Stanford Law Review-The Myth of Personal Liability, March 2020
Download original document:
Document text
Document text
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Stanford Law Review Volume 72 March 2020 ARTICLE The Myth of Personal Liability: Who Pays When Bivens Claims Succeed James E. Pfander, Alexander A. Reinert & Joanna C. Schwartz* Abstract. In Bivens v. Six Unknown Named Agents, the Supreme Court held that federal law creates a right to sue federal officials for Fourth Amendment violations. For the last three decades, however, the Court has cited the threat of individual liability and the burden of government indemnification on agency budgets as twin bases for narrowing the right of victims to secure redress under Bivens. In its most recent decisions, Ziglar v. Abbasi and Hernandez v. Mesa, the Court said much to confirm that it now views personal liability less as a feature of the Bivens liability rule than as a bug. But, to date, there has been no empirical examination of who pays when Bivens claims succeed. This Article studies the financial threat that successful Bivens claims pose to federal officers and their employing federal agency. Information supplied by the Federal Bureau of Prisons in response to a Freedom of Information Act request identified successful Bivens actions over a ten-year period; in the vast majority of cases (over 95%), individual defendants contributed no personal resources to the resolution of the claims. Nor did the responsible federal agency pay the claims through indemnification. The data suggest, in short, that recent hostility to Bivens litigation rests on a perceived threat of personal liability that is much more theoretical than real. The data also raise important questions about the adequacy of existing constitutional remedies and the manner in which the Department of Justice exercises its settlement authority under the Federal Tort Claims Act and the Judgment Fund. * James E. Pfander is the Owen L. Coon Professor of Law, Northwestern Pritzker School of Law; Alexander A. Reinert is the Max Freund Professor of Litigation & Advocacy and the Director of the Center for Rights and Justice, Benjamin N. Cardozo School of Law; and Joanna C. Schwartz is a Professor of Law, UCLA School of Law. We thank Dick Fallon, Paul Figley, Greg Sisk, and participants in faculty workshops at the Northwestern Pritzker School of Law and the Benjamin N. Cardozo School of Law for comments on early drafts. We are also grateful to Wade Formo for able research assistance, and to Ethan Amaker, Hannah Begley, Nicole Collins, Lori Ding, Nathan Lange, Elizabeth Reetz, Aletha Smith, Gregory Terryn, Reid Whitaker, and the editors of the Stanford Law Review for excellent editorial assistance. 561 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Table of Contents Introduction ............................................................................................................................................................563 I. Constitutional Torts, Personal Liability, and Ziglar ...............................................................569 A. Constitutional Claims and the Bivens Action ...................................................................569 B. Suits for Damages Under the Federal Tort Claims Act ..............................................571 C. Personal Liability and the Decline of the Bivens Action .............................................574 II. Bivens in Practice: A Study of Who Pays to Resolve Well-Grounded Constitutional Tort Claims ..................................................................................................................578 A. Cases in Which Employees Contributed to the Payment of Settlements and Judgments ...................................................................................................................................580 B. Cases in Which Employees Made No Compensating Payments ..........................583 1. Dismissal of Bivens claims during litigation ...........................................................584 2. Settlement of Bivens and FTCA claims with payment by the U.S. government ..............................................................................................................................585 3. Substitution of FTCA claims for Bivens claims as a condition of settlement...................................................................................................................................586 a. Formal substitution ...................................................................................................587 b. Informal substitution ................................................................................................593 C. The Impact of Judgment Fund Payments on the Bureau of Prisons ...................594 III. Implications ...................................................................................................................................................596 A. Constitutional Torts and Individual Deterrence After Ziglar ................................597 B. Agency Incentives and the Role of Indemnification....................................................601 C. The Department of Justice Narrative of Personal Liability ....................................605 D. Payment Practices, Transparency, and Congressional Oversight of the Judgment Fund..................................................................................................................................609 E. On the Need for Future Research and the Uncertain Future of the Bivens Action .....................................................................................................................................................615 Conclusion................................................................................................................................................................621 Appendix A: Methodology ..............................................................................................................................623 Appendix B: Data ..................................................................................................................................................627 562 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Introduction Since its 1971 decision in Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics,1 the Supreme Court has been of two minds about the impact of constitutional tort litigation on the workaday incentives of federal officials. On one side, the Court has emphasized the importance of deterring constitutional violations through the imposition of personal, tort-based liability payable by the officer herself.2 On the other side, the Court has increasingly worried about the burden of such liability, fearing it will overdeter federal officials and undermine the government’s ability to respond in times of crisis.3 Reflected in the Court’s 2017 decision in Ziglar v. Abbasi, and echoed more recently in Hernandez v. Mesa,4 such worries about official liability have fueled an expansion of immunity defenses, as well as a growing hostility to the recognition of any right to sue under the Bivens doctrine.5 The Court and its scholarly interlocutors display a similar ambivalence about the question of who ultimately bears the burden of Bivens liability. The Court, for its part, often treats Bivens as posing a threat of substantial personal liability that counsels against recognizing new rights to sue. On other occasions, the Court has sounded notes of caution for a different reason: It has worried that the expansion of Bivens liability would impose substantial 1. 403 U.S. 388 (1971). 2. See Carlson v. Green, 446 U.S. 14, 21 (1980) (noting that the Bivens right to sue serves both a compensatory and deterrent purpose); Butz v. Economou, 438 U.S. 478, 505 (1978) (same). 3. See Ziglar v. Abbasi, 137 S. Ct. 1843, 1863 (2017) (“If Bivens liability were to be imposed, high officers who face personal liability for damages might refrain from taking urgent and lawful action in a time of crisis.”); see also Hernandez v. Mesa, 140 S. Ct. 735, 742 (2020) (describing Congress as the institution best situated to evaluate the imposition of “monetary and other liabilities” on “individual officers and employees of the Federal Government” (quoting Ziglar, 137 S. Ct. at 1856)); Schweiker v. Chilicky, 487 U.S. 412, 425 (1988) (speaking of difficulties created by the “prospect of personal liability” if a Bivens remedy is recognized); Bush v. Lucas, 462 U.S. 367, 389-90 (1983) (declining to find a Bivens remedy in part because of the impact of “added risk of personal liability”); Harlow v. Fitzgerald, 457 U.S. 800, 809, 814-15 (1982) (reshaping immunity in part to moderate the threat of personal liability). 4. 140 S. Ct. at 742 (stating that judicial recognition of a damages remedy “must rest at bottom on a statute enacted by Congress”). 5. See Ziglar, 137 S. Ct. at 1858-69 (discussing and narrowing the right to sue recognized in Bivens and applying a broad conception of qualified immunity). The Ziglar Court expressed some concern about the administrative burdens of participating in discovery and trial, but those burdens would also fall on government officials in suits seeking habeas corpus or injunctive and declaratory relief—forms of relief apparently accepted by the Court. See id. at 1862-63 (discussing the availability of habeas and injunctive relief to address detention policy claims). In the interest of full disclosure, we note that Alexander Reinert was one of the lawyers who represented the respondents in Ziglar before the Supreme Court, and he continues to represent them in district court. 563 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) indemnification costs on the government and burden the fisc.6 Scholars have been similarly nimble; they often (but do not invariably) assume that the government will indemnify its officers, thereby shifting the incidence of liability from the individual defendant to the indemnifying agency.7 While scholars debate the incentive effects of competing liability rules,8 these debates have been mostly theoretical. We know of no study that examines how the government resolves successful Bivens claims and where the burden of compensating victims of federal officials’ constitutional torts eventually falls.9 6. See id., 137 S. Ct. at 1856 (highlighting the costs to the government of defense and indemnification of official defendants). Prior to Ziglar, the Court had addressed the potential for indemnification in the Bivens context, but not necessarily as a reason to deny access to a Bivens remedy. See infra note 69 and accompanying text. 7. For scholarship on the incentive effects of constitutional tort liability that makes different assumptions about indemnification practices, compare Daniel J. Meltzer, Essay, State Sovereign Immunity: Five Authors in Search of a Theory, 75 NOTRE DAME L. REV. 1011, 1021 (2000) (arguing that indemnity is not necessarily assured), with Cornelia T.L. Pillard, Taking Fiction Seriously: The Strange Results of Public Officials’ Individual Liability Under Bivens, 88 GEO. L.J. 65, 78 & n.61 (1999) (concluding, on the basis of an interview with an official in the Department of Justice, an internal Department of Justice memo, and the content of federal regulations, that officers sued under Bivens were routinely indemnified both for the cost of legal representation and for the payment of any final award or settlement), Carlos Manuel Vázquez, Eleventh Amendment Schizophrenia, 75 NOTRE DAME L. REV. 859, 880-81, 881 n.92 (2000) (noting that officers are commonly indemnified), and Carlos Manuel Vázquez, What Is Eleventh Amendment Immunity?, 106 YALE L.J. 1683, 1795-96, 1796 n.464 (1997) (same). 8. Some argue for the assignment of liability to the agency or department. See, e.g., PETER H. SCHUCK, SUING GOVERNMENT: CITIZEN REMEDIES FOR OFFICIAL WRONGS 98-106 (1983) (emphasizing departmental ability to supervise employees so as to reduce the likelihood of a violation); Catherine Fisk & Erwin Chemerinsky, Civil Rights Without Remedies: Vicarious Liability Under Title VII, Section 1983, and Title IX, 7 WM. & MARY BILL RTS. J. 755, 758 (1999) (urging purposive interpretation of all civil rights statutes to impose vicarious liability on government agencies); Larry Kramer & Alan O. Sykes, Municipal Liability Under § 1983: A Legal and Economic Analysis, 1987 SUP. CT. REV. 249, 277-80 (articulating a preference for entity liability when the individual tortfeasor enjoys immunity based on cost-benefit analysis). Some commentators have argued in favor of an individual liability model for constitutional violations. See, e.g., Oren Gross, Chaos and Rules: Should Responses to Violent Crises Always Be Constitutional?, 112 YALE L.J. 1011, 1113 (2003) (defending Bivens’s individual liability model even if indemnification is likely); Jeffrey Standen, The Exclusionary Rule and Damages: An Economic Comparison of Private Remedies for Unconstitutional Police Conduct, 2000 BYU L. REV. 1443, 1446, 144951 (positing that damages for Fourth Amendment violations deter at least as effectively as the exclusionary rule). 9. Each of us has asked and answered adjacent questions. One of us has shown that federal officials were historically indemnified by the federal government. James E. Pfander & Jonathan L. Hunt, Public Wrongs and Private Bills: Indemnification and Government Accountability in the Early Republic, 85 N.Y.U. L. REV. 1862, 1904-05 (2010) (finding an antebellum indemnity rate of roughly 60%). Another of us has shown that state and local law enforcement officers are virtually always indemnified. Joanna C. Schwartz, Police Indemnification, 89 N.Y.U. L. REV. 885, 912-13 (2014) (reporting that state and local governments paid 99.98% of the damages plaintiffs recovered in lawsuits alleging civil footnote continued on next page 564 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) To answer these important questions, we studied successful lawsuits brought against the Federal Bureau of Prisons (BOP) and its officers. Invoking the Freedom of Information Act (FOIA), we sought data on Bivens claims brought against BOP employees that resulted in payments to plaintiffs. The BOP produced documents that revealed payments made in connection with settlements and judgments in some 209 cases that were closed over a ten-year period from 2007 through 2017—101 cases that alleged claims only under the Federal Tort Claims Act (FTCA)10 or FOIA,11 and 108 cases that included Bivens claims. Through independent research, we identified another 63 successful Bivens cases brought against BOP officials during the relevant time period.12 rights violations by law enforcement). And the third of us has shown that plaintiffs succeed in Bivens cases far more often than had previously been assumed. Alexander A. Reinert, Measuring the Success of Bivens Litigation and Its Consequences for the Individual Liability Model, 62 STAN. L. REV. 809, 839, 841 (2010) (reporting that nonfrivolous Bivens claims in which an answer or motion is filed succeed at a rate of about 30%, and that counseled Bivens claims succeed at a rate of about 39%). 