case . . . . The law might be deprived of this explanation were a court simply to skip ahead to the question whether the law clearly established that the officer’s conduct was unlawful in the circumstances of the case.”48

Eight years later, in Pearson v. Callahan, the Court reversed itself and concluded that Saucier’s two-step process was not mandatory.49In reaching this conclusion, Justice Alito, writing for the Court, relied heavily on the fact that courts considered the process mandated by Saucier to be unduly burdensome.50 Justice Alito also explained that the process wasted the parties’ resources, writing that “Saucier’s two-step protocol ‘disserve[s] the purpose of qualified immunity’ when it ‘forces the parties to endure additional burdens of suit—such as the costs of litigating constitutional questions and delays attributable to resolving them—when the suit otherwise could be disposed of more readily.’”51 Concerns about the burdens of litigation therefore led the Court to allow lower courts not to decide the first question—whether the conduct was unconstitutional—if they could grant the motion on the ground that the right was not clearly established.

3. Interlocutory Appeals

The Court’s interest in protecting government officials from the burdens of discovery and trial also motivated its decision to allow interlocutory appeals of qualified immunity denials.52 Generally speaking, litigants in federal court can only appeal final judgments; interlocutory appeals are not allowed unless a right “cannot be effectively vindicated after the trial has occurred.”53 The question decided by the Court in Mitchell v. Forsyth was whether qualified immunity should be understood as an entitlement not to stand trial that cannot be remedied by an appeal at the end of the case. In concluding that a denial of qualified immunity could be appealed immediately, the Court relied on its assertion in Harlow that qualified immunity was “an entitlement not to stand trial or face the other burdens of litigation.”54 If qualified immunity protected only against the financial burdens of liability, there would be no need for interlocutory appeal; defendants denied qualified immunity could appeal after a final judgment and before the payment of any award to a plaintiff. Instead, the Court concluded, qualified immunity “is an immunity from suit rather than a mere defense to liability; and . . . it is effectively lost if a case is erroneously permitted to go to trial.”55 Only an interest in protecting officials from discovery and trial can justify this holding.

C. The Lack of Empirical Support for the Court’s Concerns and Interventions

The Supreme Court’s qualified immunity decisions over the past thirty-five years have relied on the assumptions that discovery and trial impose substantial burdens on government officials, and that qualified immunity can shield government officials from these burdens. Four years after it decided Harlow, the Court asserted that the decision had achieved the Court’s goal of facilitating dismissal at summary judgment.56 In subsequent years, the Court’s repeated invocation of the burdens of discovery and trial, and repeated reliance on qualified immunity doctrine to protect defendants from those assumed burdens, suggest the Court’s continued faith in these positions. Yet the Court has relied on no empirical evidence to support its views.

Scholars have decried the lack of empirical evidence about the realities of civil rights litigation relevant to questions about the proper scope of qualified immunity doctrine and the extent to which the doctrine achieves its intended purposes. Twenty years ago, Alan Chen complained that the Court and its critics make assertions about the role of qualified immunity in constitutional litigation without evidence to support their claims.57 Although scholars have empirically examined some questions about qualified immunity—paying particular attention to the impact of Pearson on the development of constitutional law—the same is largely true today.58 As Richard Fallon has observed, “[W]e could make far better judgments of how well qualified immunity serves the function of getting the right balance between deterrence of constitutional violations and chill of conscientious official action if we had better empirical information.”59This Article, and my research more generally, aims to fill that gap.

II. study methodology

To evaluate the role that qualified immunity plays in the litigation of Section 1983 suits, I reviewed the dockets of cases filed from January 1, 2011 to December 31, 2012 in five districts: the Southern District of Texas, Middle District of Florida, Northern District of Ohio, Eastern District of Pennsylvania, and Northern District of California. Several considerations led me to study these five districts.

I chose to look at decisions from district courts in the Third, Fifth, Sixth, Ninth, and Eleventh Circuits because I expected that judges from these circuits might differ in their approach to qualified immunity and to Section 1983 litigation more generally. This expectation was based on my review of district court qualified immunity decisions from each of the circuits, as well as a view, shared by others, that judges in these circuits range from conservative to more liberal.60 Moreover, commentators believe that courts in these circuits vary in their approach to qualified immunity, with judges in the Third and Ninth Circuits favoring plaintiffs, and judges in the Eleventh Circuit so hostile to Section 1983 cases that they are described as applying “unqualified immunity.”61

I chose these five districts within these five circuits for two reasons. First, I expected that these five districts would have a large number of cases to review: from 2011 to 2012, these districts were among the busiest in the country, as measured by case filings.62 Second, these five districts have a range of small, medium, and large law enforcement agencies and agencies of comparable sizes.63

I chose to review dockets instead of relying on the most obvious alternative—decisions available on Westlaw.64 Although Westlaw can quickly sort out decisions in which qualified immunity is addressed by district courts, Westlaw could not capture information essential to my analysis about the frequency with which qualified immunity protects government officials from discovery and trial. First, a Westlaw search could capture no information about the number of cases in which qualified immunity was never raised. In addition, a Westlaw search could not capture information about the number of cases in which qualified immunity was raised by the defendant in his motion but was not addressed by the court in its decision. Even when a defendant raises a qualified immunity defense and the district court addresses qualified immunity in its decision, the decision may not appear on Westlaw—Westlaw does not capture motions resolved without a written opinion, and includes only those opinions that are selected to appear on the service.65 In other words, opinions on Westlaw can offer insights about the ways in which district courts assess qualified immunity when they choose to address the issue in a written opinion and the opinion is accessible on Westlaw, but can say little about the frequency with which qualified immunity is raised, the manner in which all motions raising qualified immunity are decided, and the impact of qualified immunity on case dispositions.

I reviewed the dockets of cases filed in 2011 and 2012 in the five districts in my study.66 I searched case filings in the five districts in my study through Bloomberg Law, an online service that has dockets otherwise available through PACER and additionally provides access to documents submitted to the court—complaints, motions, orders, and other papers.67 Within Bloomberg Law, I limited my search to those cases that plaintiffs had designated under the broad term “Other Civil Rights,” nature-of-suit code 440.68 This search generated 462 dockets in the Southern District of Texas, 465 dockets in the Northern District of Ohio, 674 dockets in the Middle District of Florida, 712 dockets in the Northern District of California, and 1,435 dockets in the Eastern District of Pennsylvania. I reviewed the complaints associated with these 3,748 dockets and included in my dataset those cases, brought by civilians, alleging constitutional violations by state and local law enforcement agencies and their employees.69

I limited my study to cases brought by civilians against law enforcement defendants for several reasons. First, many of the Supreme Court’s qualified immunity decisions have involved cases brought against law enforcement. Of the twenty-nine qualified immunity cases that the Supreme Court has decided since 1982, almost half have involved constitutional claims against state and local law enforcement.70 Because the Court has developed qualified immunity doctrine (and articulated its underlying purposes) primarily in cases involving law enforcement, it makes sense to examine whether the doctrine is meeting its express goals in these types of cases.

Limiting my study to Section 1983 cases against state and local law enforcement also creates some substantive consistency across the cases in my dataset. Most Section 1983 cases against state and local law enforcement allege Fourth Amendment violations—excessive force, false arrest, and wrongful searches—and, less frequently, First and Fourteenth Amendment violations. Restricting my study to suits by civilians against state and local law enforcement facilitates direct comparison of outcomes in similar cases across the five districts in my study. Finally, much of my own prior research has focused on lawsuits against state and local law enforcement, and maintaining this focus here allows for future synthesis of my findings.71

The resulting dataset includes a total of 1,183 cases from these five districts: 131 cases from the Southern District of Texas, 225 cases from the Middle District of Florida, 172 cases from the Northern District of Ohio, 248 cases from the Northern District of California, and 407 cases from the Eastern District of Pennsylvania. For each of these dockets, I tracked multiple pieces of information relevant to this study, including whether the plaintiff(s) sued individual officers and/or the municipality, the relief sought by the plaintiff(s), whether the law enforcement defendant(s) filed one or more motions to dismiss on the pleadings or for summary judgment, whether and when the defendant(s) raised qualified immunity, how the court decided the motions raised by the defendant(s), whether there was an interlocutory or final appeal of a qualified immunity decision, and how the case was ultimately resolved.72 Although some of this information was available from the docket sheet, I obtained much of the information by reading motions and opinions linked to the dockets on Bloomberg Law.

Although some of my coding decisions were straightforward, others involved less obvious choices. Because my coding decisions may make most sense when reviewed in context, I have described those decisions in detail in the footnotes accompanying the data.73 Throughout, my coding decisions were guided by my focus on the role that qualified immunity played in the resolution of cases and the frequency with which the doctrine meets its goal of shielding government officials from discovery and trial.

My dataset is comprehensive in the five chosen districts. It includes most—if not all—Section 1983 cases filed by civilians against state and local law enforcement in these federal districts over a two-year period, and it offers insights about how frequently qualified immunity is raised in these cases, how courts decide these motions, and how the cases are resolved. There are, however, several limitations of the data. First, although I selected these five districts in part to capture regional variation, they may not represent the full range of court and litigant behavior nationwide. The marked differences in my data across districts do, however, suggest a considerable degree of regional variation. Second, the data offer no information about the role of qualified immunity in state court litigation. This is in part because Bloomberg Law does not offer much information about the litigation of constitutional cases in state courts.74

Third, although this study sheds light on the litigation of constitutional claims against state and local law enforcement officers, it does not necessarily describe the role qualified immunity plays in the litigation of constitutional claims against other types of government employees. It may be that the types of constitutional claims often raised in cases against law enforcement—Fourth Amendment claims alleging excessive force, unlawful arrests, and improper searches—are particularly difficult to resolve on qualified immunity grounds in advance of trial. Fourth Amendment claims may be comparatively easy to plead in a plausible manner (and so could survive a motion to dismiss), and such claims may be particularly prone to factual disputes (making resolution at summary judgment difficult). If so, perhaps qualified immunity motions in cases raising other types of claims would be more successful. On the other hand, John Jeffries has argued that it may be particularly difficult to clearly establish that a use of force violates the Fourth Amendment because Fourth Amendment analysis requires a fact-specific inquiry about the nature of the force used and the threat posed by the person against whom force was used, viewed from the perspective of an officer on the scene.75Further research should explore whether qualified immunity plays a different role in cases brought against other government actors, or cases alleging different types of constitutional violations.