10. The FTCA assigns liability to the government for many of the torts its employees commit in the course and scope of their employment. See 28 U.S.C. § 1346(b)(1) (2018) (extending federal jurisdiction to, and accepting government responsibility for, “loss of property, or personal injury or death caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his office or employment”). Originally enacted to accept federal government liability for the negligent and wrongful acts of its officers and employees, the FTCA initially excluded a specified set of intentional torts. See Federal Tort Claims Act, ch. 753, tit. IV, § 421(h), 60 Stat. 842, 846 (1946) (codified as amended at 28 U.S.C. § 2680(h)). The statute was amended in 1974 to accept government responsibility for certain intentional torts committed by law enforcement officers. See Act of March 16, 1974, Pub. L. No. 93-253, sec. 2, § 2680(h), 88 Stat. 50, 50 (codified as amended at 28 U.S.C. § 2680(h)) (identifying “assault, battery, false imprisonment, false arrest, abuse of process, or malicious prosecution,” where committed by “investigative or law enforcement officers,” as exceptions to the general rule barring government liability for intentional torts and defining law enforcement officers as those with power “to execute searches, to seize evidence, or to make arrests”). For an introduction to the Federal Tort Claims Act, see GREGORY C. SISK, LITIGATION WITH THE FEDERAL GOVERNMENT 109-96 (2016) (sketching the origins and current application of the FTCA); James E. Pfander & Neil Aggarwal, Bivens, the Judgment Bar, and the Perils of Dynamic Textualism, 8 U. ST. THOMAS L.J. 417, 424-27 (2011) (explaining the interplay between personal and government liability under the FTCA). On the circumstances surrounding the 1974 amendments, see James E. Pfander & David Baltmanis, Rethinking Bivens: Legitimacy and Constitutional Adjudication, 98 GEO. L.J. 117, 131-36 (2009) (tracing the impetus for, and practical effect of, the expansion of liability under the FTCA). On the implications for Bivens, see Carlson v. Green, 446 U.S. 14, 19-20, 19 n.5 (1980) (concluding that the expansion of the FTCA in 1974 was meant to supplement, rather than displace, the Bivens remedy). 11. See 5 U.S.C. § 552 (2018). FOIA makes no provision for money suits against the government, although it does permit individuals in some circumstances to recover their attorney’s fees. Id. § 552(a)(4)(E). 12. We cannot conclusively determine that we have identified every single successful Bivens claim brought against BOP officials during this timeframe, but we have exhaustively canvassed multiple sources. As described in greater detail in Appendix A, footnote continued on next page 565 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Our study focuses on the 171 cases with Bivens claims, where personal liability is assumed. By examining these 171 cases, we were able to determine whether individual defendants contributed any personal resources in the course of resolving the claims of misconduct. We were also able to determine the frequency with which payments were made by the BOP.13 Despite the study’s limitations, applying as it does only to the practices of a single agency over a specified period of time, we can draw important, if qualified, conclusions about who pays when Bivens litigation succeeds. Among other striking conclusions, the data reveal that individual government officials almost never contribute any personal funds to resolve claims arising from allegations that they violated the constitutional rights of incarcerated people. Indeed, of the 171 successful cases in our dataset asserting Bivens claims, we found only eight in which the individual officer or an insurer was required to make a compensating payment to the claimant.14 Of the more than $18.9 million paid to plaintiffs in these 171 cases, federal employees or their insurers were required to pay approximately $61,163—0.32% of the total.15 Echoing the conclusion one of us reached in a study of the way local governments pay settlements and judgments in § 1983 claims against state and local law enforcement officers,16 we find that the federal government effectively held its officers harmless in over 95% of the successful cases brought against them, and paid well over 99% of the compensation received by plaintiffs in these cases. A second important finding emerged from our study. Just as individual officers were almost invariably shielded from personal liability, we found that the BOP and its budget were similarly protected from financial responsibility for constitutional tort claims. The settlement agreements we reviewed made 13. 14. 15. 16. we conducted an independent review of federal court dockets during the study period using the Bloomberg Law electronic database and identified sixty-three Bivens cases filed between 2005 and 2014 which resulted in a settlement and which were not included in the BOP’s FOIA disclosures. We included those additional cases in our dataset. The BOP additionally provided information about settlements paid in disputes that were resolved prelitigation, but after canvassing those settlements, we concluded none involved Bivens claims. Although confidential settlement agreements are common in litigation between private parties, the Department of Justice has adopted a policy against entering into such agreements in civil litigation. See 28 C.F.R. § 50.23(a) (2019). The BOP made no assertion that an exception to this policy applied in response to our FOIA requests under 28 C.F.R. § 50.23(b). Nor did we locate any example of a confidential settlement agreement involving the BOP in our independent review of civil dockets. See infra Part II.A. See infra notes 79-91 and accompanying text (discussing these findings). See Schwartz, supra note 9, at 912-13 (reporting that state and local governments paid 99.98% of the damages plaintiffs recovered in lawsuits alleging civil rights violations by law enforcement). 566 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) clear that the government almost always satisfied claims brought under Bivens by arranging to have the agreed-upon amounts paid through the Judgment Fund of the United States Treasury,17 rather than by the agency responsible for the conduct of its employees—in this case, the BOP. The federal government’s practice of resolving Bivens claims through payments from the Judgment Fund has several significant implications. First, the litigation and settlement practices we report here conflict with the Supreme Court’s assumptions about the ways in which Bivens cases are resolved—these cases simply do not threaten individual employees with financial ruin or trigger indemnifying payments from their agencies. In predicating its refusal to recognize a right to sue under Bivens in part on the perceived threat of exorbitant personal, agency, or systemic liability, the Ziglar Court proceeded in error. The Hernandez Court took no steps to correct the error.18 Second, our findings have important implications for the way the political branches manage the payment of successful Bivens claims. Under longstanding Department of Justice regulations, employees sued for job-related conduct cannot seek indemnifying protection from personal liability until after the litigation concludes with the entry of an adverse judgment.19 Department of Justice attorneys often emphasize these limitations in representing to courts and to opposing counsel that federal officers face a substantial threat of personal liability in Bivens litigation.20 But our findings indicate that settlements frequently occur during the pendency of litigation and before judgment, with the amounts being paid not through agency indemnification but through the Judgment Fund.21 In some of these cases, Department of 17. See 31 U.S.C. § 1304 (2018) (appropriating money “to pay final judgments, awards, 18. 19. 20. 21. compromise settlements, and interest and costs specified in the judgments or otherwise authorized by law”). Established in 1956 to speed and simplify the payment of judgments against the United States, and extended in 1961 to authorize the payment of settlements, the Judgment Fund provides for the payment of final judgments under the FTCA, the Tucker Act, and other federal statutes that provide for the adjudication of money claims against the United States. See Paul F. Figley, The Judgment Fund: America’s Deepest Pocket & Its Susceptibility to Executive Branch Misuse, 18 U. PA. J. CONST. L. 145, 147, 158-64 (2015). Figley reports that Congress once required agencies to pay FTCA judgments from their own appropriations, but later removed that restriction and made the Judgment Fund available to pay virtually all claims under the FTCA, with no requirement that agencies reimburse the Fund. See id. at 161-67. We discuss the role of the Judgment Fund in paying for the settlement of Bivens claims in Part II.C below. See Hernandez v. Mesa, 140 S. Ct. 735, 742-43 (2020) (following Ziglar in describing the Bivens action as disfavored, in part due to the perception that Congress should decide when to impose liability on federal officials). See infra Part III.C. See infra Part III.C. See infra Parts II.B.2-.3. 567 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Justice attorneys instruct plaintiffs to substitute an FTCA claim for the Bivens claim in an amended complaint as a condition of settlement; in other cases, the settlement agreement is framed as a settlement under the Federal Tort Claims Act although there is no FTCA claim in the case. In cases in which FTCA claims were formally added and cases simply treated as though brought under the FTCA, there were often jurisdictional bars to relief; most of the FTCA claims were added well after the statute of limitations had run and without any indication that necessary administrative exhaustion procedures had been followed.22 Such practices appear to run counter to the limits imposed by Congress on the way agencies exercise their settlement authority.23 While Congress has authorized settlements under the FTCA, it has never accepted Judgment Fund liability for Bivens claims or for any agency payments made to employees to hold them harmless from personal liability. This Article proceeds in three parts. Part I describes the history and current framework of Bivens litigation as narrowed by the Court’s recent decisions in Ziglar v. Abbasi and Hernandez v. Mesa. We focus in particular on the consequences of the Court’s hostility to money claims, a hostility apparently driven in part by the perception that such claims threaten wellmeaning government officials with personal liability and their employing agencies with the burden of indemnification. Part II describes the results of our study of successful Bivens actions.24 Part II tracks representative cases, showing how the ultimate resolution of the claims by settlement or judgment sometimes corresponded with the submission of an amended pleading that restated the claims in terms of the FTCA, and sometimes resulted in an agreement for the United States simply to pay for the acts of BOP employees. Part III explores the implications of our findings in multiple areas. The data we report challenge conventional understandings of the Bivens regime, cast doubt on aspects of the Supreme Court’s Bivens jurisprudence, and invite congressional redesign of rights and remedies moving forward. More research should explore why the Department of Justice handles Bivens claims against BOP employees in the manner we have uncovered, how widely that practice applies to claims against other bureaus in the Department, and who bears the burden of liability in Bivens litigation against federal officials who work in other agencies of the government. Recognizing the value of further research and greater transparency on the part of the federal agencies, we nonetheless believe that these initial findings suggest that the Supreme Court and Congress should rethink their approach to Bivens claims. The Court cannot sensibly 22. For further description of these jurisdictional bars, see notes 120-37 below and accompanying text. See also infra Part III.D. 23. See infra Part III.D. 24. The methodology of the study is detailed in Appendix A. 568 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) predicate its hostility to Bivens cases on a concern with the threat of personal liability and indemnity. Nor should Congress, assuming it intends to encourage closer supervision of federal employees through the assignment of liability, leave federal agencies free to pass along all of the costs associated with agency misconduct to the general revenues that supply the Judgment Fund. I. Constitutional Torts, Personal Liability, and Ziglar The Supreme Court’s hostility to the personal liability model of constitutional remediation did not appear overnight. When Bivens was first announced, it was seen as consistent with contemporaneous approaches to statutory interpretation in which the Court found causes of action to be “implied” from federal law.25 Beginning in the 1980s, however, just as the Court began narrowing its implied-cause-of-action doctrine in the statutory context, it also began limiting the application of Bivens doctrine. Indeed, over the past thirty-five years, the Court has expressed hostility to Bivens actions at every opportunity. The Court’s recent opinion in Ziglar is a product and reflection of that hostility, even as it could be read to expand the justifications for questioning the original Bivens Court’s logic. This Part will briefly sketch the growth of judicial hostility towards any sort of Bivens claim. It then concludes with a close reading of Ziglar, which invoked the threat of personal and agency liability in narrowing the right to seek compensation for constitutional torts. A. Constitutional Claims and the Bivens Action In Bivens, the Supreme Court created a federal right to sue individual officers for constitutional violations.26 Webster Bivens claimed to have been the victim of an unlawful search of his home and an unlawful strip search of his person.27 He sued federal drug enforcement agents in federal court, seeking 25. Before Bivens, the Supreme Court had found implied causes of action in numerous federal statutes. See, e.g., J. I. Case Co. v. Borak, 377 U.S. 426, 430-34 (1964) (finding an implied cause of action for damages to enforce section 14(a) of the Securities Exchange Act of 1934), abrogated by Alexander v. Sandoval, 532 U.S. 275 (2001). The Bivens Court took this accepted model of statutory interpretation and applied it to the Constitution. See Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S. 388, 397 (1971) (citing Borak and noting that “we have here no explicit congressional declaration” that a remedy should not be available). 26. 403 U.S. 388 (1971). For more on Bivens, see generally James E. Pfander, The Story of Bivens v. Six Unknown Named Agents of the Federal Bureau of Narcotics, in FEDERAL COURTS STORIES 275 (Vicki C. Jackson & Judith Resnik eds., 2010). 27. Bivens, 403 U.S. at 389. 569 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) damages under the Fourth Amendment.28 The district court ultimately dismissed in part on jurisdictional grounds, finding that the claim arose under state trespass law and that Bivens should have filed suit in state court.29 The Supreme Court disagreed, concluding that the Fourth Amendment gave rise to an implied federal right of action for damages.30 The Court’s key move was not its recognition of a right to sue for damages payable by officers in their personal capacity or its opening of the federal courts to such litigation—state law provided such a right, and those common law tort claims, even if filed in state court, were often litigated in federal court after removal.31 Rather, the key was the Court’s recognition that federal officers could be sued under federal law for constitutional violations.32 For the next decade, the Supreme Court and lower courts read Bivens broadly as creating a general claim for damages caused by constitutional violations, akin to § 1983 claims against state and local government officials.33 28. Although Mr. Bivens’s pro se complaint did not specifically invoke the Fourth 29. 