Fourth, qualified immunity may be influencing the litigation of constitutional claims in ways that cannot be measured through the examination of case dockets.76 For example, my study does not measure how frequently qualified immunity causes people not to file lawsuits. It also does not capture information about the frequency with which plaintiffs’ decisions to settle or withdraw their claims are influenced by the threat of a qualified immunity motion or decision. Exploration of these issues is critical to a complete understanding of the role qualified immunity plays in constitutional litigation. I discuss these issues in more depth in Part IV, and future research should explore these questions.77 Yet this Article illuminates several important aspects of qualified immunity’s role in Section 1983 cases. Moreover, by measuring the frequency with which qualified immunity motions are raised, granted, and dispositive, this Article reveals the extent to which the doctrine functions as the Supreme Court expects and critics fear.

III. findings

The Supreme Court has explained that a goal of qualified immunity is to “avoid ‘subject[ing] government officials either to the costs of trial or to the burdens of broad-reaching discovery’ in cases where the legal norms the officials are alleged to have violated were not clearly established at the time.”78 Logically, qualified immunity will only achieve this goal in a case if four conditions are met.

First, the case must be brought against an individual officer and must seek monetary damages. Qualified immunity is not available for claims against municipalities or claims for noneconomic relief. Second, the defendant must raise the qualified immunity defense early enough in the litigation that it can protect him from discovery or trial. If the defendant seeks to protect himself from discovery, he must raise qualified immunity in a motion to dismiss or a motion for judgment on the pleadings; if a defendant seeks to protect himself from trial, he can raise qualified immunity at the pleadings or at summary judgment.79 Third, for a qualified immunity motion to protect government officials against burdens associated with discovery or trial, the court must grant the motion on qualified immunity grounds.80 Finally, the grant of qualified immunity must completely resolve the case. If qualified immunity is granted for an officer on one claim but not another, that officer will continue to have to participate in the litigation of the case. Even when a grant of qualified immunity results in the dismissal of all claims against a defendant, that defendant may still have to participate in the litigation of claims against other defendants. To be sure, the government official who has been dismissed from the case may no longer feel the same psychological burdens associated with the litigation and may have lesser discovery burdens than he would have had as a defendant. But the grant of qualified immunity will not necessarily shield him from the burdens of participating in discovery and trial.

This Part describes my findings regarding the frequency with which each of these conditions is met. I empirically examine six topics: (1) the number of cases in which qualified immunity can be raised by defendants; (2) the number of cases in which defendants choose to raise qualified immunity; (3) the stage(s) of litigation at which defendants raise qualified immunity; (4) the ways in which district courts decide qualified immunity motions; (5) the frequency and outcome of qualified immunity appeals; and (6) the frequency with which qualified immunity is the reason that a case ends before discovery or trial.

My findings regarding these six topics show that, at least in filed cases, qualified immunity rarely functions as expected. Qualified immunity could not be raised in more than 17% of the cases in my dataset, either because the cases did not name individual defendants or seek monetary damages, or because the cases were dismissed sua sponte by the court before the defendants had an opportunity to answer or otherwise respond. Defendants raised qualified immunity in 37.6% of the cases in my dataset in which the defense could be raised. Defendants were particularly disinclined to raise qualified immunity in motions to dismiss: they did so in only 13.9% of the cases in which they could raise the defense at that stage. Courts granted (in whole or part) less than 18% of the motions that raised a qualified immunity defense. Qualified immunity was the reason for dismissal in just 3.9% of the cases in my dataset in which the defense could be raised, and just 3.2% of all cases in my dataset. The remainder of this Part describes each of these findings in more detail.

A. Cases in Which Qualified Immunity Cannot Play a Role

There are certain types of cases in which qualified immunity cannot play a role. The Supreme Court has held that qualified immunity does not apply to claims against municipalities and claims for injunctive or declaratory relief.81 Accordingly, qualified immunity cannot protect government officials from discovery or trial in cases asserting only these types of claims. In my docket dataset of 1,183 cases, ninety-nine cases (8.4%) were brought solely against municipalities and/or sought only injunctive or declaratory relief.82

table 1.

frequency with which qualified immunity can be raised, in five districts

Even when cases are brought against individual officers and seek monetary relief, there are some cases in which defendants have no need to raise qualified immunity as a defense—cases dismissed sua sponte by the court before the defendants respond to the complaint. In these cases, qualified immunity is unnecessary to protect defendants from discovery and trial. In the five districts in my docket dataset, 105 (8.9%) complaints naming individual law enforcement officers and seeking damages were dismissed sua sponte by district courts before defendants answered or responded. Most often, district courts dismissed these cases pursuant to their statutory power to review pro se plaintiffs’ complaints and dismiss actions they conclude are frivolous or meritless.83 Other cases were dismissed by the court at this preliminary stage because the plaintiffs never served the defendants or failed to prosecute the case, or because the court remanded the case to state court for lack of subject matter jurisdiction before the defendants were served or responded.

Qualified immunity can only protect government officials from discovery and trial in cases in which government defendants can raise the defense. Defendants could not raise qualified immunity in 8.4% of cases in my docket dataset because those cases did not name individual defendants and/or seek damages. Qualified immunity was unnecessary to shield government officials from discovery or trial in another 8.9% of cases in my dataset because these cases were dismissed by the district courts before defendants could raise the defense. Accordingly, defendants could raise a qualified immunity defense in a total of 979 (82.8%) of the 1,183 complaints filed in the five districts during my two-year study period.

B. Defendants’ Choices: The Frequency and Timing of Qualified Immunity Motions

Qualified immunity can only protect a defendant from the burdens of discovery and trial if she raises the defense in a dispositive motion. Accordingly, this Section examines the frequency with which defendants raise qualified immunity and the stage of litigation at which they raise the defense.84

table 2.

frequency with which qualified immunity is raised

Defendants raised qualified immunity one or more times in 368 (37.6%) of the 979 cases in which defendants could raise the defense. The frequency with which defendants raised qualified immunity varied substantially by district. Defendants in the Southern District of Texas and the Middle District of Florida were most likely to raise the qualified immunity defense; in these districts, defendants brought one or more motions raising qualified immunity in approximately 54% of the cases in which the defense could be raised. Defendants in the Eastern District of Pennsylvania were least likely to raise the qualified immunity defense; defendants brought one or more motions raising qualified immunity in approximately 24% of cases in which the defense could be raised. Defendants in the Northern District of California brought qualified immunity motions in 33.8% of possible cases, and in the Northern District of Ohio defendants raised qualified immunity in 47.5% of possible cases.

I also explored the stage(s) of litigation at which qualified immunity was raised. Of the 368 cases in which qualified immunity was raised at least once, defendants in ninety-five (25.8%) cases raised qualified immunity only in a motion to dismiss or motion for judgment on the pleadings, defendants in 229 (62.2%) cases raised qualified immunity only in a motion for summary judgment, and defendants in forty-one (11.1%) cases raised qualified immunity at both the motion to dismiss and summary judgment stages. Based on my review of motions and opinions available on Bloomberg Law, I can confirm only three cases in which defendants included qualified immunity in a motion at or after trial for judgment as a matter of law. My data almost certainly underrepresent the role qualified immunity plays at or after trial, however, as Bloomberg Law does not include oral motions or court decisions issued without a written opinion.85

table 3.

timing of qualified immunity motions

Across the five districts in my study, defendants raised qualified immunity at summary judgment approximately twice as often as they did at the motion to dismiss stage. In cases where defendants brought one or more qualified immunity motions, defendants in 73.9% of the cases raised qualified immunity at summary judgment, whereas defendants in 37.0% of the cases raised qualified immunity in a motion to dismiss. There is, however, regional variation in this regard. Defendants in the Middle District of Florida were equally likely to raise qualified immunity at the pleadings stage and at summary judgment, whereas in the Northern District of Ohio and the Northern District of California defendants were more than three times more likely to raise qualified immunity at summary judgment than they were to raise the defense in a motion to dismiss.

Defendants in the Middle District of Florida were also more likely to raise qualified immunity at more than one stage of litigation—they raised qualified immunity at multiple stages of litigation in nineteen (22.6%) of the cases in which they raised the defense. Defendants in the other districts less frequently raised qualified immunity at multiple stages of litigation; they did so in six (10.3%) of the cases in which the defense was raised in the Southern District of Texas, in seven (9.5%) of the cases in which the defense was raised in the Northern District of California, in eight (9.3%) of the cases in which the defense was raised in the Eastern District of Pennsylvania, and in three (4.5%) of the cases in which the defense was raised in the Northern District of Ohio.

I additionally sought to calculate how frequently defendants chose to raise qualified immunity motions in all the cases in which such motions could be brought. This calculation is relatively straightforward regarding motions to dismiss. Defendants could have brought motions to dismiss on qualified immunity grounds in any of the 979 cases in which the defense could be raised and did so in 136 (13.9%) of these cases.

Calculating the number of possible summary judgment motions on qualified immunity grounds is more complicated. Although defendants could bring a summary judgment motion in any case in which they could offer some evidence in support, defendants generally do not move for summary judgment without first engaging in at least some formal discovery.86 It is difficult to discern from case dockets to what extent parties have engaged in discovery, but the dockets do reflect whether a case management order has been issued, which generally sets the discovery schedule and is the first step of the discovery process. If entry of a case management order can serve as an indication that a case has entered discovery, and if one accepts that defendants in cases that have conducted some discovery could move for summary judgment, then there are 577 cases in my dataset in which defendants could have moved for summary judgment. Defendants brought summary judgment motions on qualified immunity grounds in 272 (47.1%) of these cases.

I also calculated the total number of qualified immunity motions brought by defendants. Defendants sometimes raised qualified immunity in multiple motions to dismiss or summary judgment motions that were resolved by the court in separate opinions: if, for example, defendants moved to dismiss on qualified immunity grounds, the court granted the motion with leave to amend, and the plaintiff filed an amended complaint, the defendants might again move to dismiss on qualified immunity grounds.87 Defendants filed a total of 440 qualified immunity motions in the 368 cases in which they raised the defense. Table 4 reflects the stage of litigation at which these 440 motions were brought and, again, reflects that defendants file significantly more qualified immunity motions at summary judgment than at the motion to dismiss stage. Of the 440 qualified immunity motions filed, 154 (35.0%) were filed in a motion to dismiss or motion for judgment on the pleadings, and 283 (64.3%) were filed at summary judgment.

table 4.

total qualified immunity motions filed, by stage of litigation

table 5.

number of qualified immunity motions per case

Table 5 reflects the distribution of these 440 motions among the 368 cases in which the defense was raised. Table 5 shows that when defendants raise qualified immunity they usually do so in only one motion, but that defendants in the Southern District of Texas and Middle District of Florida are more likely than defendants in the other three districts to file multiple motions raising qualified immunity.

Finally, I explored how frequently defendants raise other types of defenses in motions to dismiss or for judgment on the pleadings and in summary judgment motions. Qualified immunity is usually one of several arguments defendants make in their motions to dismiss and for summary judgment. Indeed, defendants sometimes move to dismiss or for summary judgment without raising qualified immunity at all.

Of the 979 cases in my docket dataset in which defendants could raise qualified immunity, defendants filed a total of 462 motions to dismiss, and 154 (33.3%) included a qualified immunity argument.88 Defendants in the Middle District of Florida were the most likely to raise qualified immunity in motions to dismiss or for judgment on the pleadings—defendants included a qualified immunity argument in 45.4% of their motions, compared with 39.0% of the motions filed by defendants in the Southern District of Texas, 32.1% of the motions filed by defendants in the Northern District of Ohio, 26.2% of the motions filed by defendants in the Eastern District of Pennsylvania, and 23.5% of the motions filed by defendants in the Northern District of California. Motions to dismiss or for judgment on the pleadings that did not raise qualified immunity argued instead that the complaint did not satisfy plausibility pleading requirements, concerned a claim that was barred by a criminal conviction, or otherwise did not state a legally cognizable claim.89

figure 1.

motions to dismiss/for judgment on the pleadings

figure 2.

summary judgment motions

Defendants in all five districts were far more likely to include a qualified immunity argument in their summary judgment motions. Defendants filed 374 motions for summary judgment, and 283 (75.7%) of those motions included an argument based on qualified immunity. There was some variation among the districts in this area as well, although the regional variation was less pronounced here than in other aspects of qualified immunity litigation practice.90

C. District Courts’ Decisions: The Success Rate of Qualified Immunity Motions

This Section examines how frequently district courts grant motions to dismiss and for summary judgment on qualified immunity grounds. As I have shown, qualified immunity is almost always raised in conjunction with other arguments in motions to dismiss or for summary judgment. My focus here is on the way the district court evaluates the qualified immunity argument.

table 6.

success of motions raising qualified immunity

In the five districts in my docket dataset, defendants raised qualified immunity in a total of 440 motions. Table 6 reflects the way in which district courts resolved those motions.91Across the five districts in my study, qualified immunity motions were denied 31.6% of the time.92 Qualified immunity motions in these five districts were granted in part—on some claims or defendants but not others—5.9% of the time and granted in full on qualified immunity grounds 12.0% of the time. In another 11.4% of the decisions, courts concluded that the plaintiff had not met her burden of establishing a constitutional violation and either declined to reach the second step of the qualified immunity analysis (whether a reasonable officer would have believed that the law was clearly established) or granted qualified immunity in the alternative.93 Courts in 14.1% of the decisions granted defendants’ motions on other grounds without addressing qualified immunity, and in another 2.0% of the decisions the courts offered little or no rationale. Courts in 5.9% of the decisions granted the motion in part without mentioning qualified immunity, or on qualified immunity in the alternative. And district courts in my study did not decide 17.0% of the motions raising qualified immunity, usually because the cases settled or were voluntarily dismissed while the motions were pending.

There was substantial variation in courts’ decisions across the districts in my study. The Southern District of Texas had the lowest rate of qualified immunity denials (21.7%). In the remaining four districts, judges denied 30-38% of defendants’ qualified immunity motions. The Southern District of Texas also had the highest rate of qualified immunity grants: courts in the Southern District of Texas granted 33.3% of defendants’ qualified immunity motions in part or full on qualified immunity grounds. In contrast, courts in the Eastern District of Pennsylvania granted only 6.1% of the qualified immunity motions in whole or part on qualified immunity grounds.94

table 7.

rulings on motions to dismiss/on the pleadings that raised qualified immunity

table 8.

rulings on summary judgment motions that raised qualified immunity

I additionally evaluated differences in courts’ decisions at the motion to dismiss and summary judgment stages.95 Of the 154 motions to dismiss and motions for judgment on the pleadings raising qualified immunity, courts granted seventy-nine (51.3%) of the motions in whole or part. Twenty-one (26.6%) of those seventy-nine full or partial grants were decided on qualified immunity grounds. Of the 283 summary judgment motions raising qualified immunity, courts granted 146 (51.6%) in whole or part. Fifty-eight (39.7%) of those 146 full or partial grants were decided on qualified immunity grounds. In other words, although courts were equally likely to grant summary judgment motions and motions to dismiss, courts were more likely to grant summary judgment motions on qualified immunity grounds than they were to grant motions to dismiss on qualified immunity grounds. But courts more often than not granted both types of motions on grounds other than qualified immunity.

D. Circuit Courts’ Decisions: The Frequency and Success of Qualified Immunity Appeals

A complete examination of the role qualified immunity plays in constitutional litigation must examine the frequency and outcome of qualified immunity appeals. Defendants can appeal denials of qualified immunity immediately, and any qualified immunity decision can be appealed after a final judgment in the case.96

table 9.

interlocutory appeals of qualified immunity denials

Defendants immediately appealed 41 of the 189 qualified immunity decisions in my docket dataset that were denied or granted in part and thus could have been appealed at this stage of the litigation—an interlocutory appeal rate of 21.7%. Across the five districts in my dataset, more than one-third of the lower courts’ decisions were affirmed on interlocutory appeal, 12.2% were reversed in whole, 7.3% were reversed in part, and 39.0% were withdrawn by the parties without a decision by the court of appeals.

table 10.

final appeals of qualified immunity grants

I also tracked the frequency with which plaintiffs appealed qualified immunity grants after a final judgment in the case.97 Plaintiffs appealed twenty-six (32.9%) of the seventy-nine decisions granting defendants’ motions on qualified immunity grounds in whole or part.98 Lower court decisions granting qualified immunity were affirmed 65.4% of the time and reversed 7.7% of the time. Almost 27% of the appeals were withdrawn without a decision.

E. The Impact of Qualified Immunity on Case Dispositions

A final question concerns the frequency with which a grant of qualified immunity results in the dismissal of Section 1983 cases. There are multiple ways to frame this inquiry. First, there is the question of which cases should be counted in the numerator—cases dismissed on qualified immunity grounds. I have included qualified immunity grants in this category unless the court ended its qualified immunity analysis after concluding that the plaintiff could not establish a constitutional violation, or granted the motion on qualified immunity in the alternative. Although the question of whether a constitutional violation occurred is the first step of the qualified immunity analysis, the court would also need to resolve this question in the absence of qualified immunity. And although a court’s decision to grant qualified immunity in the alternative may influence its dispositive holding in some manner, the qualified immunity decision was not necessary to resolve the case.99

In addition, I have counted a case as dismissed on qualified immunity grounds only if the entire case has been dismissed as a result of the motion. One might assume that a grant of qualified immunity will always end a case. Yet there are multiple scenarios in which a case can continue after a grant of qualified immunity. At the pleadings stage, a court may grant a motion to dismiss on qualified immunity but also grant the plaintiff an opportunity to amend her complaint.100 Not all defendants in a case will necessarily move to dismiss on qualified immunity grounds,101 or a defendant may seek qualified immunity regarding some but not all claims against him.102 State law claims may also remain for which qualified immunity is not available, and these claims may proceed in federal court or be remanded to and pursued in state court.103 In addition, municipalities cannot assert qualified immunity; accordingly, if there is a municipality named in the case at the time qualified immunity is granted, the case will continue.104 Under each of these circumstances, government officials still face the possibility that they will be required to participate in discovery and trial as defendants, representatives of the defendants’ agency, and/or witnesses to the events in question.105

table 11.

impact of qualified immunity, by stage of litigation

As Table 11 shows, there are fifty-three motions in my dataset that district courts granted in full on qualified immunity grounds—fourteen at the pleadings stage and thirty-nine at summary judgment. Of those fifty-three motions, thirty-four (64.2%) were dispositive, meaning that the cases were dismissed as a result of the qualified immunity decision. Half of qualified immunity grants at the pleadings stage led to case dismissals, and 69.2% of qualified immunity grants at summary judgment led to case dismissals. Defendants brought forty-one interlocutory appeals of qualified immunity denials, and courts of appeals reversed five (12.2%) of those decisions. All five reversals were of summary judgment decisions, and four of the five resulted in case dismissals. In total, qualified immunity led to dismissal of thirty-eight cases in my dataset.

The next question, when thinking about the impact of qualified immunity on case disposition, is how to frame the denominator—the universe of cases against which to measure the cases dismissed on qualified immunity grounds. It is my view that the broadest definition of the denominator—all 1,183 Section 1983 cases filed against law enforcement—offers the most accurate picture of the role qualified immunity plays in Section 1983 litigation. Yet, as I will show, there are at least three ways to frame the denominator, and each answers a different question about the extent to which qualified immunity achieves its intended goals.