30. 31. 32. 33. Amendment, see Pfander, supra note 26, at 280-81, the Court located the source of his claims there, see Bivens, 403 U.S. at 389. See Pfander, supra note 26, at 282-83 (recounting the district court’s disposition); see also Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 409 F.2d 718, 721 (2d Cir. 1969), rev’d, 403 U.S. 388. See Bivens, 403 U.S. at 395-97. On the creation of a broad federal officer removal provision in 1948, see RICHARD H. FALLON JR. ET AL., HART AND WECHSLER’S THE FEDERAL COURTS AND THE FEDERAL SYSTEM 426 (7th ed. 2015). The Department of Justice’s practice in relation to individual liability claims may have taken shape in the early years of the FTCA, when the statute accepted government responsibility only for claims sounding in negligence. In such a world, the Department of Justice would understandably disclaim government liability for intentional tort claims, much the way an insurance company might disclaim matters that fell outside the scope of coverage. Individual officials were on their own as to such intentional tort claims; indeed, the claims Webster Bivens filed against federal drug enforcement agents were ultimately settled on remand when the defendants wrote Mr. Bivens personal checks. The threat of such personal, common-law-tortbased liability ended in 1988, when Congress immunized federal employees and provided for such suits to go forward, if at all, against the government under the FTCA. See Pfander & Baltmanis, supra note 10, at 133-34 (describing the adoption and operation of immunity under the 1988 Westfall Act). See Bivens, 403 U.S. at 397. The Court was careful to distinguish the functions performed and interests protected by the common law suit for tort damages from those central to a claim to vindicate constitutional rights under the Fourth Amendment. See id. at 39394 (describing it as “clear beyond peradventure that the Fourth Amendment is not tied to the niceties of local trespass laws”). On the use of § 1983 as a vehicle for the assertion of constitutional tort claims against state actors, see Monroe v. Pape, 365 U.S. 167, 169, 183-87, 192 (1961) (allowing suit to proceed against city police officers, without exhaustion of state remedies, for violation of federal constitutional rights), overruled in part by Monell v. Dep’t of Soc. Servs., 436 U.S. 658 (1978). Since Monroe, § 1983 has served as the statutory predicate for much constitutional litigation against state and local government officials. See JOHN C. footnote continued on next page 570 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) The Supreme Court extended Bivens from the Fourth Amendment to sex discrimination claims under the Fifth Amendment, and to claims brought against prison officials under the Eighth Amendment.34 During this time, lower courts often assumed that Bivens extended liability against federal officials that was as broad as the liability created by statute in § 1983.35 At the same time, the Court described Bivens as establishing that “the victims of a constitutional violation by a federal agent have a right to recover damages against the official in federal court despite the absence of any statute conferring such a right.”36 B. Suits for Damages Under the Federal Tort Claims Act While courts were reading Bivens broadly to authorize constitutional tort claims against federal officials, Congress was moving to assume responsibility for some common law torts committed by those same officials. Congress took its first such steps well before Bivens, enacting the FTCA in 1946.37 Under the FTCA, the United States agreed to waive sovereign immunity and accept vicarious liability for the torts, as defined in state tort law, that its officers committed within the scope of their employment.38 Primarily aimed at official 34. 35. 36. 37. 38. JEFFRIES, JR. ET AL., CIVIL RIGHTS ACTIONS: ENFORCING THE CONSTITUTION 16-17 (4th ed. 2018) (tracing the arc of § 1983 litigation after Monroe). The Court later limited an officer’s personal liability by installing a qualified immunity defense applicable to constitutional tort claims against state actors (and federal officials sued on a Bivens theory). See Harlow v. Fitzgerald, 457 U.S. 800, 809, 813, 818 (1982). While the federal Judgment Fund has no application to § 1983 liability, state and local governments typically use government funds to indemnify § 1983 defendants, holding them harmless from any personal liability. See Schwartz, supra note 9, at 887 n.1, 890 (analyzing personal liability in suits brought against police officers). In the end, then, government entities typically bear the financial burden associated with the imposition of constitutional tort liability under § 1983. See Carlson v. Green, 446 U.S. 14, 17-18 (1980) (recognizing a Bivens claim for an Eighth Amendment violation); Davis v. Passman, 442 U.S. 228, 234-35, 243-49 (1979) (same for Fifth Amendment Due Process Clause violations stemming from employment discrimination). For a discussion of the expansion of Bivens remedies, see Alexander A. Reinert & Lumen N. Mulligan, Asking the First Question: Reframing Bivens After Minneci, 90 WASH. U. L. REV. 1473, 1484-85 (2013). For a discussion of the pre-1980 treatment of Bivens in lower courts, see Reinert, supra note 9, at 821-22 & nn.54-57. Carlson, 446 U.S. at 18. For an introduction to the Federal Tort Claims Act, ch. 753, tit. IV, 60 Stat. 842 (1946) (codified as amended in scattered sections of 28 U.S.C), see SISK, supra note 10, at 109-96. See 28 U.S.C. § 2674 (2018) (declaring that “[t]he United States shall be liable . . . in the same manner and to the same extent as a private individual under like circumstances, but is not liable for interest prior to judgment or for punitive damages”); see also id. § 1346(b) (providing federal courts with jurisdiction over such claims, but calling for the application of the law of the state “where the act or omission occurred”). Congress footnote continued on next page 571 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) negligence, the FTCA originally included express language declaring the statute inapplicable to a long list of intentional torts.39 For these excluded intentional torts, individuals were left to whatever relief they might obtain by suing responsible officers in their personal capacity under state law or, after constitutional tort liability was recognized, by invoking Bivens. By the 1960s, successful FTCA claims were routinely payable by the federal government, through the Judgment Fund.40 But personal capacity claims were payable by the individual, subject to the possibility that the employing agency might indemnify officers held personally liable for action taken in the line of duty.41 In 1974, Congress broadened the FTCA by accepting vicarious liability for an array of common law intentional torts, liability that overlapped to some degree with the Bivens remedy the Court had announced three years earlier. Concerned about a series of no-knock federal drug enforcement raids, Congress amended the FTCA to cover specified intentional torts committed by “investigative” and “law enforcement” officials.42 Such intentional tort claims were to be predicated upon the rules of common law liability in the state where the injury occurred. Congress did not, however, accept governmental liability for Bivens claims, leaving the individual liability model in place for constitutional torts.43 The availability of these overlapping remedies created 39. 40. 41. 42. 43. has also permitted similar suits to be brought in the United States Court of Federal Claims when the subject matter is contract rather than tort. See 28 U.S.C. § 1491(a)(1) (conferring jurisdiction on the Court of Federal Claims to hear any claim against the United States founded on the Constitution, any federal statute or regulation, or any express or implied contract with the United States). For an account, see SISK, supra note 10, at 240-41. See Federal Tort Claims Act § 421(h), 60 Stat. at 846 (codified as amended at 28 U.S.C. § 2680(h)) (declaring the FTCA inapplicable to claims arising out of “assault, battery, false imprisonment, false arrest, malicious prosecution, abuse of process, libel, slander, misrepresentation, deceit, or interference with contract rights”). On the origin and operation of the Judgment Fund, see supra note 17. For the Supreme Court’s acceptance of a compensation model that then imposed some limits (later removed) on the amount of the routine Judgment Fund payments, see Glidden Co. v. Zdanok, 370 U.S. 530, 570-71 (1962) (plurality opinion) (upholding federal judicial power to adjudicate subject to the Judgment Fund compensation scheme, which at the time provided for routine payment of judgments up to only $100,000, against the argument that it did not ensure full payment with the certainty required by the finality requirements of Article III). Congress eliminated the cap on Judgment Fund payments in 1978 and now allows routine payment of judgments in any amount. See Figley, supra note 17, at 162-64. For the rules governing indemnification in the Department of Justice, see notes 187-88 below and accompanying text. See Act of March 16, 1974, Pub. L. No. 93-253, sec. 2, § 2680(h), 88 Stat. 50, 50 (codified as amended at 28 U.S.C. § 2680(h)). For an account, see SISK, supra note 10, at 170-72. See Carlson v. Green, 446 U.S. 14, 19-20, 19 n.5 (1980) (concluding that the expansion of the FTCA in 1974 was meant to supplement, rather than displace, the Bivens remedy). 572 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) the possibility that an injured individual might sue both the responsible officer under Bivens and the government under the FTCA for the same tortious misconduct.44 The existence of parallel and to some degree overlapping remedies for intentional and constitutional torts has presented puzzles of coordination.45 Indeed, the government argued in Carlson v. Green that the existence of intentional tort remedies under the FTCA displaced the officers’ personal Bivens liability for the same misconduct.46 The case arose from the actions of officers at a federal prison in Indiana, who were said to have failed to provide appropriate medical treatment for Marie Green’s imprisoned son.47 Basing her claim on the Eighth Amendment, Green sought damages from those responsible for her son’s death, a claim comparable to one she might have pursued under the FTCA against the government itself.48 The Court identified several reasons why the Bivens remedy survived the government’s acceptance of liability for parallel common law torts: Congress had written the FTCA to supplement rather than displace Bivens; it was important to preserve the liability of officials so as to deter them from engaging in unconstitutional activity; the Bivens remedy (unlike the FTCA) afforded an opportunity to secure punitive damages; and, unlike the FTCA claim, the Bivens action was triable to a jury.49 The Carlson Court thus reaffirmed personal liability as a defining feature of the Bivens action and maintained that constitutional tort claims made an important and distinctive contribution to the remedial scheme and did not merely duplicate common law tort-based liability.50 44. Courts typically allowed claimants to secure but a single compensatory award for any 45. 46. 47. 48. 49. 50. such invasion, thereby preventing double recoveries. See Pfander & Aggarwal, supra note 10, at 418-20 (examining how courts coordinate remedies and prevent double recoveries). One much debated tool of coordination, the FTCA’s judgment bar, was sometimes broadly applied to block any Bivens action that arose from the same subject matter as an FTCA claim that had been previously dismissed. See, e.g., Manning v. United States, 546 F.3d 430, 437-38 (7th Cir. 2006) (invoking the judgment bar to invalidate a jury verdict on a Bivens claim after concluding in a bench trial that the government had no liability under the FTCA for the FBI conduct in question). But see Simmons v. Himmelreich, 136 S. Ct. 1843, 1850 (2016) (holding that FTCA dismissals on the basis that claims fall within exceptions to government liability do not trigger the judgment bar). See Carlson, 446 U.S. at 19-20 (rejecting these arguments regarding the FTCA). Id. at 16 & n.1. See id. As a general matter, the FTCA incorporates state law as the measure of compensatory damages in wrongful death cases. See 28 U.S.C. § 2674 (2018). See Carlson, 446 U.S. at 20-22. See id. at 20; see also Pfander & Aggarwal, supra note 10, at 422. Several other features distinguish the FTCA remedy from that available through a Bivens action. For starters, claimants pursuing remedies under the FTCA must provide a notice of claim to the government, identifying the nature of the claim and initiating an administrative footnote continued on next page 573 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Eight years later, in the Westfall Act of 1988, Congress made one important change in the scheme of available remedies but otherwise largely ratified the system that was in place. Responding to a decision by the Supreme Court, Congress immunized federal officers from liability for all claims based on state law, thereby displacing common law tort actions brought directly against individual officers for conduct in the line of duty.51 Congress preserved the government’s liability at common law under the FTCA and the prospect of a Bivens action. Indeed, Congress created an explicit exception to the rule of personal immunity for suits alleging violations of the Constitution (or a federal statute), thereby preserving the Bivens action against individual officials.52 Today, one can sue the government under the FTCA for the (statutorily specified) intentional torts of its law enforcement officers and sue federal officers themselves under Bivens for their constitutional torts.53 C. Personal Liability and the Decline of the Bivens Action Since Carlson, the Court has turned away Bivens claims for a variety of reasons, more and less openly articulated. Many factors seem central to this trend, among them a changing conception of the role of federal courts in recognizing judge-made rights to sue; a reluctance to recognize a Bivens action in a setting where the litigant has other available remedies; and a refusal to constitutionalize and oversee special relationships (such as those between superior and subordinate members of the military).54 For the purposes of this Article, however, the Court’s growing concern with personal liability is most relevant. Worrying about the threat of overdeterrence and questioning how important money claims and retrospective relief are to a healthy system of constitutional law, a majority of the Court has come to view the Bivens action with skepticism.55 51. 52. 53. 54. 