One way to think about the impact of qualified immunity is to consider the frequency with which a defendant’s motion to dismiss or for judgment on the pleadings, for summary judgment, or for judgment as a matter of law on qualified immunity grounds actually leads to the dismissal of a case—whether because the motion is granted or because the motion is denied by the district court but reversed on appeal. Presumably, a defendant will only bring a qualified immunity motion when two conditions are met: he has a non-frivolous basis for the motion, and he believes that the costs of bringing the motion are justified by the likelihood of success or some other benefit associated with the motion. Accordingly, this framework assesses the frequency with which qualified immunity results in the dismissal of cases in which both these things are true.

Defendants brought 440 qualified immunity motions in a total of 368 cases in the five districts in my study: defendants raised qualified immunity in 154 motions to dismiss and raised qualified immunity in 283 summary judgment motions. Courts granted 9.1% of the motions to dismiss on qualified immunity grounds, and 4.5% of the motions resulted in case dismissals. Courts granted 13.8% of the summary judgment motions on qualified immunity grounds, and 9.5% of the motions resulted in case dismissals. Defendants brought forty-one interlocutory appeals of qualified immunity denials, courts of appeals reversed five (12.2%) of those decisions, and four of the five were dismissed as a result. In total, thirty-eight (8.6%) of the 440 qualified immunity motions raised by defendants in my dataset resulted in case dismissals, and 10.3% of the 368 cases in which qualified immunity was raised were dismissed on qualified immunity grounds.

Another way to assess the impact of qualified immunity on case outcomes is to examine what percentage of the 979 cases in my dataset in which qualified immunity could be raised were in fact dismissed on qualified immunity grounds. One objection to this framing might be that it includes cases that defendants declined to challenge on qualified immunity grounds. But qualified immunity motions would not necessarily have failed in these cases; rather, defendants in these cases concluded that the costs of raising the defense were not justified by the likelihood of success or other benefits of bringing the motions. Moreover, this broader framework illustrates the frequency with which qualified immunity doctrine serves its intended and expected role of shielding government officials from burdens associated with discovery and trial. Evaluated in this manner, qualified immunity is less frequently successful. Qualified immunity was the basis for dismissal in 3.9% of the 979 cases in which the defense could be raised: just seven (0.7%) of cases were dismissed on qualified immunity grounds at the motion to dismiss stage, and thirty-one (3.2%) of cases were dismissed on qualified immunity grounds at summary judgment—either by the district court or on appeal.

Indeed, to evaluate fully the role that qualified immunity plays in the resolution of constitutional claims against law enforcement, the most appropriate denominator is the complete universe of 1,183 cases in my dataset. This approach includes cases that could not be resolved on qualified immunity grounds—because the cases were either brought only against municipalities or sought only equitable relief. But to the extent that the Court views qualified immunity doctrine as a shield for all government officials—not only defendants—from burdens associated with discovery and trial, a thorough assessment of qualified immunity’s role should take account of all the cases in which government officials must participate. Qualified immunity was the basis for dismissal in 3.2% of the 1,183 cases in my dataset: 0.6% of cases were dismissed on qualified immunity grounds at the motion to dismiss stage, and 2.6% of cases were dismissed on qualified immunity grounds at summary judgment—either by the district court or on appeal.106

My data show that qualified immunity is rarely the formal reason that Section 1983 cases are dismissed. How, then, are Section 1983 suits against law enforcement resolved? Table 12 reports case outcomes for the 1,183 cases in the five districts in my study.

table 12.

case dispositions

If one adopts the standard definition of plaintiff “success” to include jury verdicts, settlements, and voluntary or stipulated dismissals, the plaintiffs in my dataset succeeded in 682 (57.7%) cases.107 This success rate is similar to the results of Theodore Eisenberg and Stewart Schwab’s studies of non-prisoner Section 1983 cases.108 The remaining 42.3% of cases resolved in various ways: 256 (21.6%) were dismissed on motions to dismiss or for judgment on the pleadings, at summary judgment, or at or after trial on grounds other than qualified immunity; 173 (14.6%) were dismissed sua sponte before defendants answered, dismissed as a sanction, or dismissed for failure to prosecute; and thirty-seven (3.1%) were dismissed for other reasons or remain open. Thirty-eight (3.2%) were dismissed on qualified immunity grounds.

My data do not capture how frequently qualified immunity influences plaintiffs’ decisions to settle, or how frequently cases are decided on qualified immunity grounds even though other defenses are available. Instead, my data reflect the frequency with which a grant of qualified immunity formally ends a case. There is, once again, marked regional variation in the frequency with which qualified immunity leads to the dismissal of Section 1983 actions.109 But despite this regional variation, grants of qualified immunity motions infrequently end Section 1983 suits before discovery, and are infrequently the reason suits are dismissed before trial.

IV. implications

My findings undermine prevailing assumptions about the role qualified immunity plays in the litigation of Section 1983 claims. Accordingly, in this Part I consider the implications of my findings for ongoing discussions about the proper scope of qualified immunity in relation to its underlying purposes. First, I revisit empirical claims implicit in the Supreme Court’s qualified immunity decisions in light of my findings. Next, I consider why qualified immunity disposes of so few cases before trial. Armed with this more realistic appraisal of qualified immunity’s role, I argue that the Court has struck the wrong balance between fairness and accountability for law enforcement officers. Finally, I suggest that qualified immunity doctrine should be adjusted to comport with available evidence about the role the doctrine plays in constitutional litigation.

A. Toward a More Accurate Description of Qualified Immunity’s Role in Constitutional Litigation

The Court’s qualified immunity decisions paint a clear picture of the ways in which the Court believes the doctrine should operate: it should be raised and decided at the earliest possible stage of the litigation (at the motion to dismiss stage if possible), it should be strong (protecting all but the plainly incompetent or those who knowingly violate the law), and it should, therefore, protect defendants from the time and distractions associated with discovery and trial in insubstantial cases. Commentators similarly believe that qualified immunity is often raised by defendants, usually granted by courts, and causes many cases to be dismissed.110

My study shows that, at least in filed cases, qualified immunity rarely functions as expected. Defendants could not or did not need to raise qualified immunity in 17.3% of the 1,183 cases in my docket dataset, either because the cases did not name individual defendants or seek monetary damages, or because the cases were dismissed sua sponte by the court before the defendants had an opportunity to answer. Defendants raised qualified immunity in motions to dismiss and motions for judgment on the pleadings in only 13.9% of the cases in which the defense could be raised.111 Courts granted those motions on qualified immunity grounds 9.1% of the time, but those grants were not always dispositive because additional claims or defendants remained, or because plaintiffs were given the opportunity to amend. As a result, just seven of the 1,183 cases in my docket dataset were dismissed at the motion to dismiss stage on qualified immunity grounds.

Qualified immunity more often prevented cases from proceeding past summary judgment. Defendants were more likely to include qualified immunity in motions for summary judgment than in motions to dismiss, and courts were more likely to grant summary judgment motions than motions to dismiss on qualified immunity grounds.112 Moreover, courts of appeals reversed five denials of summary judgment motions on interlocutory appeal and granted qualified immunity in these cases. Yet qualified immunity motions at the summary judgment stage rarely shield government officials from discovery because most summary judgment motions require at least some depositions or document exchange.113 And grants of qualified immunity at summary judgment relatively rarely achieved their goal of protecting government officials from trial—such decisions by the district courts or courts of appeals disposed of plaintiffs’ cases just thirty-one times across the five districts in my study, amounting to just 2.6% of the 1,183 cases in my dataset.

Qualified immunity is likely raised more often at or after trial than my data suggest. But even if many more qualified immunity motions are made during or after trial, and even if qualified immunity regularly convinces judges and juries to enter defense verdicts, qualified immunity would still fail to serve its expected role. Qualified immunity doctrine is intended to shield government officials from burdens associated with litigation and trial. A grant of qualified immunity entered during or after trial has come too late to shield government officials from these assumed burdens.

My data demonstrate considerable regional differences in the litigation and adjudication of qualified immunity across the country. Scholars have observed that the federal circuits interpret qualified immunity standards differently.114 My findings suggest that regional differences in qualified immunity doctrine affect the decisions of courts and litigants. Defendants in the Southern District of Texas and the Middle District of Florida were more likely to raise qualified immunity than defendants in the Eastern District of Pennsylvania and the Northern District of California; courts in the Southern District of Texas and the Middle District of Florida were more likely to grant defendants’ qualified immunity motions than were judges in the Eastern District of Pennsylvania and the Northern District of California; and grants of qualified immunity ended more cases in the Southern District of Texas and the Middle District of Florida than in the Eastern District of Pennsylvania and the Northern District of California. But even in the Southern District of Texas—the district in my dataset most likely to dismiss cases on qualified immunity grounds—just 2.3% of all suits were dismissed on qualified immunity grounds at the motion to dismiss stage, and 6.9% of all suits were dismissed at summary judgment on qualified immunity grounds.115 Unless the vast majority of law enforcement officer defendants in the Southern District of Texas are “plainly incompetent” or have “knowingly violate[d] the law,”116 qualified immunity is not playing its expected role even in the district in my dataset most sympathetic to the defense.

Although qualified immunity is rarely the reason that Section 1983 cases end, there are other ways in which qualified immunity doctrine might influence the litigation of constitutional claims against law enforcement. For example, qualified immunity may discourage people from ever filing suit. Available evidence suggests that just 1% of people who believe they have been harmed by the police file lawsuits against law enforcement.117 We do not know how frequently qualified immunity doctrine plays a role in the decision not to sue. But available evidence suggests that qualified immunity often factors into plaintiffs’ attorneys’ decisions about whether to accept potential clients. When Alexander Reinert interviewed plaintiffs’ attorneys about qualified immunity in Bivens cases, attorneys reported that “the qualified immunity defense play[s] a substantial role at the screening stage.”118 Attorneys described being discouraged from accepting civil rights cases both because qualified immunity motions can be difficult to defeat and because the costs and delays associated with litigating qualified immunity can make the cases too burdensome to pursue.119 But attorneys also described qualified immunity as one of many factors they considered when deciding whether to accept a case, and we do not know how attorneys weigh these different considerations.120

Even when cases are filed, qualified immunity may influence litigation decisions in ways that are not easily observable through docket review. For example, it may be that a pending qualified immunity motion will cause a plaintiff to settle her claims. Consistent with this theory, seventy-five (17.0%) of the qualified immunity motions in my dataset were never decided, presumably because the parties settled while the motions were pending.121 Of the sixty-seven qualified immunity interlocutory and final appeals in my dataset, twenty-three (34.3%) were withdrawn or dismissed without decision, which suggests that many of those cases settled while on appeal.122 When the Supreme Court has described the ways in which it expects qualified immunity to shield government officials from discovery and trial, it has never suggested that the doctrine might serve this function by discouraging people from filing lawsuits or pursuing their claims. But these are certainly ways in which qualified immunity could achieve this goal.

A complete understanding of the frequency with which qualified immunity protects government officials from discovery and trial would measure these other potential litigation effects. For the time being, available evidence suggests that qualified immunity may make it more difficult for plaintiffs to secure representation and may encourage plaintiffs to settle, but it is infrequently the formal reason that cases end.

B. Why Qualified Immunity Disposes of So Few Cases

The Supreme Court designed qualified immunity to protect “all but the plainly incompetent or those who knowingly violate the law.”123 Why, then, does it lead to the dismissal of so few cases? One possibility is that qualified immunity doctrine discourages people from filing cases that are unlikely to meet qualified immunity’s exacting standard.124 But even if qualified immunity has this selection effect, plaintiffs would continue to file cases in which qualified immunity motions might be successful. Consistent with this theory, defendants raised qualified immunity in more than one-third of the Section 1983 cases in which the defense could be asserted, and courts granted 51.4% of motions raising qualified immunity in full or part.125 Yet most of these motions to dismiss and summary judgment motions raised multiple arguments, and courts only granted 17.9% of these motions in part or whole on qualified immunity grounds. Ultimately, qualified immunity resulted in the dismissal of only 3.9% of the cases in which the defense could be raised. Although the threat of qualified immunity may cause some people not to sue, this selection effect does not explain why qualified immunity plays such a limited role in the resolution of motions raising qualified immunity and in the disposition of cases that are filed.

The Supreme Court’s decisions suggest another theory that could partially explain why qualified immunity disposes of few cases: because courts improperly deny defendants’ qualified immunity motions. For this reason, and because of the “importance of qualified immunity ‘to society as a whole,’” the Supreme Court has taken the unusual step of “often correct[ing] lower courts when they wrongly subject individual officers to liability.”126 Yet qualified immunity grant rates are lower than expected even in the circuits generally believed to be the most amenable to qualified immunity: 33.3% of motions raising qualified immunity were granted in whole or part on qualified immunity grounds in the Southern District of Texas, and 22.5% of motions raising qualified immunity were granted in whole or part on qualified immunity grounds in the Middle District of Florida.127 Moreover, only 9.2% of cases from the Southern District of Texas and 6.7% of cases from the Middle District of Florida were actually dismissed on qualified immunity grounds. Unless one believes that the Southern District of Texas and the Middle District of Florida, as well as the Fifth and Eleventh Circuits, are regularly flouting the letter and spirit of the Supreme Court’s qualified immunity doctrine, error in the lower courts is an unconvincing—or at least incomplete—explanation for these findings.

My data suggest two additional explanations for why qualified immunity disposes of so few cases: the doctrine is not well suited to dismiss many claims before trial, and qualified immunity is often unnecessary to serve its intended role.

1. Qualified Immunity Is Ill Suited To Dispose of Cases

Qualified immunity motions are infrequently dispositive in part because the doctrine is ill suited to dispose of many cases before trial. Although qualified immunity doctrine creates a seemingly insurmountable barrier for plaintiffs, the standards for review at the motion to dismiss and summary judgment stages may prevent courts from granting defendants’ motions. At the motion to dismiss stage, a defendant’s qualified immunity motion should be denied so long as the plaintiff has plausibly alleged a violation of a clearly established right. As one district judge from the Middle District of Tennessee observed,

The rationale for the existence of qualified immunity is to avoid imposing needless discovery costs upon government officials, so determining whether the immunity applies must be made at an early stage in the litigation. At the same time, the determination of qualified immunity is usually dependent on the facts of the case, and, at the pleadings stage of the litigation, there is scant factual record available to the court. Since plaintiffs are not required to anticipate a qualified immunity defense in their pleadings, and since at this stage of the litigation the exact contours of the right at issue—and thus the degree to which it is clearly established—are unclear, the Sixth Circuit advises that qualified immunity should usually be determined pursuant to a summary judgment motion rather than a motion to dismiss.128

This is a common refrain in circuit courts across the country129 and decisions in my dataset.130

District courts also find that factual disputes prevent resolution on qualified immunity grounds at summary judgment. Alan Chen has argued that the Supreme Court’s qualified immunity decisions “have embedded a central paradox into the doctrine”: although the Court repeatedly writes that “qualified immunity claims can and should be resolved at the earliest stages of litigation,” it ignores the fact that these determinations “inherently entail nuanced, fact-sensitive, case-by-case determinations involving the application of general legal principles to a particular context.”131 My data offer anecdotal evidence to support Chen’s observation. In the five districts in my study, courts repeatedly found that factual disputes prevented summary judgment on qualified immunity grounds.132 In these decisions, courts duly recited the benefits of resolving qualified immunity at the earliest possible stage and qualified immunity’s intended role as protection from discovery and trial. Yet the same courts found that factual disputes made summary judgment inappropriate.

The Supreme Court’s recent decision in White v. Pauly provides additional anecdotal evidence of this underappreciated phenomenon. In White v. Pauly, the Supreme Court held that it would be appropriate to grant summary judgment on qualified immunity grounds to an officer who shot and killed a suspect without first identifying himself and ordering the suspect to drop his gun, because “[n]o settled Fourth Amendment principle requires that officer to second-guess the earlier steps already taken by his or her fellow officers in instances like the one [the defendant] confronted here.”133 The decision has been described as evidence that the Supreme Court “wants fewer lawsuits against police to go forward.”134 This may well be true. Yet the decision in White v. Pauly did not end Daniel Pauly’s lawsuit; as Justice Ginsburg notes in her concurrence, the Court’s decision “leaves open the propriety of denying summary judgment” based on various factual disputes about the officer’s conduct.135

Plaintiffs’ decisions about how to frame their cases also make qualified immunity ill suited to dispose of many cases. Defendants could not raise a qualified immunity defense in 8.4% of the cases in my study because the plaintiffs did not sue an individual officer for money damages.136 Even in cases in which defendants could raise qualified immunity, plaintiffs’ other pleading decisions sometimes diminished the impact of qualified immunity. In the vast majority of cases asserting claims against individual officers for money damages, plaintiffs also included claims against municipalities, claims for injunctive relief, and/or state law claims that could not be dismissed on qualified immunity grounds.137 Even when a plaintiff brings a claim for damages against an individual defendant (for which qualified immunity is available), the defendant raises a qualified immunity defense, and the court grants the motion, claims against the municipality, claims for injunctive relief, and state law claims may remain.138

2. Qualified Immunity Is Unnecessary To Dispose of Cases

My data also suggest that qualified immunity may lead to the dismissal of few cases because cases are so often resolved on other grounds. Qualified immunity could not be raised in 126 (10.7%) of the cases in my study because the judges dismissed the cases sua sponte before the defendants could answer or otherwise respond.139 In these cases, qualified immunity doctrine was unnecessary to shield defendants from discovery and trial.

Qualified immunity was also often unnecessary to dispose of cases at the motion to dismiss stage. Defendants in the cases in my dataset clearly held this view: even when defendants could raise qualified immunity at the motion to dismiss stage, they often chose not to do so.140 More often than not, when defendants moved to dismiss or for judgment on the pleadings, they did not include a qualified immunity argument. Instead, defendants moved to dismiss for failure to plead plausible claims for relief or failure to assert a constitutional violation, among other grounds. Even when defendants raised qualified immunity at the motion to dismiss stage, and courts concluded that the cases should be dismissed, courts often resolved the motions on other grounds. Courts granted, in whole or part, seventy-nine (51.3%) out of the 154 motions to dismiss or for judgment on the pleadings that raised qualified immunity. Of those seventy-nine grants, twenty-one (26.6%) were granted on qualified immunity grounds, and fifty-eight (73.4%) were granted on grounds other than qualified immunity.141

Qualified immunity played a more substantial role at summary judgment. Defendants raised qualified immunity arguments in most of their summary judgment motions.142 And when courts granted defendants’ summary judgment motions in whole or part, they relied on qualified immunity 39.7% of the time.143 Still, courts decided a clear majority of the motions on other grounds. Most often, these summary judgment motions were granted in whole or part because the plaintiff could not establish a genuine dispute about a material question of fact. This finding should not come as a surprise to at least one member of the Court—Justice Kennedy noted in Wyatt v. Cole that the Court’s summary judgment decisions reduced the need for qualified immunity to shield government officials from trial. As Justice Kennedy explained:

Harlow was decided at a time when the standards applicable to summary judgment made it difficult for a defendant to secure summary judgment regarding a factual question such as subjective intent, even when the plaintiff bore the burden of proof on the question; and in Harlow we relied on that fact in adopting an objective standard for qualified immunity. However, subsequent clarifications to summary-judgment law have alleviated that problem . . . . Under the principles set forth in Celotex and related cases, the strength of factual allegations such as subjective bad faith can be tested at the summary-judgment stage.144

When the Supreme Court discusses qualified immunity, it appears to presume that qualified immunity is the only barrier standing between government officials and discovery and trial. Instead, my study illustrates that there are other tools that parties can—and often do—use to resolve Section 1983 cases before trial.145

***

In this Section, I have offered some possible explanations for why cases are infrequently dismissed on qualified immunity grounds. This phenomenon is not solely attributable to plaintiffs’ decisions not to file cases in which qualified immunity motions might be successful. Nor can lower courts be shouldered with all the blame for the low rate of qualified immunity dispositions. Instead, my data suggest that qualified immunity doctrine is ill suited in some cases and unnecessary in others to serve its intended role.