55. process that serves as a prelude to litigation. Claimants who fail to provide timely notices may be barred from recovery. For further discussion of these requirements, see notes 205-06 below and accompanying text. For the terms of Westfall Act immunity, see 28 U.S.C. § 2679(b)(2). These common law actions had been a feature of American jurisprudence since the early nineteenth century. See JAMES E. PFANDER, CONSTITUTIONAL TORTS AND THE WAR ON TERROR 3-17 (2017) (tracing early decisions that held federal officers personally accountable for onthe-job torts, subject to the prospect of congressional indemnification). See 28 U.S.C. § 2679(b). For an account, see Pfander & Baltmanis, supra note 10, at 136-38. Note, however, that recent developments suggest Bivens furnishes a right of action only for a narrow range of constitutional torts. See Ziglar v. Abbasi, 137 S. Ct. 1843, 1855-60 (2017) (applying special factors analysis to reject recognition of claims in a new context, as is more fully discussed in notes 244-45 below and accompanying text). See infra notes 56-62 (collecting cases). See Hernandez v. Mesa, 140 S. Ct. 735, 741-43 (2020); Ziglar, 137 S. Ct. at 1856-57. 574 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Familiar cases illustrate these changes in the Court’s view of the doctrine and together they convey the impression that Bivens skepticism has fed upon itself and gained momentum with each claim the Court has rejected. In a series of decisions spanning the early 1980s to the early 1990s, the Court sometimes emphasized the availability of congressionally approved alternatives that could provide remedies similar to those contemplated by Bivens.56 At other times, it appeared more concerned with the separation of powers implications of the judicial creation of a cause of action in areas uniquely committed to a coordinate branch.57 Read on its own terms, each of these decisions tends to confirm the viability of the Bivens doctrine but finds it inapplicable to the case at hand. Recent decisions take a more critical view. Lately, the Court has tended to emphasize its reluctance to fashion implied rights of action, noting that the statutory implied right of action framework on which the Bivens Court relied in part has been disbanded.58 Consider how far the Court has come since Wilkie v. Robbins, which treated the implied right of action analysis as one that lay squarely within the competence of judges making what was essentially a common law decision.59 Although Wilkie declined to recognize a Bivens claim in the specific context of retaliatory takings, it did not denigrate the enterprise of weighing factors relevant to the wisdom of allowing the suit to proceed. More recently, the Court has been openly skeptical of such analyses, viewing the recognition of implied rights to sue as a “disfavored” judicial activity.60 At the same time, the Court has described its past decisions in broader terms. Thus, the Court has restated some fact-bound decisions as reflecting a broader 56. See Schweiker v. Chilicky, 487 U.S. 412, 424-26 (1988) (emphasizing the adequacy of 57. 58. 59. 60. existing remedies for the denial of Social Security benefits in a procedural due process case); Bush v. Lucas, 462 U.S. 367, 368, 390 (1983) (treating comprehensive civil service remedies as meaningful and as sufficient to displace any need for a Bivens action in the First Amendment context). Notwithstanding the Court’s emphasis on the availability of constitutionally created alternative remedies, Carlson’s holding that the FTCA does not displace Bivens actions remains intact. See Ziglar, 137 S. Ct. at 1864-65 (distinguishing Carlson but treating it as authoritative). See Chappell v. Wallace, 462 U.S. 296, 304 (1983) (finding that a Bivens remedy was not available for racial discrimination claims brought by enlisted military personnel against superior officers in the military). Compare Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S. 388, 402-03, 402 n.4 (1971) (Harlan, J., concurring in the judgment) (invoking the implied right of action doctrine relied upon in J. I. Case Co. v. Borak, 377 U.S. 426 (1964)), with Corr. Servs. Corp. v. Malesko, 534 U.S. 61, 75 (2001) (Scalia, J., concurring) (noting that the Court had abandoned the implied right of action doctrine in Alexander v. Sandoval, 532 U.S. 275 (2001)). See 551 U.S. 537, 554 (2007) (describing the task at hand as “weighing reasons for and against the creation of a new cause of action, the way common law judges have always done”). See Ashcroft v. Iqbal, 556 U.S. 662, 675 (2009). 575 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) reluctance to extend Bivens liability “to any new context or new category of defendants.”61 One constant in the Court’s narrowing of Bivens litigation has been its concern with the threat of personal liability. That unease first surfaced in decisions that created a doctrine of qualified immunity from suit for executive officers doctrine at the same time that the Court took steps away from expansive Bivens liability. One can see this in the Court’s 1982 decision in Harlow v. Fitzgerald, which altered qualified immunity in large part because of the Court’s judgment that Bivens claims “frequently run against the innocent as well as the guilty—at a cost not only to the defendant officials, but to society as a whole.”62 The perceived threat of personal liability also came to shape a more general judicial hostility to Bivens claims. One finds the concern expressed in Bush and Schweiker, decisions that rejected Bivens claims where other remedies were deemed adequate.63 In Bush, the Court declined to recognize a right for federal employees to sue their employers for First Amendment violations, observing: “[I]t is quite probable that if management personnel face the added risk of personal liability for decisions that they believe to be a correct response to improper criticism of the agency, they would be deterred from imposing discipline in future cases.”64 In Schweiker, the Court similarly found that plaintiffs could not bring a damages action for a due process violation against federal agency administrators for fear that “[t]he prospect of personal liability for official acts . . . would undoubtedly lead to new difficulties and expense in recruiting administrators for the programs Congress has established.”65 The Court expressed similar concerns in Chappell and Wilkie.66 61. Corr. Servs. Corp., 534 U.S. at 68 (majority opinion) (describing Bush and Schweiker in this manner, despite their reliance on assessments of the adequacy of alternative remedies). 62. 457 U.S. 800, 814 (1982). Similar concerns with protecting officials from personal 63. 64. 65. 66. liability inform later qualified immunity decisions, including those that establish an objective standard for the evaluation of official immunity defenses and those that apply that standard in ever-more restrictive ways. See, e.g., Ashcroft v. al-Kidd, 563 U.S. 731, 741 (2011) (finding qualified immunity unless “existing precedent . . . placed the statutory or constitutional question beyond debate”). Schweiker v. Chilicky, 487 U.S. 412, 425 (1988); Bush v. Lucas, 462 U.S. 367, 389-90 (1983) (referring to overdeterrence from the “added risk of personal liability”). Bush, 462 U.S. at 389. Schweiker, 487 U.S. at 425. See Wilkie v. Robbins, 551 U.S. 537, 562 (2007) (expressing fear that if a Bivens action is recognized, it will lead to a “tide of suits threatening legitimate initiative on the part of the Government’s employees”); Chappell v. Wallace, 462 U.S. 296, 304 (1983) (describing how the “special nature of military life . . . would be undermined by a judicially created remedy exposing officers to personal liability at the hands of those they are charged to command”). 576 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) The Court’s recent decision in Ziglar v. Abbasi again highlighted the threat of personal liability as a reason to proceed cautiously before recognizing a claim in a new Bivens context.67 The Court in Ziglar feared that “[i]f Bivens liability were to be imposed, high officers who face personal liability for damages might refrain from taking urgent and lawful action in a time of crisis.”68 At the same time, the Ziglar Court introduced a new concern: that Bivens actions may ultimately impose financial liability on government bodies that represent and indemnify their employees.69 The Court referred vaguely to “economic and governmental concerns” raised by these costs, concluding that Congress is the proper decisionmaker for whether “monetary and other liabilities should be imposed upon individual officers and employees of the Federal Government.”70 The Ziglar Court finally combined these two issues— the threat of “personal liability” and the “projected costs and consequences to the Government itself”—to locate “systemwide” reasons to limit Bivens expansion.71 To summarize, then, the Ziglar majority invoked the threat of personal liability to make two kinds of arguments against the availability of a Bivens remedy. First, the Court returned to its oft-stated worry that individual liability might overdeter government officials and prevent them from taking appropriate action to meet an urgent crisis. Such an argument rests, at bottom, 67. Ziglar v. Abbasi, 137 S. Ct. 1843, 1858 (2017). 68. Id. at 1863. 69. The Court has addressed the potential for indemnification in the Bivens context before, but not necessarily as a reason to deny access to a Bivens remedy. In FDIC v. Meyer, for example, the Court acknowledged that the federal government might indemnify individuals in Bivens cases, but made no assumptions about how that might impact the federal fisc. 510 U.S. 471, 486 (1994). And in Cleavinger v. Saxner, the Court declined to provide absolute immunity to members of a federal prison’s disciplinary committee sued under Bivens, recognizing that if a “flood of litigation” burdened these individuals, the government could always decide to provide indemnification. 474 U.S. 193, 207-08 (1985). 70. Ziglar, 137 S. Ct. at 1856. This represents an important shift from Bush, where the Court’s decision to defer to Congress was driven not by a threat of personal liability alone but by the proposed addition of personal liability to a legislative scheme that already supplied money remedies for federal whistleblowers. See Bush v. Lucas, 462 U.S. 367, 368, 385 (1983) (describing the applicable statute as providing “meaningful remedies” against the United States for the constitutional claim in question). 71. See Ziglar, 137 S. Ct. at 1858. The Court in Hernandez v. Mesa reemphasized the concern with preserving congressional primacy in deciding when to subject government officials to personal liability. See 140 S. Ct. 735, 742 (citing Ziglar, 137 S. Ct. at 1856). More significantly, a clear majority expressed serious doubts about the legitimacy of judge-made rights to sue. See id. at 742 (emphasizing that a damages remedy must rest on a statute and that no statute authorizes a Bivens action); id. at 752 (Thomas, J., concurring) (characterizing the recognition of Bivens actions as a “usurpation of the legislative power” and calling on the Court to overturn the doctrine (quoting Gamble v. United States, 139 S. Ct. 1960, 1984 (2019) (Thomas, J., concurring))). 577 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) on the perception that the liability imposed under the Bivens doctrine ultimately falls on the officers themselves with sufficient regularity to shape their behavior. Second, the Court appeared to recognize the possibility that Bivens liability may occasion the payment of indemnity by the affected agency or the federal government as a whole. The Court pointed to both possibilities as reason to limit the right to sue until Congress acts. In adopting this either-or approach, the Court implicitly acknowledged that it does not know where the burdens of liability lie in Bivens cases.72 Despite the Court’s apparent agnosticism on this point, we share the scholarly consensus that where liability ultimately falls matters a great deal to the systemic effectiveness of constitutional remedies.73 The strength and nature of an award’s deterrent effect will depend in good measure on whether an individual officer, a federal agency, or the U.S. government as a whole will satisfy damages awards in Bivens actions. The incidence of liability also bears on the question of how seriously one should view the harms supposedly associated with expansive conceptions of Bivens liability. By learning who pays a viable Bivens claim, we can evaluate whether the Court’s hostility to money claims is warranted, and whether the liability rule properly shapes the incentives of government agencies and officials. II. Bivens in Practice: A Study of Who Pays to Resolve Well-Grounded Constitutional Tort Claims We know what the applicable statutes and regulations say about where liability falls. Claims brought under the FTCA are supposed to be paid from the U.S. Treasury’s Judgment Fund.74 Bivens claims run against the individual government official named as a defendant.75 Although that official can seek indemnification from their employer, in the absence of exceptional circumstances they can only do so after a judgment has been reached in the 72. Other considerations doubtless informed the Ziglar Court’s parsimonious view of the availability of Bivens remedies. For starters, the Court expressed suspicion of litigation that seeks to resolve broad policy issues in the context of suits to impose damages liability. See id. at 1860-61 (explaining that Bivens litigation does not provide an appropriate vehicle for testing the constitutionality of government policy decisions). See generally James E. Pfander, Dicey’s Nightmare: An Essay on the Rule of Law, 107 CALIF. L. REV. 737, 781-83 (2019) (describing the Court’s preference for injunctive-style declaratory approaches to constitutional adjudication). Moreover, the Court’s emphasis on the importance of statutory rights of action has led it to question the legitimacy of the Bivens remedy. See Ziglar, 137 S. Ct. at 1855 (recounting the Court’s growing reluctance to recognize implied rights of action). 73. See supra notes 7-8. 