My data also make clear that qualified immunity’s role in Section 1983 litigation is the product of decisions made by multiple actors—judges, defendants, plaintiffs, and the litigants’ attorneys. Moreover, there is at least some evidence to suggest that district judges’ varying inclinations to grant qualified immunity motions may influence defendants’ and plaintiffs’ litigation decisions. In jurisdictions with judges who most often granted defendants’ qualified immunity motions—the Southern District of Texas and the Middle District of Florida—defendants brought qualified immunity motions more frequently, and plaintiffs more frequently crafted their cases in ways that prevented defendants from raising the defense. Conversely, in jurisdictions with judges who less frequently granted defendants’ qualified immunity motions—the Eastern District of Pennsylvania and the Northern District of California—defendants less frequently brought qualified immunity motions, and plaintiffs less frequently crafted their cases to avoid the defense. A complete understanding of the role of qualified immunity in constitutional litigation against law enforcement must attend to regional differences in the dynamic interactions between judges, defendants, and plaintiffs. I plan to explore these interactions in future work.

C. Implications for the Balance Struck by Qualified Immunity

The Supreme Court has explained that qualified immunity is intended to balance “the need to hold government officials accountable when they exercise power irresponsibly and the need to shield officials from harassment, distraction, and liability when they perform their duties reasonably.”146 Many have argued, and I agree, that the Court’s qualified immunity doctrine puts a heavy thumb on the scale in favor of government interests, and disregards the interests of individuals whose rights have been violated.147 My research offers an additional reason to believe that the Supreme Court has gotten the balance wrong: qualified immunity doctrine does not appear to be necessary or well suited to protect government officials “from harassment, distraction, and liability when they perform their duties reasonably.”148 This observation makes it even more difficult to justify the burdens the doctrine appears to place on plaintiffs.

1. Interests in Protecting Government Officials

The Supreme Court explained in Harlow that qualified immunity was necessary to protect government officials from four harms: 1) “the expenses of litigation”; 2) “the diversion of official energy from pressing public issues”; 3) “the danger that fear of being sued will ‘dampen the ardor of all but the most resolute, or the most irresponsible [public officials], in the unflinching discharge of their duties’”; and 4) “the deterrence of able citizens from acceptance of public office.”149 The Court has relied on no empirical evidence to support its conclusions that these threats exist, or that qualified immunity can protect against them. Although questions remain about the government interests served by qualified immunity, this study and my prior research suggest that qualified immunity doctrine is often unfit to protect against some of these harms, and often unnecessary to protect against others.

The first—and frequently repeated—justification for qualified immunity is that it protects government officials from the burdens of financial liability. But my prior research has shown that qualified immunity is unnecessary to serve this role—virtually all law enforcement defendants are provided with counsel free of charge, and are indemnified for settlements and judgments entered against them. In the six-year period from 2006 to 2011, law enforcement officers in forty-four of the seventy largest law enforcement agencies paid just 0.02% of the dollars awarded to plaintiffs in police misconduct suits.150 In thirty-seven small and midsized agencies, no officer contributed to settlements or judgments to plaintiffs awarded during this period. Officers were indemnified even when they were disciplined, fired, and criminally prosecuted for their misconduct. And no officer paid a penny of the punitive damages awarded to plaintiffs in these jurisdictions. I could confirm only two jurisdictions in which officers contributed to settlements and judgments during the study period—New York City and Cleveland.151 In these jurisdictions, the median contribution was $2,250, and no officer contributed more than $25,000.152 Given this evidence, qualified immunity cannot be justified as a means of protecting officers from personal financial liability.

In recent years, the Supreme Court has described “the ‘driving force’ behind creation of the qualified immunity doctrine” to be resolving “‘insubstantial claims’ against government officials . . . prior to discovery.”153 But qualified immunity resulted in the dismissal of just 0.6% of the cases in my dataset before discovery, and resulted in the dismissal of just 3.2% of the 1,183 cases in my dataset before trial.

Indeed, qualified immunity may actually increase the costs and delays associated with Section 1983 litigation. Although qualified immunity terminated only 3.9% of the 979 cases in my dataset in which qualified immunity could be raised, the defense was in fact raised by defendants in more than 37% of these cases—and was sometimes raised multiple times, at the motion to dismiss stage, at summary judgment, and through interlocutory appeals.154 Each time qualified immunity is raised, it must be researched, briefed, and argued by the parties and decided by the judge. And litigating qualified immunity is no small feat. John Jeffries describes qualified immunity doctrine as “a mare’s nest of complexity and confusion.”155 Lower courts are “hopelessly conflicted both within and among themselves” as a result.156 One circuit court judge reported that “[w]ading through the doctrine of qualified immunity is one of the most morally and conceptually challenging tasks federal appellate court judges routinely face.”157

The time and effort necessary to resolve qualified immunity motions could nevertheless further the goals of qualified immunity doctrine if it effectively protected defendants from discovery and trial. But in the five districts in my study, just 8.6% of qualified immunity motions brought by defendants in my docket dataset resulted in case dismissals.158 The remaining 91.4% of qualified immunity motions brought by defendants required the parties and judges to dedicate time and resources to briefing, arguing, and deciding the motions without shielding defendants from discovery and trial.

Even in the cases in which qualified immunity motions resulted in case dismissals, it is far from certain that qualified immunity saved the parties and the courts time. As Alan Chen has observed, when considering the efficiencies of qualified immunity, “the costs eliminated by resolving the case prior to trial must be compared to the costs of trying the case . . . . [T]he pretrial litigation costs caused by the invoking of the immunity defense may cancel out the trial costs saved by that defense.”159 In this study, I have not calculated how much time was spent litigating qualified immunity motions, or compared that time with the amount of time spent preparing for and conducting a trial. Yet—given the complexity of qualified immunity doctrine, the use of interlocutory appeals of qualified immunity denials, the fact that most trials in my docket dataset lasted just a few days, and the possibility that a case will settle instead of going to trial even when qualified immunity is denied—the aggregate benefits of qualified immunity do not necessarily outweigh its costs for government officials.

In Pearson, the Supreme Court wrote that the Saucier two-step qualified immunity analysis “‘disserve[s] the purpose of qualified immunity’ when it ‘forces the parties to endure additional burdens of suit—such as the costs of litigating constitutional questions and delays attributable to resolving them—when the suit otherwise could be disposed of more readily.’”160 Given the costs and delays associated with qualified immunity motion practice and the infrequency with which qualified immunity motions terminate Section 1983 cases, the doctrine arguably disserves its own purposes.

Although qualified immunity doctrine appears to do little to shield defendants from burdens associated with litigation in filed cases—and may in fact increase the amount of time spent on a substantial number of those cases—my data leave open the possibility that qualified immunity doctrine shields government officials from burdens associated with discovery and trial in other ways, namely by causing people never to file insubstantial claims or to settle them quickly.161 The possibility that qualified immunity doctrine serves its intended purpose in these ways, however, does not mean that it actually does. At least two pressing questions would have to be answered before qualified immunity doctrine could be justified on these grounds.

First, what are the merits of cases that are never filed or settled quickly because of qualified immunity? If qualified immunity doctrine discourages people from filing or pursuing insubstantial cases, the doctrine is meeting its express goals.162 But if the doctrine discourages people from filing or pursuing meritorious cases because the briefing and interlocutory appeals associated with qualified immunity would be too expensive, the doctrine is not sorting cases in the way anticipated by the Court. Although more research is necessary to answer this question, available evidence offers reason for concern. Alexander Reinert’s interviews with attorneys who bring Bivens actions suggest that people with strong claims may sometimes be unable to find a lawyer because the cost of litigating qualified immunity is too high or because the conduct at issue has not been clearly established by prior cases.163 Some people who do file their cases may settle at a discount, not because their cases are weak but because they cannot afford to litigate qualified immunity in the district court or on interlocutory appeal.

Second, how frequently does qualified immunity cause plaintiffs not to file or to settle insubstantial cases? The costs associated with litigating qualified immunity and the difficulty of overcoming a qualified immunity motion may cause plaintiffs not to file some insubstantial cases. But other, independent considerations may cause plaintiffs not to file such cases, including rigorous pleading requirements, stringent standards for proving underlying constitutional violations, and minimal potential damages awards. Without further study, it is not possible to conclude that qualified immunity, rather than these alternative considerations, is responsible for plaintiffs’ decisions to settle or never file insubstantial cases.

In short, there is limited evidence to support the hypothesis that qualified immunity serves its purpose through screening cases or coercing settlement. Indeed, some evidence suggests that the doctrine may be discouraging plaintiffs from filing or pursuing meritorious cases because qualified immunity would take too long or cost too much to litigate. Our existing knowledge about qualified immunity’s effects on filing and settlement decisions cannot justify the doctrine on these grounds.

The Supreme Court has mentioned, but dwelled little upon, two other possible benefits of qualified immunity doctrine—that it lessens “the danger that fear of being sued will ‘dampen the ardor of all but the most resolute, or the most irresponsible [public officials], in the unflinching discharge of their duties’” and that it mitigates “the deterrence of able citizens from acceptance of public office.”164 The available evidence casts doubt on these rationales as well. The Court has written that dangers of overdeterrence should dissipate for officials who are not financially responsible for settlements and judgments.165 Consistent with this observation, studies have found that “the prospect of civil liability has a deterrent effect in the abstract survey environment but that it does not have a major impact on field practices.”166 Further, civil liability does not appear to play a sizable role in people’s decisions to apply to become police officers. Police departments around the country report difficulties finding recruits, but the long list of reasons police officials believe people are not applying does not include the threat of being sued.167 These speculative benefits cannot justify qualified immunity’s highly restrictive standards.

Perhaps the Court believes that qualified immunity doctrine serves other interests that it has failed to mention. Even if officers are almost always indemnified, and cases are rarely dismissed on qualified immunity grounds, qualified immunity doctrine may somehow reduce the costs of litigation for the municipalities that end up paying the settlements and judgments on behalf of their officers.168 Qualified immunity doctrine may encourage the development of constitutional law because it allows courts to announce new constitutional rules without fear of subjecting defendants to financial liability.169 In this Article, I do not evaluate the sensibility of—or empirical support for—these alternative justifications for qualified immunity. Neither has been relied upon by the Court. If these or other policy interests motivate the Supreme Court’s qualified immunity jurisprudence, the Court should be explicit about those motivations so that courts, practitioners, and scholars can evaluate the sensibility of these interests and measure the extent to which qualified immunity advances them. Until then, we are left with the justifications for qualified immunity doctrine that the Court has offered. Available evidence suggests that the doctrine is unnecessary to serve some of qualified immunity’s key goals and ill suited for others.

2. Interests in Government Accountability

My research indicates that filed lawsuits are rarely dismissed on qualified immunity grounds. As I have argued, this finding suggests that qualified immunity doctrine rarely achieves its intended function as a shield for government officials against discovery and trial in filed cases. What are the implications of this finding for the other side of qualified immunity’s balance, described by the Court both as “the importance of a damages remedy to protect the rights of citizens”170 and as “the need to hold public officials accountable when they exercise power irresponsibly”?171 Commentators have long criticized qualified immunity doctrine for protecting government officials at the expense of Section 1983’s accountability goals. If qualified immunity is not doing much to protect government officials, does that allay concerns that the doctrine compromises government accountability? In other words, do my data suggest that qualified immunity does little of great significance, either to defendants’ benefit or to plaintiffs’ detriment?

Evidence that few cases are dismissed on qualified immunity grounds suggests that the direst descriptions of qualified immunity’s impact on plaintiffs perhaps go too far. Critics assert that qualified immunity closes the courthouse door for plaintiffs.172 And there is no shortage of decisions by the Supreme Court and lower courts dismissing cases on qualified immunity grounds.173 Yet, my study suggests that qualified immunity doctrine appears to close the courthouse door far less frequently than critics have assumed—at least once a case is filed. My findings do not, however, undermine other concerns raised about the impact of qualified immunity on plaintiffs’ claims. Qualified immunity could significantly damage law enforcement accountability without protecting officials from the burdens of discovery and trial.

First, qualified immunity doctrine may discourage people from filing their cases or may cause them to settle or withdraw their claims.174 If qualified immunity had this effect only on insubstantial cases, the doctrine would be achieving its intended role, albeit in a manner unexpected by the Court. But if qualified immunity is causing people not to file or to settle meritorious cases, as available anecdotal evidence suggests, then the doctrine is preventing people from vindicating their rights and holding government accountable.175

Moreover, my findings do not undermine other common critiques of the doctrine. Qualified immunity doctrine has been criticized by courts and scholars alike for being confusing and difficult to apply, and leading to inconsistent adjudications.176 These characteristics of qualified immunity doctrine may well increase the time it takes courts to decide qualified immunity motions, even as the decisions are infrequently dispositive.177

In addition, many are critical of the Court’s decision in Pearson to allow lower courts to grant qualified immunity without first assessing whether a defendant violated the constitutional or statutory rights of the plaintiff.178 Their fear is that if courts regularly find that the law is not clearly established without first ruling on the scope of the underlying constitutional right, the constitutional right at issue will never become clearly established. This catch-22 may lead to constitutional uncertainty and stagnation, making it more difficult for plaintiffs to prevail on constitutional claims and offering little guidance to government officials about the scope of constitutional rights.179 Scholars who have studied the impact of Pearson have found some evidence to support these concerns.180 The fact that few cases are dismissed on qualified immunity grounds is immaterial to this critique. The Supreme Court’s decision in Pearson to allow lower courts to grant qualified immunity without deciding whether a right has been violated may still lead to constitutional uncertainty, particularly in cases involving new technologies or practices.181

Finally, many have argued that the Supreme Court’s qualified immunity decisions protect bad actors. The Court’s disregard of subjective intent protects officers who act in bad faith, so long as their conduct does not violate clearly established law.182 In addition, a government official who has acted in a clearly unconstitutional manner can be shielded from liability simply because no prior case has held similar conduct to be unconstitutional. The Supreme Court’s recent decisions have made it increasingly difficult to meet this standard.183 It is, as John Jeffries has written, “as if the one-bite rule for bad dogs started over with every change in weather conditions.”184 Even this critique of qualified immunity is left largely intact by my findings. Qualified immunity’s disregard for officials’ subjective intent, and the need for precedent that “place[s] the statutory or constitutional question beyond debate,”185 may insulate bad actors from financial liability, but still expose them to discovery and trial if other claims or defendants remain.

McKay v. City of Hayward,186 a case from the Northern District of California in my docket dataset, illustrates how qualified immunity can impair government accountability in these ways without shielding defendants from discovery or trial. On May 29, 2011, officers from the Hayward Police Department used a police dog to track an armed suspect who had robbed a restaurant.187 The dog guided the officers to an eight-foot wall. Without any warning, the officers lifted the dog over the wall. On the other side of the wall was the backyard of a mobile home belonging to Jesse Porter, an 89-year-old who had no connection to the robbery. The dog bit Porter on the leg, leaving a wound so severe that Porter’s leg had to be amputated. Mr. Porter was then moved into a residential care facility, where he died two months later. Mr. Porter’s children sued the involved officers and the City of Hayward under federal and state law.

At summary judgment, the district court in McKay granted the officers qualified immunity.188 The court found that, to survive summary judgment, the plaintiffs had to be able to show that the failure to warn before seizure by a police dog constitutes a Fourth Amendment violation. The court surveyed Ninth Circuit cases involving police dogs and found that “[n]o Ninth Circuit case holds explicitly that failure to warn before seizure by a police dog constitutes a violation of the Fourth Amendment.”189 The court surveyed other circuits and found some variation: the Fourth and Eighth Circuits had held that the failure to give a warning before using a police dog violates the Fourth Amendment, but the Eleventh, Seventh, and Tenth Circuits had held that failure to warn before deploying a police dog was “not dispositive of the reasonableness of seizing an individual with a police dog.”190 Because of this variation among circuits, the court in the Northern District of California concluded that that the unconstitutionality of the officers’ conduct had not been clearly established.

The decision granting qualified immunity in McKay did not shield government officials from burdens associated with either discovery or trial. In McKay’s case, qualified immunity was raised at summary judgment, after the officers had already participated in discovery. The motion was granted less than two weeks before trial was scheduled to begin.191 Moreover, even after the court granted qualified immunity to the individual officers, the officers still faced the prospect of trial. In addition to the Section 1983 claims against the two individual officers, the plaintiffs brought state law claims against the individual officers and state and federal claims against the City—the qualified immunity defense did not apply to any of these claims.192 In the days following the court’s summary judgment decision, the parties drafted and submitted voir dire questions, multiple motions in limine, and briefs regarding whether the trial should be separated into three stages.193 The case settled and the court entered a conditional dismissal the day trial was scheduled to begin.194

Although the district court’s qualified immunity decision in McKay did not shield officials from discovery and was not formally the reason the case did not go to trial, it did negatively affect interests in government accountability. The qualified immunity motion likely increased the amount of time spent by the attorneys for the plaintiffs and defendants.195 The grant of qualified immunity in McKay may also have ripple effects that extend far beyond the parties to the litigation. The district court found that it was not clearly established in the Ninth Circuit that deploying police dogs without a prior warning violates the Constitution. This decision may cause lawyers to decline to represent people with similar claims. One could argue that qualified immunity is serving its intended role by discouraging people from bringing Section 1983 cases when the underlying constitutional rights have not been clearly established. But this position goes further than the Court’s own justification for qualified immunity doctrine: to protect government officials from insubstantial claims.196 That no prior court has decided a given constitutional issue does not imply that a case raising it lacks merit.

Uncertainty about the constitutionality of deploying a police dog without a prior warning may also influence police departments’ policy and training decisions. Although the Supreme Court appears confident that police departments can regulate themselves,197 police officials look to court decisions to guide their policies and trainings.198 Were, for example, the Ninth Circuit to hold that officers should give prior warnings before using police dogs, departments in the jurisdiction of the Ninth Circuit would likely train their officers to issue warnings under these circumstances. Without such a decision, and with the McKay court’s conclusion that there is no clearly established constitutional right to such a warning, departments may be less likely to train their officers to give such warnings.199 These costs to government accountability accrue whether or not qualified immunity protects government officials from discovery and trial.

D. Moving Forward

The Supreme Court has written that evidence undermining its assumptions about the realities of constitutional litigation might “justify reconsideration of the balance struck” in its qualified immunity decisions.200 My research has, indeed, undermined the Court’s assumptions about the purposes served by qualified immunity doctrine. In this Section, I consider how these findings should shape qualified immunity doctrine moving forward.