74. See supra notes 17, 40 and accompanying text. 75. See supra notes 33-36, 41 and accompanying text. 578 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) case.76 And Department of Justice officials have made very clear in filings and representations to courts at all levels that indemnification is far from certain.77 Based on the applicable statutes and rules, one would assume that the Treasury’s Judgment Fund is used to satisfy successful FTCA claims, federal agencies satisfy successful Bivens claims if they agree to indemnify their employees, and individual officers pay when successful Bivens claims are settled before trial and when agencies refuse to indemnify. Yet no data exist to test these three assumptions. In short, we know little about who actually pays when Bivens claims succeed. In order to answer these pressing questions about the realities of Bivens litigation, we submitted a FOIA request to the BOP, seeking the litigation records from successful claims, including information about the amount paid in each successful case and about which entity or person made the payments in question.78 The BOP responded with the litigation records of cases closed over a ten-year period from 2007 to 2017. Through independent research, we uncovered information about sixty-three additional cases not initially produced by the BOP.79 We then submitted to the BOP requests for additional information about these cases. From our research, and the BOP’s responses, we have constructed a dataset of 171 cases in which the plaintiffs presented a Bivens claim at some point during the course of litigation and secured a settlement or payment after judgment. In only 8 of the 171 cases (less than 5%), BOP employees and their insurers paid a share of the settlement amount.80 That share amounted to just 0.32% of the more than $18.9 million paid to plaintiffs to resolve these 171 claims. Furthermore, we found no case in which the BOP itself appears to have contributed agency funds to plaintiffs’ settlements in successful Bivens claims. Instead, government attorneys arranged to have these matters resolved with payments from the Judgment Fund, which is funded by the Treasury of the United States. In this Part, we discuss what our data reveal about the manner in which these cases were resolved. We begin with a sketch of the handful of claims in which the employees contributed personal or insurance funds to the settlement pot. We next examine the vast majority of cases (more than 95%) See infra notes 187-90 and accompanying text. See infra Part III.C. Our methodology is set out in more detail in Appendix A. As described in Appendix A, we utilized the Bloomberg Law search engine to canvass federal court dockets using search terms likely to uncover settlements involving the BOP and its employees. See also supra note 12 and accompanying text. 80. For a description of these eight cases and the amount paid, see Part II.A below. See also infra Appendix B, Table 1. 76. 77. 78. 79. 579 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) in which the officials named as defendants made no contribution to the payment of settlements and judgments. We focus particularly on the various ways in which claims evolved over the course of their litigation lifespan from suits seeking to impose personal liability on federal employees to suits that were ultimately resolved with compensating payments from the Treasury. A. Cases in Which Employees Contributed to the Payment of Settlements and Judgments Individual officers were required to contribute to the resolution of 8 of 171 Bivens cases in our dataset.81 Three of the eight cases involved claims alleging sexual assault by BOP employees.82 In another three of the eight cases, plaintiffs sued BOP employees for improper medical care.83 In one case, the plaintiff accused the defendant officer of putting thorns on a bench 81. See infra Appendix B, Table 1. 82. The first of the three cases is Doe v. United States, settling for $70,500 with the officer— who was in prison—paying $3000. Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, at 2, Doe v. United States, No. 1:08-cv-00517 (D. Haw. Dec. 30, 2013); Order Granting Defendant Markell Milsap’s Motion for Approval of Good Faith Settlement at 1-2, Doe, No. 1:08-cv-00517 (D. Haw. Dec. 18, 2009), ECF No. 46. The second case is Doe v. United States, settling for $40,000, with $25,000 paid by the officer. Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, at 2, Doe v. United States, No. 1:12-cv-00640 (D. Haw. Apr. 28, 2015); Complaint for Damages at 4, Doe, No. 1:12cv-00640 (D. Haw. Nov. 29, 2012), ECF No. 1. The third case is Harrison v. Jackson, settling for $11,000 and paid by the officer. Joint Motion for Entry of Consent Judgment Against Defendant Lewis Jackson Exhibit 1, at 3, Harrison v. Jackson, No. 1:12cv-04459 (N.D. Ga. Mar. 4, 2016), ECF No. 121; Complaint at 6-7, Harrison, No. 1:12-cv04459 (N.D. Ga. Dec. 27, 2012), ECF No. 1. 83. The first of the three cases is Ortiz v. Bezy, settling for $10,000 paid either by the defendant doctor or his insurer. Motion for Time to File Stipulation of Dismissal at 2, Ortiz v. Bezy, No. 2:05-cv-00246 (S.D. Ind. June 27, 2013), ECF No. 219; Civil Rights Complaint at 2, Bezy, No. 2:05-cv-00246 (S.D. Ind. Oct. 5, 2005), ECF No. 1. The second case is Bolden v. Marberry with a confidential settlement paid by the defendant officer. Complaint for Injunctive & Declaratory Relief & Damages from Violations of Constitutional Rights at 2, Bolden v. Marberry, No. 2:09-cv-00312 (S.D. Ind. Sept. 25, 2009), ECF No. 1; Email from C. Darnell Stroble, Assistant Gen. Counsel/FOIA Pub. Liaison, Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Mar. 28, 2019, 5:27 AM) (on file with authors) (explaining that the settlement amount in Marberry was confidential). The third case is Jones v. Caraway, settling for $662.95 and paid by the officer. Settlement Agreement & Release at 1-2, Jones v. Caraway, No. 2:14-cv-00319 (S.D. Ind. Jan. 18, 2017); Complaint at 3, Caraway, No. 2:14cv-00319 (S.D. Ind. Oct. 24, 2014), ECF No. 1. 580 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) where he sat, injuring him.84 And in one case, the plaintiff sued three defendant officers for physical assault.85 We have information about the amounts officers were required to pay in seven of the eight cases. In those seven cases, eight officers were obligated to pay a total of $61,163. The average required contribution for those eight officers was $7645, and their median required contribution was $5000. It bears noting that three of the eight officers paid only a portion of the settlement awards, with the government contributing the remainder from the Judgment Fund.86 Another three of the eight BOP employees were financially shielded in other ways: One was a doctor whose settlement was likely paid by an insurer, effectively holding the individual employee harmless from any personal liability; the other two BOP employees never paid the $5000 each agreed to pay through a Rule 68 judgment.87 Just two BOP employees paid the entirety of the settlement awards resulting from their misconduct: One settlement for $11,000 was paid by a doctor accused by the plaintiff of repeated sexual assaults;88 the other settlement for $663 was paid by an officer alleged to have assigned the plaintiff to a top bunk despite knowing he suffered from seizures.89 One can understand why the conduct alleged in these eight cases, if proven, might give rise to successful constitutional tort claims. But when we 84. Amended Complaint at 2, Shirley v. Manning, No. 3:13-cv-00236 (D. Or. Nov. 4, 2013), 85. 86. 87. 88. 89. ECF No. 17 (describing plaintiff’s allegations); U.S. Fed. Bureau of Prisons, Case Details: Shirley v. Manning et al (2016) (on file with authors) (showing a settlement for $7500, with the officer paying $1500). Complaint at 4, Monclova-Chavez v. McEachern, No. 1:08-cv-00076 (E.D. Cal. Jan. 15, 2008), ECF No. 1; Email from Elizabeth Alexander, Attorney, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Dec. 23, 2018, 4:25 PM) (on file with authors) (noting the acceptance of a Rule 68 award by two of the officers in Monclova-Chavez for $10,000 but also noting that the officers never satisfied the judgment, meaning the plaintiff ultimately received nothing). See infra Appendix B, Table 1. See Motion for Time to File Stipulation of Dismissal at 2, Ortiz, No. 2:05-cv-00246 (S.D. Ind. June 27, 2013), ECF No. 219; Email from Elizabeth Alexander to Joanna C. Schwartz, supra note 85 (describing the outcome in Monclova-Chavez); see also infra Appendix B, Table 1. The records do not reflect whether other officers had professional liability insurance that they used to satisfy their settlement obligations. We are aware that insurance companies market liability insurance to federal employees. See, e.g., STARR WRIGHT USA, https://perma.cc/GA4M-5C7T (archived Jan. 22, 2020). Further research should explore the frequency with which federal officers purchase such insurance, and the frequency with which insurers satisfy federal officers’ legal liabilities. See Consent Judgment Against Defendant at 3, Harrison v. Jackson, No. 1:12-cv-04459 (N.D. Ga. Mar. 8, 2016), ECF No. 122; Complaint at 6, Harrison, No. 1:12-cv-04459 (N.D. Ga. Dec. 27, 2012), ECF No. 1. See Settlement Agreement & Release at 1-2, Jones v. Caraway, No. 2:14-cv-00319 (S.D. Ind. Jan. 18, 2017) (on file with authors); Complaint at 3, Caraway, No. 2:14-cv-00319 (S.D. Ind. Oct. 24, 2014), ECF No. 1. 581 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) compare these eight cases to the great mass of cases in which the individuals made no contribution to the settlements, it is unclear why BOP employees were required to contribute to the resolution of these cases and not others. It may be that BOP employees are more likely to be required to contribute to certain types of cases: Three of the eight cases involved sexual assault.90 It may also be that there is some regional variation in U.S. Attorneys’ or BOP attorneys’ demands that employees contribute: Three of the eight cases in which officers were required to contribute were filed in the Southern District of Indiana, and two were filed in the District of Hawaii. But there are other cases of sexual assault, and other cases brought in the Southern District of Indiana and the District of Hawaii, in which the involved officers were shielded from financial liability.91 And, as the examples in the next Subpart suggest, one cannot confidently say that the official misconduct in these eight 90. This hypothesis is consistent with plaintiffs’ attorneys’ anecdotal evidence—attorneys with whom we spoke report that individual officers are very rarely held personally liable in these cases, and the only exceptions to this rule were in cases of sexual assault by officers. As one attorney told us, In my experience—many cases over more than 40 years—I have never heard of an individual federal employee having to contribute to tort settlements or judgments paid by the government, and that includes medical malpractice cases, rapes and sexual assaults in prisons, wrongful deaths, etc. I have participated in two cases in which male prison guards were named individually in cases in which they and a federal correctional institution were both sued for an alleged sexual relationship involving a female inmate, and the individuals settled separately from the government, but those are the only occasions of which I am aware that an employee/actor contributed monetarily to a settlement or judgment in a federal tort claim matter. Email from Eric Seitz, Attorney, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Oct. 25, 2017, 1:57 PM). Note that Seitz was an attorney for the plaintiffs in both Doe v. United States, No. 12-cv-0640 (D. Haw.), and Doe v. United States, No. 1:08-cv00517 (D. Haw.)—two of the three sexual assault cases in which individual officers contributed to the settlements. 91. Officers did not contribute to settlements in two other Bivens actions in the dataset involving sexual assault and rape by the officers—Zepeda v. United States and Houston v. United States. For information about Zepeda, see Stipulation for Compromise Settlement & Release at 2, Zepeda v. United States, No. 1:06-cv-00676 (D. Haw. Apr. 10, 2008); and Complaint at 3, Zepeda, No. 1:06-cv-00676 (D. Haw. Dec. 27, 2006), ECF No. 1. For information about Houston, see Stipulation for Compromise Settlement & Proposed Order at 2, Houston v. United States, No. 2:08-cv-01076 (C.D. Cal. Feb. 11, 2009); and Complaint at 3, Houston, No. 2:08-cv-01076 (C.D. Cal. Feb. 15, 2008), ECF No. 1. Moreover, officers were not required to contribute in all cases that arose in the Southern District of Indiana and the District of Hawaii. Zepeda was brought in the District of Hawaii, with no contribution by the officer required to resolve the Bivens claim. Three other Bivens cases were brought in the Southern District of Indiana in which individual officers were not required to contribute, but the Bivens claims in each of those cases were dismissed before settlements were entered. See Stipulation of Dismissal, Johnson v. Merritt, No. 2:13-cv-00441 (S.D. Ind. July 22, 2016), ECF No. 107; Final Judgment, Barker v. McPherson, No. 2:10-cv-00314 (S.D. Ind. May 4, 2015), ECF No. 106; Penick v. United States, No. 2:12-cv-00341, 2014 WL 5431594 (S.D. Ind. Oct. 24, 2014). 582 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) cases was more egregious or constitutionally problematic than that in the cases in which individuals made no compensating payments. B. Cases in Which Employees Made No Compensating Payments In 163 of the 171 cases in our dataset, BOP employees did not contribute any amount to the resolution of successful Bivens actions. But as noted above, these no-contribution cases do not necessarily differ in terms of the severity or nature of the claims being asserted from those in which employees made payments. Thus, in the no-contribution cases, we find allegations that prison guards used excessive force, committed sexual assault, and turned a blind eye in allowing people confined in prison to assault each other, and that officials failed to provide adequate medical care to incarcerated people. The records we reviewed do not explain why the settlement of these claims was entirely underwritten by the government, whereas a handful of apparently similar claims were settled at the employee’s (partial) expense. The data do reveal, though, the manner in which these no-contribution cases evolved from suits against officers to settlements and judgments paid by the government. Aside from one outlier,92 these cases took one of three paths from filing to resolution that shielded individuals from liability. In 59 (36.2%) of the 163 no-contribution cases, courts dismissed Bivens claims during the course of litigation, or plaintiffs voluntarily dismissed Bivens claims.93 In 63 (38.7%) of these cases, plaintiffs settled their cases—which alleged both Bivens and FTCA claims—in return for payments made by the U.S. government.94 And in 40 (24.5%) of these cases, the government appears to have restyled Bivens claims in various ways as FTCA claims at or around the time of settlement to facilitate payments through the Judgment Fund.95 These restylings in many instances required the government and the district court to overlook jurisdictional bars to the filing of the FTCA claims that replaced the Bivens 92. In one case, the plaintiff won at trial on his Bivens claim, the individual defendants sought and received indemnification for the judgments, and the plaintiff was paid by the U.S. government based on that indemnification agreement. Judgment in a Civil Action, Pronin v. Duffey, No. 6:13-cv-03423 (D.S.C. Aug. 17, 2016), ECF No. 211; Jury Instructions, Pronin, No. 6:13-cv-03423 (D.S.C. Aug. 16, 2016), ECF No. 210; Letter from Benjamin C. Mizer, Principal Deputy Assistant Attorney General, U.S. Dep’t of Justice, to Michael Frazier, Associate Gen. Counsel, Fed. Bureau of Prisons (Nov. 10, 2016) (on file with authors). We discuss the Department of Justice regulations governing employee indemnification in Part III. 93. See infra Part II.B.1; Appendix B, Table 2. 94. See infra Part II.B.2; Appendix B, Table 3. 95. See infra Part II.B.3; Appendix B, Table 4. 583 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) claims.96 We describe these broad practices, along with certain differentiating refinements, in the Subparts that follow. 