My findings suggest that the Court’s efforts to advance its policy goals through qualified immunity doctrine has been an exercise in futility. In Harlow v. Fitzgerald, the Supreme Court “completely reformulated qualified immunity along principles not at all embodied in the common law, replacing the inquiry into subjective malice so frequently required at common law with an objective inquiry into the legal reasonableness of the official action.”201 The Court believed that “[t]he transformation was justified by the special policy concerns arising from public officials’ exposure to repeated suits.”202 Some—including Justice Thomas—have argued that this transformation was a mistake because the scope of qualified immunity doctrine should mirror the common law defenses that existed in 1871, and should not reflect the Court’s policy preferences at all.203

This Article offers an additional reason to conclude this transformation was a mistake: the doctrine does not serve its intended policy objectives. Although the Supreme Court repeatedly describes qualified immunity doctrine as a means of shielding government officials from the costs and burdens of litigation, I have found officers are virtually always indemnified, and that qualified immunity is rarely the reason that Section 1983 cases end. Future research can explore whether qualified immunity causes plaintiffs not to file or pursue insubstantial claims, or advances the doctrine’s goals in other ways. At this point, however, available evidence contradicts the Court’s assumptions about the role qualified immunity plays in constitutional litigation.

Justices sympathetic to qualified immunity’s policy goals might conclude based on my findings that they should further strengthen qualified immunity doctrine to protect defendants. I would discourage this approach for several reasons. First, it is far from clear that qualified immunity doctrine is well designed to weed out only “insubstantial” cases. Available evidence suggests that some people may decline to file or pursue their claims because of the cost of litigating qualified immunity, even when they might succeed on the merits.204 And cases alleging clearly unconstitutional behavior may be dismissed on qualified immunity grounds simply because no prior case has held sufficiently similar conduct to be unconstitutional.205 Strengthening qualified immunity doctrine would presumably aggravate these preexisting concerns.

Setting aside the question of whether such a shift is desirable, I am not convinced that it is feasible. It is hard to imagine how the Court could make qualified immunity doctrine any stronger than it already is.206 Perhaps members of the Court believe that lower courts are not applying qualified immunity doctrine as expansively as they should. Indeed, the Court’s flurry of recent summary reversals suggests that it is attempting to encourage lower courts to follow course.207

But even if all judges applied qualified immunity doctrine as expansively as does the Supreme Court, qualified immunity doctrine would likely still fall short of its intended role in many cases filed against law enforcement. Plaintiffs could often still plead a plausible entitlement to relief at the motion to dismiss stage, and could often still raise factual disputes at summary judgment that prevent dismissal on qualified immunity grounds. Plaintiffs would continue to include claims against municipalities, claims for declaratory or injunctive relief, and state law claims in their cases that qualified immunity cannot resolve.208 Defendants would still sometimes conclude that other defenses or an inexpensive settlement is preferable to the added costs of qualified immunity motion practice. And courts would continue to dismiss cases for multiple other reasons besides qualified immunity. Presumably the number of cases dismissed on qualified immunity grounds would increase somewhat, but given litigation dynamics and other applicable doctrines, many cases would remain in which qualified immunity never shielded government officials from discovery or trial. Qualified immunity is the Supreme Court’s hammer. But many civil rights damages actions against law enforcement are not nails.

The fact that qualified immunity is often ill suited and unnecessary to advance the Court’s policy objectives provides additional reason to adopt Justice Thomas’s view and realign the doctrine with historical common law defenses. According to those who have studied the common law at the time Section 1983 was passed, little would remain of qualified immunity if the Court adopted this approach.209 But other defenses would remain—including arguments that plaintiffs cannot state plausible claims for relief in their complaint or cannot establish material factual disputes at summary judgment. Defendants would still be able to argue that plaintiffs cannot meet the Court’s exceedingly rigorous standards for constitutional violations.210 Even in the absence of qualified immunity, these other procedural and substantive barriers would prevent many Section 1983 cases from being filed, proceeding to discovery, or advancing to trial.

If the Court is unwilling to eliminate or dramatically restrict qualified immunity, it could make more modest alterations that would align the doctrine with evidence of its role in constitutional litigation. For example, the Court could undo adjustments to qualified immunity doctrine that were expressly motivated by an interest in shielding government officials from discovery and trial in filed cases. In Harlow, the Court eliminated consideration of officers’ subjective intent because it believed doing so would “avoid ‘subject[ing] government officials either to the costs of trial or to the burdens of broad-reaching discovery’ in cases where the legal norms the officials are alleged to have violated were not clearly established at the time.”211 My study shows that the Court’s elimination of the subjective prong of qualified immunity in Harlow should be viewed as a failed experiment. Despite courts’ and commentators’ assumptions to the contrary,212 the decision in Harlow appears to have done little to shield government officials from discovery and trial in filed cases.

Restoring the subjective prong to qualified immunity analysis could also mitigate at least one serious concern with the doctrine. Currently, government officials acting in bad faith or with knowledge of the unconstitutionality of their behavior can be shielded from liability simply because no prior case proscribed their conduct. If the subjective prong were restored to the qualified immunity analysis, government officials would not be entitled to qualified immunity if they knew or should have known that their conduct was unlawful. A recent Supreme Court case, Mullenix v. Luna, illustrates how reversing Harlow might address this concern.213

The facts relevant to Mullenix began when Tulia Police Department officers attempted to arrest Israel Leija, Jr. for violating misdemeanor probation.214 Leija fled the scene in his car, and officers from several agencies participated in the pursuit. Officers set up spike strips on the highway to puncture Leija’s tires as he drove by—a strategy they had been trained to use in just this type of situation. Texas Department of Public Safety Trooper Chadrin Mullenix decided that instead of setting up spike strips he would try to disable Leija’s car by shooting at it.215 He had received no training in shooting at a car to disable it and was instructed by his supervisor not to do so.216 Nevertheless, Mullenix fired six rounds at Leija’s car as it passed under the bridge where Mullenix was standing. Leija died, with one of the shots determined to be the cause of death.217 Soon after the shooting, Mullenix remarked to his supervisor, “How’s that for proactive?”—an apparent reference to a conversation they had had early in the day in which the superior had criticized the officer for not taking enough initiative.218

The district court denied Mullenix’s motion for summary judgment on qualified immunity grounds, Mullenix filed an interlocutory appeal, and the Fifth Circuit affirmed the district court. The Supreme Court granted Mullenix’s petition for certiorari and reversed. The Court did not answer whether Mullenix violated the Constitution but instead held that prior cases had not clearly established that his conduct was unconstitutional.219 Mullenix’s remark to his supervisor played no role in the analysis, as “an officer’s actual intentions are irrelevant” to the qualified immunity analysis.220 Restoring the subjective prong to the qualified immunity analysis would likely change the outcome of a case like Mullenix. Mullenix’s “How’s that for proactive?” statement would once again be relevant to the qualified immunity analysis, and would constitute at least triable evidence of bad faith.221

The Court could also reconsider other adjustments to qualified immunity made with the express goal of shielding defendants from burdens of discovery and trial. For example, the Court granted defendants the right to immediately appeal denials of qualified immunity as a means of shielding defendants from burdens of discovery and trial.222 Yet my data show interlocutory appeals of qualified immunity denials infrequently serve that function. Defendants filed interlocutory appeals of 21.7% of decisions denying qualified immunity in whole or part. Of the appeals that were filed, just 12.2% of the lower court decisions were reversed in whole, and just 9.8% of the interlocutory appeals filed resulted in case dismissals. Interlocutory appeals may have prompted case resolutions in another way—39.0% of interlocutory appeals were never decided, apparently because the cases were settled while the motions were pending.223 But defendants’ interlocutory appeals rarely resulted in case dismissals on qualified immunity grounds. It is far from clear that interlocutory appeals shield defendants from litigation burdens—the time and money spent briefing and arguing interlocutory appeals may in fact exceed the time and money saved in the relatively few reversals on interlocutory appeal. If so, the policy objectives motivating Mitchell militate in favor of eliminating the right of interlocutory appeal.

Finally, and still more modestly, the Court could reconsider the restrictive manner in which it defines “clearly established law.” John Jeffries has written that the Court’s narrow definition of clearly established law is inspired by its interest in facilitating qualified immunity dismissals at summary judgment.224 My data show that the Court’s decisions are not having their intended effect. Yet, as others have pointed out, the Court’s doctrinal framework creates confusion in the lower courts and protects bad actors when there is no prior case on point.225 Jeffries’s proposed solution is to focus the qualified immunity inquiry not on whether the law was clearly established but, instead, on whether the defendant’s conduct was “clearly unconstitutional.”226 I believe that my data support a more complete transformation of the doctrine, but this adjustment would at least be a step in the right direction.

At this point, it is impossible to predict what impact these proposed changes to qualified immunity doctrine would have on the litigation of constitutional claims against law enforcement. Perhaps narrowing the qualified immunity defense, restoring the subjective prong, or eliminating qualified immunity altogether would dramatically increase the number of suits filed against the police, or increase the number of filed cases that were settled or tried. On the other hand, these changes might inspire courts to place other limits on Section 1983 claims to maintain the status quo.227 This Article does not predict how changes to qualified immunity doctrine might influence the collection of doctrines relevant to constitutional litigation, or suggest the ideal ways in which they should relate. My suggestions are motivated by a less lofty ambition—to achieve greater consistency across qualified immunity doctrine’s structure, intended policy goals, and actual role in constitutional litigation.

Conclusion

In recent years, the Supreme Court has dedicated an outsized portion of its docket to qualified immunity motions in cases against law enforcement because, it has explained, the doctrine is so “important to ‘society as a whole.’”228 But the Court relies on no evidence to back up this fervently held position. Instead, my research shows that qualified immunity doctrine infrequently plays its intended role in the litigation of constitutional claims against law enforcement. Qualified immunity doctrine is unnecessary to shield law enforcement officers from financial liability, and the doctrine infrequently protects government officials from burdens associated with discovery and trial in filed cases. Further exploration of dynamics unobservable through my dataset could reveal other ways in which qualified immunity influences the litigation of civil rights actions against law enforcement. At this point, however, available evidence indicates that qualified immunity often is not functioning as assumed, and is not achieving its intended goals. In an ideal world, all open empirical questions about Section 1983 litigation would be answered before any applicable doctrine was adjusted. But it is my view that the perfect should not be the enemy of the good.229 The Supreme Court, as well as lower courts, should adjust their qualified immunity decisions to comport with this evidence.