1. Dismissal of Bivens claims during litigation In fifty-nine cases, plaintiffs pursued both Bivens and FTCA claims, but the Bivens claims were dismissed at some point during the course of litigation.97 In thirty-seven of these cases, individual defendants were dismissed by the court sua sponte, at the motion to dismiss or summary judgment stages, or on appeal, and the FTCA claims proceeded and settled or went to judgment against the United States.98 In two cases, both Bivens and FTCA claims were tried and judgment was entered against the United States—but not against the individual defendants—after trial.99 In twenty cases, the plaintiffs voluntarily moved to dismiss the individual defendants during the course of litigation. These dismissals occurred months or years before the resolution of the FTCA claims, and the available records do not indicate whether the dismissals of the Bivens claims were related in any way to settlement negotiations.100 But, given the government’s practice of 96. See infra notes 120-33 and accompanying text. 97. Each of these cases is set out in Appendix B, Table 2. 98. Many of these cases involved both Bivens and FTCA claims at the time of filing, but the Bivens claim was dismissed at some point in the litigation. See, e.g., M.G. v. United States, 603 F. App’x 616 (9th Cir. 2015) (reversing on interlocutory appeal the order denying defendants’ motion to dismiss the Bivens claim); Duran v. Lindsay, No. 1:09-cv-05238, 2015 WL 4994315, at *4, *6 (E.D.N.Y. Aug. 12, 2015) (dismissing the Bivens claim at summary judgment); Order, Zidell v. Kanan, No. 4:10-cv-00106 (N.D. Tex. July 28, 2010), ECF No. 12 (dismissing the Bivens claim sua sponte); Manswell v. United States, No. 1:09-cv-04102, 2010 WL 3219156, at *1 (S.D.N.Y. Aug. 12, 2010) (granting motion to dismiss the Bivens claim); Almashleh v. United States, No. 1:06-cv-00106, 2007 WL 2406965, at *1 (W.D. Pa. Aug. 21, 2007) (dismissing the Bivens claim at summary judgment). 99. For relevant information about the two cases—Barker v. McPherson and Northington v. Hawk-Sawyer, see Final Judgment, Barker v. McPherson, No. 2:10-cv-00314 (S.D. Ind. May 4, 2015), ECF No. 106; Barker, No. 2:10-cv-00314, 2015 WL 2093459, at *5-6 (S.D. Ind. May 4, 2015); Order Granting Joint Motion to Vacate, Northington v. HawkSawyer, No. 2:04-cv-01032 (C.D. Cal. Oct. 7, 2014), ECF No. 419; and Amended Judgment, Northington, No. 2:04-cv-01032 (C.D. Cal. Mar. 20, 2014), ECF No. 410, vacated, Order Granting Joint Motion to Vacate, Northington, No. 2:04-cv-01032 (C.D. Cal. Oct. 7, 2014). 100. See, e.g., Order at 2, Hildebrand v. United States, No. 1:13-cv-01233 (C.D. Ill. Jan. 12, 2015), ECF No. 61 (dismissing voluntarily individual defendants before the case settled); Final Judgment as to Certain Defendant, Lee v. United States, No. 4:12-cv-00197 (N.D. Tex. June 18, 2012), ECF No. 17; Stipulated Order That Defendants Fausto & Dalmasi Are Dropped as Defendants Pursuant to Fed. R. Civ. P. 21, Williams v. United States, No. 2:11-cv-05612 (E.D. Pa. Nov. 29, 2011), ECF No. 9; Stipulation & Order of Partial Dismissal Without Prejudice, Banks v. United States, No. 1:10-cv-05308 (E.D.N.Y. Mar. 24, 2011), ECF No. 7; Stipulation for Partial Dismissal at 1-2, Vincent v. Fed. Bureau of Prisons, No. 2:08-cv-03286 (C.D. Cal. Sept. 16, 2008), ECF No. 10 (dismissing footnote continued on next page 584 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) formally and informally substituting claims under the FTCA for Bivens claims, and the government’s apparent view that it is preferable to settle cases under the FTCA, it is certainly possible that at least some of the voluntary dismissals in these twenty cases were made with an eye toward eventual settlement of the cases through the FTCA.101 2. Settlement of Bivens and FTCA claims with payment by the U.S. government In sixty-three cases, both Bivens and FTCA claims remained up until the time of settlement, but the settlements were paid by the United States government.102 In forty-nine of these cases, both the individual defendants and the United States are described as parties to the settlement agreements, but the language of the agreements or other available evidence makes clear that the United States was the only party to pay to resolve the claims.103 In the remaining fourteen cases, both Bivens and FTCA claims remained at settlement, but the parties agreed that the plaintiff would dismiss the Bivens claims with prejudice and the settlement was executed solely with the United States of America. In four of these cases, motions to dismiss the individual defendants were filed with the court and then settlement agreements between the plaintiffs and the United States were filed the same day or a few days later.104 In the other ten cases, the settlement agreements explicitly state that 101. 102. 103. 104. voluntarily individual defendants without prejudice before the case settled); Dismissal Without Prejudice, Clark v. United States, No. 3:06-cv-00016 (S.D. Ill. Mar. 14, 2008), ECF No. 82 (dismissing voluntarily individual defendants without prejudice before the case settled). Note that in Hildebrand one of the defendants was dismissed for lack of personal jurisdiction. Order & Opinion at 1-2, Hildebrand, No. 1:13-cv-01233 (C.D. Ill. Oct. 11, 2013), ECF No. 37. For further description of these practices and preferences, see notes 106-33 below and accompanying text. Each of these cases is set out in Appendix B, Table 3. We have settlement agreements in the majority of these cases that make clear the United States paid the settlements. In some cases, a representative from the Bureau of Prisons confirmed via email that payments were made by the U.S. government. All settlement agreements and email confirmations are on file with the authors. For relevant documents regarding these four cases—Lee v. Pfister, Hill v. United States, Fitz v. Malatinsky, and Morris v. Jones—see Stipulation to Dismiss Lt. Pfister with Prejudice, Lee v. Pfister, No. 5:12-cv-00794 (C.D. Cal. June 13, 2017), ECF No. 194 (dismissing voluntarily individual defendant the same day the parties filed a settlement agreement regarding the FTCA claim); Settlement Agreement at 7, Hill v. United States, No. 1:13-cv-03404 (D. Colo. May 29, 2015) (providing that the plaintiff agrees to “dismiss the Bivens Defendants with prejudice and will file an amended complaint that contains an injunctive relief claim against the BOP and a negligence claim for failure to protect against the United States under the Federal Tort Claims Act” and that “Plaintiff will sign a standard FTCA stipulation and a stipulation of dismissal” once the complaint has been amended); Stipulated Order of Dismissal, Fitts v. Malatinsky, footnote continued on next page 585 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) payment to resolve the FTCA claim was to be made on the condition of dismissing the individual Bivens claims, but there was no separate documentation of the dismissal of the individual defendants.105 3. Substitution of FTCA claims for Bivens claims as a condition of settlement In forty cases, plaintiffs had only Bivens claims against individual officers immediately preceding the time of settlement, but the U.S. government paid the settlements nevertheless.106 In twenty-one of these cases, the parties formally substituted an FTCA claim for the Bivens claim or added an FTCA claim to the complaint at or around the time of settlement.107 In the other nineteen cases, there was no formal substitution of an FTCA claim for a Bivens claim, but the settlements were paid by the United States as though the cases were brought under the FTCA.108 These formal and informal substitutions of 105. 106. 107. 108. No. 2:10-cv-11100 (E.D. Mich. July 23, 2012) (dismissing voluntarily the only individual defendant one day before the voluntary dismissal of the case), ECF No. 42; and Stipulation of Dismissal of Certain Defendants, Morris v. Jones, No. 2:08-cv-03842 (E.D. Pa. Aug. 25, 2011) (dismissing voluntarily individual defendants four days before the voluntary dismissal of the case), ECF No. 101. For relevant documents related to four of these cases—Chicarielli v. United States, Luna v. Jordan, Cottini v. United States, and Gil v. Reed—see Stipulation & Order of Settlement & Dismissal at 2-3, Chicarielli v. United States, No. 1:14-cv-06765 (S.D.N.Y. May 17, 2016), ECF No. 51 (“WHEREAS, Plaintiff has decided to voluntarily dismiss with prejudice all claims against the Agency Defendants and the Individual Defendants, and all claims against the United States other than the FTCA Claim . . . [t]he United States agrees to pay to Plaintiff the sum of thirty-six thousand five hundred dollars ($36,500.00) (the ‘Settlement Amount’) in connection with the FTCA Claim.”); Stipulation for Compromise Settlement & Release of Tort Claims at 2-5, Luna v. Jordan, No. 1:14-cv02028 (M.D. Pa. Jan. 27, 2015) (providing that the United States agrees to pay $5350 and that “[t]he parties further understand and agree that all individual Defendants who have been named in this case are dismissed from the case with prejudice, and that no claims against any individual Defendants arising from the acts and allegations complained of in the complaint survive this settlement agreement and release”) (on file with authors); Stipulation for Compromise Settlement at 2, Cottini v. United States, No. 2:10-cv-00294 (C.D. Cal. Mar. 13, 2012), ECF No. 118 (“In exchange for a payment of $200,000, paid on behalf of the United States only, Plaintiffs agree to dismiss with prejudice the Bivens claims against the individual defendants . . . .”); and Stipulation for Compromise Settlement and Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677 at 2-3, Gil v. Reed, No. 3:00-cv-00724 (W.D. Wis. Feb. 23, 2009) (“The United States of America agrees to pay the sum of $20,000 which sum shall be in full settlement and satisfaction of any and all claims . . . . Plaintiffs agree that all claims against the individual defendants, James Reed and Jaime Penaflor, will be dismissed with prejudice as part of this settlement.”) (on file with authors). Each of these cases is set out in Appendix B, Table 4. See infra notes 109-19 and accompanying text. See infra notes 135-37 and accompanying text. 586 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) the United States for individual defendants occurred in a variety of ways, outlined in the Subparts that follow. a. Formal substitution In twenty-one cases, an FTCA claim was formally substituted for the Bivens claim or added to the complaint as a condition of settlement. In seven of these twenty-one cases, the plaintiffs amended their complaints to dismiss the Bivens defendants and add FTCA claims against the United States at or around the time of settlement.109 In two of the twenty-one cases, the plaintiffs moved to amend their complaint to add an FTCA claim without dismissing the Bivens claims, and the United States agreed to pay the settlement.110 In three of the 109. For documents relevant to Shepherd v. Palmer, see Letter from Surinder K. Aggarwal to Noel L. Hillman, Judge, U.S. Dist. Court, Dist. of N.J., Shepherd v. Palmer, No. 1:14-cv02992 (D.N.J. Aug. 17, 2016), ECF No. 51 (providing notice of settlement); and Consent Order Nunc Pro Tunc to Grant Plaintiff Leave to File a Second Amended Complaint, Shepherd, No. 1:14-cv-02992 (D.N.J. July 28, 2016), ECF No. 49. For documents relevant to Montoya v. Wall, see Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, Montoya v. Wall, No. 1:11-cv01414 (C.D. Ill. Oct. 30, 2014), ECF No. 51-1; and Unopposed Motion for Leave to File an Amended Complaint at 2, Montoya, No. 1:11-cv-01414 (C.D. Ill. Oct. 22, 2014), ECF No. 49. For documents relevant to Stine v. Allred, see Third Amended Verified Prisoner Complaint, Stine v. Allred, No. 1:11-cv-00109 (D. Colo. Apr. 12, 2013), ECF No. 358; and Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2672, Stine, No. 1:11-cv-00109 (D. Colo. Apr. 12, 2013) (on file with authors). For documents relevant to Freeman v. Woolston, see Plaintiff ’s Motion to Enforce Settlement Agreement at 5, Freeman v. Woolston, No. 1:11-cv-01756 (D. Colo. Jan. 13, 2014), ECF No. 277. For documents relevant to Ellis v. United States, see Stipulation of Dismissal with Prejudice, Ellis v. United States, No. 1:08-cv-00160 (W.D. Pa. Nov. 14, 2012), ECF No. 98; and Amended Complaint, Ellis, No. 1:08-cv-00160 (W.D. Pa. July 25, 2012), ECF No. 96. For documents relevant to Johnson v. Martinez, see Second Amended Complaint at 5, Johnson v. Martinez, No. 2:04-cv-01967 (E.D. Pa. Mar. 8, 2007), ECF No. 52; and U.S. Fed. Bureau of Prisons, Case Details: Johnson v. Martinez et al (2016) (on file with authors) (showing payment of settlement in Johnson, No. 2:04-cv01967 (E.D. Pa.)).While the plaintiff in Johnson framed his claims under § 1983, they were properly understood by the court to be Bivens claims. See Johnson, No. 2:04-cv01967, 2006 WL 208640, at *2 n.9 (E.D. Pa. Jan 19, 2006). For documents relevant to Shannon v. Federal Bureau of Prisons, see Settlement Agreement at 2, Shannon v. Fed. Bureau of Prisons, No. 1:03-cv-00352 (D. Colo. Dec. 1, 2006), ECF No. 134; and Unopposed Motion for Leave to File Second Amended Complaint at 1, Shannon, No. 1:03-cv-00352 (D. Colo. Oct. 25, 2006), ECF No. 127. 110. For information about Al-Kidd v. Sugrue, see Settlement Agreement & Release at 1-2, Al-Kidd v. Sugrue, No. 5:06-cv-01133 (W.D. Okla. Mar. 24, 2008) (on file with authors); and Plaintiff ’s Consent Motion to File a Fourth Amended Complaint at 2, Al-Kidd, No. 5:06-cv-01133 (W.D. Okla. Jan. 23, 2008), ECF No. 56 (seeking leave to add the FTCA claim without removing the Bivens claim). For information about Buckley v. Harding, see Amended Complaint at 5, Buckley v. Harding, No. 1:06-cv-00413 (D. Colo. Aug. 20, 2007), ECF No. 60; and Stipulation for Compromise Settlement ¶ 4, Buckley, No. 1:06-cv-00413 (D. Colo. Aug. 8, 2007) (on file with authors). 587 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) twenty-one cases, the settlement agreements provided that the United States would pay plaintiffs to resolve an FTCA administrative claim on the condition that the pending Bivens claims were voluntarily dismissed.111 In another three cases, defendants filed a motion to substitute the United States as a party defendant for the named officers, but plaintiffs never filed new amended complaints naming the United States as the sole defendant.112 For example, in Laurent v. Castellanos, the pro se plaintiff alleged that on January 27, 2012, defendant officers put him, handcuffed, in a “reck [sic] pen” area with another prisoner and then watched without intervening as the plaintiff was beaten.113 Laurent filed a civil suit against the individual officers on May 27, 2014.114 After almost two years litigating the case, the parties came to a settlement in principle, conditioned on the substitution of the individual defendants for the United States. As the Assistant U.S. Attorney explained to the judge in the case: The parties have an agreement in principal [sic] to settle this matter, subject to formal approval of this office and the Bureau of Prisons. As part of that agreement in principal [sic], the undersigned will file a certification of scope and notice of substitution, substituting the United States for the individual defendants in this action, with respect to the negligence and intentional tort claims in 111. See Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2672, at 1-2, Caballero v. Mejia, No. 4:13-cv-00630 (N.D. Fla. Sept. 4, 2015), ECF No. 108-1 (providing in the settlement agreement that the United States will pay to resolve plaintiff ’s administratively filed FTCA claim upon dismissal of the Bivens action); Order on Motion from Relief from Agreement, Order Closing Case Pursuant to Stipulation, & Order Denying All Other Pending Motions at 2, Gillings v. Lepe, No. 1:12-cv-01533 (E.D. Cal. Feb. 26, 2015), ECF No. 72 (discussing the government’s agreement to settle a previously denied FTCA claim in exchange for the plaintiff dismissing the Bivens claims); Release & Settlement Agreement at 2, Nunez v. Lindsay, No. 3:05-cv-01763 (M.D. Pa. June 9, 2007) (on file with authors) (providing in the settlement agreement that the plaintiff will file an administrative FTCA claim and that the government will pay to settle that claim in exchange for the plaintiff dismissing the Bivens suit). 112. For information about the three cases—Laurent v. Castellanos, De Anda v. Smith, and Williams v. Warmerdorf—see Certification of Scope of Employment & Notice of Substitution of United States as Party Defendant for Officer A. Castellanos, Officer Sheperd, Officer P. Naupari, & D. Garcia at 1-2, Laurent v. Castellanos, No. 1:14-cv-03340 (E.D.N.Y. Mar. 7, 2016), ECF No. 79; Joint Motion to Dismiss Individual Defendants & Substitute the United States of America at 1, De Anda v. Smith, No. 1:10-cv-01094 (C.D. Ill. Sept. 3, 2013), ECF No. 77; and Notice of Substitution at 1, Williams v. Warmerdorf, No. 3:07-cv-01283 (M.D. Pa. Aug. 3, 2011), ECF No. 128. 113. Civil Rights Complaint at 15, Laurent, No. 1:14-cv-03340 (E.D.N.Y. May 27, 2014), ECF No. 1. 114. The complaint does not clearly describe the plaintiff ’s cause of action, see id., but was treated by the judge in the case and the defendants as a Bivens claim, see Status Report at 1, Laurent, No. 1:14-cv-03340 (E.D.N.Y. Feb. 8, 2016), ECF No. 77. 588 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) plaintiff’s Amended Complaint. Plaintiff has agreed to dismiss the claims against the individual defendants and settle this matter with the United States.115 The settlement agreement was signed by the parties, and provided that the United States would pay $5500 to the plaintiff to settle the matter.116 In six cases, there was no formal amendment of the complaint, but the settlement agreement provided that the United States was substituting itself for the individual Bivens defendants.117 For example, in McCarroll v. Matteau, the plaintiff brought a Bivens action alleging that an officer retaliated against him for engaging in legal work.118 The plaintiff never alleged an FTCA claim and the defendant never asserted that the case was properly brought under the FTCA, but the settlement agreement provided that “[t]he parties agree that plaintiff’s complaint is to be construed as a complaint for damages pursuant to the Federal Tort Claims Act . . . and that any claims filed pursuant to Bivens . . . are hereby withdrawn.”119 In at least thirteen of these twenty-one cases, government agents and courts agreed to payments on FTCA claims with no indication that the FTCA claims were administratively exhausted or filed within the appropriate statute 115. Status Report, supra note 114, at 1-2. 116. Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, at 3, Laurent, No. 1:14-cv-03340 (E.D.N.Y. May 3, 2016) (on file with authors). 117. See Stipulation & Order to Modify Caption & Dismiss Case with Prejudice at 2-3, Bolden v. Beaudouin, No. 1:14-cv-05470 (E.D.N.Y. Apr. 12, 2017), ECF No. 69 (providing in the settlement agreement that the United States “has, by operation of law, been substituted as the sole defendant with respect to any of plaintiff’s claims . . . which sound in state law tort”); Stipulation for Compromise Settlement & Release at 1, Brizard v. Terrell, No. 1:11-cv-02274 (E.D.N.Y. Dec. 3, 2013) (on file with authors) (providing in the settlement agreement that the Attorney General “certified, by the authority vested in him . . . that Defendant was acting within the course and scope of his employment as an employee of the United States of America at all times relevant to Plaintiff ’s Complaint, and thus, the United States was hereby substituted by operation of law as a party defendant for Defendant for purposes of any claims or liability under the Federal Tort Claims Act”); Stipulation for Compromise Settlement Pursuant to 28 U.S.C. § 2677, at 1-2, McCarroll v. Matteau, No. 9:09-cv-00355 (N.D.N.Y. Aug. 26, 2013), ECF No. 62; Stipulation & Order of Settlement & Dismissal at 2, Garcia v. Hicks, No. 1:08-cv-07778 (S.D.N.Y. May 7, 2013), ECF No. 108 (providing, in the settlement agreement, that “[t]he United States is substituted as a defendant for defendants Hicks and Suarez”); Notice of Substitution, Williams v. Smith, No. 1:07-cv-01382 (M.D. Pa. Apr. 15, 2013), ECF No. 115 (filing formal substitution of the United States for individual defendants before filing of the settlement); see also U.S. Fed. Bureau of Prisons, Case Details: Hammond v. Sherman (2016) (on file with authors) (reporting, in the BOP’s case details for Hammond v. Sherman, No. 1:05-cv-00339 (W.D. Pa.), that the “[c]ase [was] converted to FTCA, Bivens defendants dismissed” (capitalization altered)). 118. Complaint at 3-4, 27, McCarroll, No. 9:09-cv-00355 (N.D.N.Y. Mar. 27, 2009), ECF No. 1. 119. Stipulation for Compromise Settlement Pursuant to 28 U.S.C. § 2677, supra note 117, at 1-2. 589 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) of limitations.120 For example, in Johnson v. Martinez, the plaintiff alleged he was provided inadequate medical care in July 2002.121 The case was filed as a Bivens case, but the parties agreed after three years of litigation that the plaintiff would dismiss claims against the individual defendants and file an amended complaint against only the United States because it was “necessary to allow settlement of the case.”122 The amended complaint was filed in March 2007 and the case was voluntarily dismissed that month.123 There is no indication in the amended complaint or motion papers that the plaintiff exhausted his administrative remedies under FTCA, and no recognition of the fact that the case was filed long after the statute of limitations on the FTCA claim had run. Another case, Stine v. Allred, was brought as a Bivens claim by a pro se plaintiff who alleged inadequate dental care.124 The parties reached a settlement on April 12, 2013; as the transcript of a court conference that day makes clear, the parties agreed that the plaintiff would file an amended 120. As we discuss below in notes 205-06 and accompanying text, the filing of an 121. 122. 123. 124. administrative claim pursuant to 28 U.S.C. § 2401 is a jurisdictional requirement. And prior to the Supreme Court’s decision in United States v. Kwai Fun Wong, 135 S. Ct. 1625, 1633 (2015), filing an administrative claim within the time requirements set forth by § 2401(b) was jurisdictional in most circuits. See infra notes 210-13 and accompanying text. For three cases in which there is an indication that the plaintiff exhausted administrative requirements, see Order Denying the Motion for an Order Allocating the Settlement Proceeds at 1, Caballero v. Mejia, No. 4:13-cv-00630 (N.D. Fla. Oct. 7, 2015), ECF No. 113 (discussing the parties’ agreement to settle the FTCA claim and noting that “the practical effect [of the settlement] was to end this Bivens case as well”); Plaintiff ’s Agreed Motion to Apportion Settlement Funds Equally Between Claims Alleged Pursuant to Bivens & the Federal Tort Claims Act at 1, Caballero, No. 4:13-cv00630 (N.D. Fla. July 20, 2015), ECF No. 103 (“At the request of the Government, Plaintiff never filed a complaint to initiate a formal action for the FTCA Claim.”); Unopposed Motion for Leave to File an Amended Complaint at 2, Montoya v. Wall, No. 1:11-cv-01414 (C.D. Ill. Oct. 22, 2014), ECF No. 49 (reporting, in the motion to amend the complaint to dismiss Bivens causes of action and add an FTCA claim, that the plaintiff exhausted the FTCA claim before filing and that the original complaint was filed within the statute of limitations); Plaintiff ’s Fourth Amended Complaint at 14, AlKidd v. Sugrue, No. 5:06-cv-01133 (W.D. Okla. Jan. 23, 2008), ECF No. 56-1 (reporting, in a proposed amended complaint adding the FTCA claim, that administrative remedies were exhausted). Second Amended Complaint at 2-4, Johnson v. Martinez, No. 2:04-cv-01967 (E.D. Pa. Mar. 8, 2007), ECF No. 52. Uncontested Motion for Leave to File Amended Complaint at 1, Johnson, No. 2:04-cv01967 (E.D. Pa. Mar. 2, 2007), ECF No. 49; see also Stipulation for Voluntary Dismissal with Prejudice of All Claims Against Defendants Martinez, Zagame, & Hofferica at 1-2, Johnson, No. 2:04-cv-01967 (E.D. Pa. Mar. 2, 2007), ECF No. 48. Order, Johnson, No. 2:04-cv-01967 (E.D. Pa. Mar. 22, 2007), ECF No. 53 (reporting settlement and ordering dismissal of the case); Second Amended Complaint, Johnson, No. 2:04-cv-01967 (E.D. Pa. Mar. 8, 2007), ECF No. 52. Prisoner Complaint at 4, 4A, Stine v. Allred, No. 1:11-cv-00109 (D. Colo. Jan. 14, 2011), ECF No. 1. 590 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) complaint (that substituted the United States for the individual defendants named in the case) and then the United States would pay the plaintiff $2000 from the Judgment Fund.125 There was no indication in the amended complaint that the plaintiff had administratively exhausted his claim, and the claim was filed beyond the statute of limitations.126 Moreover, in another five of these twenty-one cases, the United States agreed to pay to resolve FTCA claims that were previously dismissed by the court or rejected through the administrative process. For example, in Shepherd v. Palmer, the court dismissed the plaintiff’s FTCA claim sua sponte 125. See Transcript of Rule 16(b) Scheduling Conference Held on April 12, 2013 at 6-8, Stine, No. 1:11-cv-00109 (D. Colo. May 16, 2013), ECF No. 362; Third Amended Verified Prisoner Complaint at 1, Stine, No. 1:11-cv-00109 (D. Colo. Apr. 12, 2013), ECF No. 358; Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2672, at 1-2, Stine, No. 1:11-cv-00109 (D. Colo. Apr. 12, 2013) (on file with authors). 126. For other cases in which FTCA claims were added without evidence of exhaustion prior to the two-year statute of limitations, see Stipulation & Order to Modify Caption & Dismiss Case with Prejudice at 2-4, Bolden v. Beaudouin, No. 1:14-cv-05470 (E.D.N.Y. Apr. 12, 2017), ECF No. 69 (showing the substitution of FTCA claim for the Bivens claim in amended complaint post-settlement, past the statute of limitations, and with no indication the FTCA claim was exhausted); Stipulation & Order of Settlement & Dismissal, Garcia v. Watts, No. 1:08-cv-07778 (S.D.N.Y. May 7, 2013), ECF No. 108 (showing the substitution of FTCA claim for the Bivens claim in stipulation of settlement with no FTCA claim—without amending the complaint—past the statute of limitations and with no indication that the FTCA claim was exhausted); Status Report at 1-2, Laurent v. Castellanos, No. 1:14-cv-03340 (E.D.N.Y. Feb. 8, 2016), ECF No. 77 (showing the substitution of FTCA claims for tort claims against individual defendants); Joint Status Report at 1, Freeman v. Woolston, No. 1:11-cv-01756 (D. Colo. Sept. 13, 2013), ECF No. 270 (showing the substitution of FTCA claim for the Bivens claim past the statute of limitations, and with no indication the FTCA claim was exhausted); Joint Motion to Dismiss Individual Defendants & Substitute the United States of America at 1-2, De Anda v. Smith, No. 1:10-cv-01094 (C.D. Ill. Sept. 3, 2013), ECF No. 77 (showing the substitution of FTCA claims for claims against individual defendants); Notice of Substitution, Williams v. Smith, No. 1:07-cv-01382 (M.D. Pa. Apr. 15, 2013), ECF No. 115 (showing the substitution of FTCA claim for the Bivens claim two days before settlement by operation of letter—without amending the complaint—past the statute of limitations and with no indication that the FTCA claim was exhausted); Amended Complaint at 2, Buckley v. Harding, No. 1:06-cv-00413 (D. Colo. Aug. 20, 2007), ECF No. 60 (substituting FTCA claim for the Bivens claims in amended complaint post-settlement, past the statute of limitations, and with no indication the FTCA claim was exhausted); Release & Settlement Agreement at 2, Nunez v. Lindsay, No. 3:05-cv-01763 (M.D. Pa. June 9, 2007) (on file with authors) (showing substitution of administrative FTCA claim for the Bivens claim); U.S. Fed. Bureau of Prisons, Case Details: Hammond v. Sherman et al (2016) (on file with authors) (describing Hammond v. Sherman, No. 1:05-cv-00339 (W.D. Pa.), as having been “converted to FTCA” and noting that the Bivens defendants were dismissed (capitalization altered)); U.S. Fed. Bureau of Prisons, Case Details:Williams v. Warmerdorf et al (2016) (on file with authors) (identifying Williams v. Warmerdorf, No. 3:07-cv-01283 (M.D. Pa.), as a “Bivens 8th Amend[ment]” type case and noting that the case has been “converted to FTCA” (capitalization altered)). 591 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) with prejudice because he had not complied with exhaustion requirements for that claim.127 After the Bivens claims were litigated for two years, the parties submitted a joint agreement that the plaintiff would file an FTCA claim and dismiss the individual defendants.128 The amended FTCA complaint was filed, and a settlement agreement between the plaintiff and the United States was filed less than two weeks later.129 In another case, Willis v. Lappin, the plaintiff’s FTCA claims and his Bivens claims against all but one officer were dismissed by the court at summary judgment.130 The United States subsequently agreed to pay $3000 to the plaintiff to resolve the remaining Bivens claim.131 Although the FTCA claim against the United States had been dismissed by the court, the settlement document was titled “Stipulation for Compromise Settlement and Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677.”132 Finally, in Gillings v. Lepe, the plaintiff filed an administrative FTCA claim, which was denied, then filed his Bivens case, after 127. Order at 3, Shepherd v. Palmer, No. 1:14-cv-02992 (D.N.J. June 25, 2014), ECF No. 10. 128. Consent Order Nunc Pro Tunc to Grant Plaintiff Leave to File a Second Amended Complaint, supra note 109, at 1-2. 129. See Letter from Surinder K. Aggarwal to Noel L. Hillman, supra note 109; Second Amended Complaint at 2, Shepherd, No. 1:14-cv-02992 (D.N.J. Aug. 4, 2016), ECF No. 50. Such a sequence of events is not uncommon. See Opinion & Order at 4-5, Brizard v. Terrell, No. 1:11-cv-02274 (E.D.N.Y. Aug. 27, 2012), ECF No. 39 (dismissing plaintiff ’s initial FTCA claim for failure to exhaust); Stipulation for Compromise Settlement & Release at 1, Brizard, No. 1:11-cv-02274 (E.D.N.Y. Dec. 16, 2013), ECF No. 60 (substituting the United States for the individual defendant at settlement); Memorandum Order at 7, Ellis v. United States, No. 1:08-cv-00160 (W.D. Pa. Sept. 28, 2009), ECF No. 41 (dismissing plaintiff ’s initial FTCA claims at summary judgment); Amended Complaint, Ellis, No. 1:08-cv-00160 (W.D. Pa. Nov. 13, 2012), ECF No. 97 (amending the complaint as a condition of settlement years later by dismissing the Bivens claims and substituting the United States for the individual defendant). 130. Order Adopting Findings & Recommendations, Order Granting in Part Defendant’s Motion to Dismiss, Order for This Case to Proceed Only Against Defendant Devere for Failure to Protect Plaintiff Under the Eighth Amendment, & Order Dismissing All Other Claims & Defendants at 3, Willis v. Lappin, No. 1:09-cv-01703 (E.D. Cal. Mar. 19, 2013), ECF No. 60. 131. See Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, at 1-2, Willis, No. 1:09-cv-01703 (E.D. Cal. May 17, 2013) (filed as an exhibit to Expedited Motion to Enforce Settlement, Willis, No. 1:09cv-01703 (E.D. Cal. July 22, 2013), ECF No. 66). 132. Id.; see also Order Adopting Findings & Recommendations, Order Granting in Part Defendant’s Motion to Dismiss, Order for This Case to Proceed Only Against Defendant Devere for Failure to Protect Plaintiff Under the Eighth Amendment, & Order Dismissing All Other Claims & Defendants at 3, Willis, No. 1:09-cv-01703 (E.D. Cal. Mar. 19, 2013), ECF No. 60. 592 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) which the U.S. government offered to settle the previously denied FTCA claim in exchange for dismissal of the Bivens case.133 Although the litigation records in most of these cases offer few clues about the negotiations that led to these formal substitutions of FTCA claims for Bivens claims, there are occasional indications that the government was more willing to settle cases brought under the FTCA than under Bivens. For example, in Al-Kidd v. Sugrue, a Bivens case challenging the plaintiff’s detention under the material witness statute, the plaintiff moved to file a Fourth Amended Complaint that dismissed the Bivens claim and added an FTCA claim, explaining that “[t]he parties agree that adding a cause of action under the FTCA against the United States, based on the same allegations that have already been pleaded against Warden Sugrue . . . would facilitate settlement of this action.”134 b. Informal substitution In nineteen cases, the United States was never formally substituted for the individual defendants in the cases, but it was clear from the settlement agreement or from our conversations with plaintiffs’ or government attorneys in these cases that the United States paid the settlements. For example, in Brown v. Laing, a Bivens case in which the plaintiff was beaten by BOP officers, the pro se plaintiff never named the United States or brought an FTCA claim, but the settlement agreement was entered into among the plaintiff, individual defendants, and the United States of America and was described as a “complete and final settlement under the Federal Torts Claims Act of matters involved in, or relating to, or arising out of, the above-captioned case.”135 Similarly, in Merriweather v. Zamora and Brown v. Blocker, the United States was not a party to either case (although it was named in the settlement agreement), but the 133. See Opposition to Motion to Alter & or Rescission of [sic] Settlement Agreement & Countermotion to Dismiss with Prejudice at 2-3, Gillings v. Lepe, No. 1:12-cv-01533 (E.D. Cal. Jan. 28, 2015), ECF No. 67. 134. Plaintiff ’s Consent Motion to File a Fourth Amended Complaint at 2, Al-Kidd v. Sugrue, No. 5:06-cv-01133 (W.D. Okla. Jan. 23, 2008), ECF No. 56; see also, e.g., Stipulation for Voluntary Dismissal with Prejudice of All Claims Against Defendants Martinez, Zagame, & Hofferica at 1-2, Johnson v. Martinez, No. 2:04-cv-01967 (E.D. Pa. Mar. 2, 2007), ECF No. 48 (noting that with the plaintiff ’s amended complaint, claims against the individual defendants would be dismissed and that plaintiff would only proceed on his claims against the United States); Second Amended Complaint at 2, Johnson, No. 2:04-cv-01967 (E.D. Pa. Mar. 8, 2007), ECF No. 52 (naming the United States as the defendant); Uncontested Motion for Leave to File Amended Complaint, Johnson, No. 2:04-cv-01967 (E.D. Pa. Mar. 2, 2007), ECF No. 49 (reporting that amending the complaint was “necessary to allow settlement of the case”). 135. Settlement Agreement & Release of All Claims at 1, Brown v. Laing, No. 6:09-cv-00392 (E.D. Ky. Sept. 21, 2012) (on file with authors). 593 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) United States paid the settlements.136 Another sixteen Bivens actions in the dataset resulted in similar agreements with the United States—the settlement agreements are described as settlements of FTCA claims or settlements with payments by the United States, but the plaintiff never filed a complaint with a cause of action under the FTCA.137 There is no indication that FTCA claims were administratively exhausted in any of these nineteen cases, and in the vast majority of them it is inconceivable that any FTCA claim could have been filed within the statute of limitations. C. The Impact of Judgment Fund Payments on the Bureau of Prisons Our research makes clear that Bureau of Prisons employees very rarely contributed to settlements and judgments in actions brought against them. These data also establish that the employing agency, the BOP, was not held financially responsible for the settlement of these cases. Instead, all available evidence suggests that the settlements were satisfied through the Judgment Fund, and that costs of settlements and judgments were not taken from the BOP’s budget. Many settlement agreements expressly state that the settlement funds were being paid from the U.S. Treasury (where the Judgment Fund is 136. Complaint at 1-2, Merriweather v. Zamora, No. 2:04-cv-71706 (E.D. Mich. May 12, 2004), ECF No. 3 (alleging the prison violated the plaintiffs’ mail privacy rights); Email from Daniel Manville, Clinical Professor, Mich. State Univ. Coll. of Law, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (May 20, 2018, 6:33 PM) (on file with authors) (“The government paid the settlement in Merriweather. I got one check and it was from the government.”); see also Stipulation for Compromise Settlement at 1-2, Brown v. Blocker, No. 2:09-cv-00434 (D.S.C. Mar. 26, 2010), ECF No. 45 (providing that “[t]he United States agrees to pay to Lloyd Eugene Brown $15,000.00 in full settlement” of his claims against the individual defendants); Complaint at 12-13, Brown, No. 2:09cv-00434 (D.S.C. Feb. 23, 2009), ECF No. 1 (alleging failure to accommodate medical condition and denial of adequate medical care). 137. See, e.g., Stipulation for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2672, at 1, Mundo v. Shaw, No. 1:12-cv-00184 (N.D. W. Va. Jan. 6, 2014) (describing the settlement as a “Stipulation for Compromise Settlement and Release of Federal Tort Claims Act Claims,” where the United States was never named as a defendant in the case alleging religious discrimination and harassment); Settlement Agreement & Release of All Claims at 2, Brown, No. 2:09-cv-00392 (E.D. Ky. Sept. 21, 2012) (noting in the settlement agreement that the $9000 settlement shall be paid “by government electronic fund transfer”); Settlement Agreement & Release Between Plaintiff & Bureau of Prisons at 4, Montgomery v. Johnson, No. 7:05-cv-00131 (W.D. Va. Apr. 27, 2010) (providing that the entirety of settlement will be paid “by a check drawn on the Treasury of the United States”); Stipulation for Compromise Settlement at 2, 4-5, Green v. Wiley, No. 1:07-cv-01011 (D. Colo. July 1, 2009), ECF No. 137 (noting plaintiff ’s allegations of inadequate medical care and providing that the $2100 to be paid by the U.S. government was “in settlement of his Federal Tort Claim Act action as set forth in the above-captioned case,” where there was no claim against the United States or assertion of an FTCA claim). 594 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) located).138 And our conversations with BOP representatives confirm that payments of FTCA and Bivens claims are almost certainly paid from the Judgment Fund.139 The records we examined additionally suggest that the BOP has only incomplete information about the cases that were settled on behalf of their officers. The case files produced by the BOP in response to our FOIA request did not consistently include the causes of action alleged, and many case files did not include a settlement agreement or other evidence of the amount paid to resolve them.140 We understand from our communications with the BOP that the agency and its officers are typically represented in these cases by lawyers in local U.S. Attorneys’ offices who negotiate and draft the settlement agreements.141 BOP attorneys “seek approval for settlement amounts but typically don’t sign the agreements themselves,” and do not consistently retain a copy of the agreements for the BOP’s records.142 We understand from communication with the BOP that the agency has “an internal litigationtracking database,” and that BOP employees who enter litigation records into 138. See, e.g., Stipulation for Compromise Settlement & Release of Federal Tort Claims Act 139. 140. 141. 142. Claims Pursuant to 28 U.S.C. § 2677, at 3, Stine v. Allred, No. 1:11-cv-00109 (D. Colo. Apr. 12, 2013) (“Payment of the settlement amount will be made by check drawn on the Treasury of the United States . . . .”); Settlement Agreement & Release Between Plaintiff & Bureau of Prisons at 4, Montgomery, No. 7:05-cv-00131 (W.D. Va. Apr. 27, 2010) (same); Stipulation & Order for Compromise Settlement & Release of Federal Tort Claims Act Claims Pursuant to 28 U.S.C. § 2677, at 3, Oleson v. United States, No. 3:05-cv-00033 (W.D. Wisc. June 6, 2006) (same); see also, e.g., Stipulation of Dismissal at 1, Kikumura v. Osagie, No. 1:03-cv-00236 (D. Colo. Nov. 5, 2008), ECF No. 177 (noting the parties’ stipulation to dismissal of the case with prejudice, “as the settlement payment ha[d] been made from the Treasury Department’s Judgment Fund”). See Email from Ian M. Guy, Supervisory Attorney-Advisor, FOIA & PA Section, Office of Gen. Counsel, Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. Of Law (Mar. 8, 2017, 10:05 AM) (on file with authors) (“If you are seeking information regarding which ‘pot’ of money was used to pay a settlement or judgment, then the records do not show that information. Instead, the general rules are applicable in a vast majority of the cases: FOIA litigation expenses are out of the BOP’s operating budget; while virtually all tort, EEO, Bivens, etc., costs are paid from the DOJ judgment fund.”). In separate research conducted using the Bloomberg Law electronic database, we identified sixty-three Bivens cases filed between 2005 and 2014 which resulted in a settlement and which were not included in the BOP’s FOIA disclosures. See supra note 12. We include these cases in our analysis, as is described in Appendix A. Most relevant for this discussion is the fact that BOP did not initially produce records from these cases in response to our FOIA requests. We have no reason to believe that the BOP was acting in bad faith; perhaps, however, the failure to produce these records indicates weaknesses in the BOP’s record collection and retention systems. Email from Ronald L. Rodgers, Senior Counsel, Info. & Remedies Processing Branch, Office of Gen. Counsel, Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Mar. 1, 2019, 7:46 AM) (on file with authors). Id. 595 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) that database can include information about the amount paid to resolve a case and the terms of the settlement agreement.143 Indeed, the BOP produced several documents—what it referred to as “face sheets”—reflecting settlement information entered into the BOP’s internal litigation-tracking database.144 But the records produced by the BOP indicate that the agency has only a partial picture of the way its activities lead to the imposition of legal liability on the U.S. government or of what steps it could take to reduce that liability risk. The 171 successful Bivens cases in our dataset take a variety of paths from filing to resolution. But, in almost every instance, cases are restructured during litigation or at settlement to substitute Bivens claims for claims under the Federal Tort Claims Act. Government attorneys pursue this approach even when doing so overlooks statute of limitations or jurisdictional defenses, and even when the same attorneys have successfully sought dismissal of plaintiffs’ FTCA claims at an earlier point of the litigation. Rather than require individual officers to bear financial liability, or seek formal indemnification by the BOP, government attorneys appear to prefer to shift liability in Bivens cases to the federal treasury’s Judgment Fund. As a result, both individual officers and the BOP are spared the financial consequences of almost all successful claims. III. Implications Our study calls for a fresh evaluation of the current state of Bivens doctrine. At the most basic level, the data contradict the Supreme Court’s repeated assertion that federal officials face a threat of significant personal financial responsibility in these cases: The threat of personal liability appears from our data to be far more theoretical than real. To the extent one can generalize from our data,145 the Court’s hostility to the Bivens right to sue and its expansive conception of qualified immunity both appear to rest on a conception of personal liability that the facts do not sustain. Apart from its implications for the personal liability model of Bivens litigation, the study casts doubt on three conventional assumptions. In contrast to the presumption that agencies must use appropriated funds to indemnify individual officials, our data indicate that the BOP enjoys virtually airtight protection from the financial consequences of its employees’ wrongdoing. Second, the findings reported here call into question the reliability of 143. See Letter from Ian M. Guy, Supervisory Attorney, Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Oct. 24, 2017) (on file with authors). 144. Id. Further research should explore whether or to what extent the BOP uses these records in the analysis of risk. 145. We explore some of the limits of our study in Part III.E below. 596 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) representations made by Department of Justice attorneys to all levels of the federal judiciary about the financial threat that Bivens cases pose to federal officials. Third, our data reveal an executive branch payment practice that may well contravene congressional expectations about the relationship between Bivens, the FTCA, and the Judgment Fund. We take up these implications in turn and conclude with thoughts about the future of Bivens liability and questions for future research. A. Constitutional Torts and Individual Deterrence After Ziglar Almost all modern theories of tort law proceed on the assumption that the risk of liability shapes primary behavior.146 True, scholars recognize that tort law’s deterrent signal can be dimmed for a variety of reasons.147 And scholars 146. See generally John C.P. Goldberg, Twentieth-Century Tort Theory, 91 GEO. L.J. 513, 525 (2003) (“And so we arrive at the baseline proposition of compensation-deterrence theory, repeated at the outset of countless law review articles published in the last fifty years: The function of tort law is to compensate and deter.”). 147. Both informational and structural barriers may make it extremely difficult to detect and remedy wrongdoing, reducing the deterrent effect of potential liability in both the constitutional and common law contexts. See Miriam H. Baer, Pricing the Fourth Amendment, 58 WM. & MARY L. REV. 1103, 1169 (2017) (describing “run-of-the-mill constitutional violations” as “difficult to detect”); Thomas C. Galligan, The Risks of and Reactions to Underdeterrence in Torts, 70 MO. L. REV. 691, 698 (2005) (identifying “difficulty of detection” as one explanation for underdeterrence in tort); Sara Sternberg Greene, Race, Class, and Access to Civil Justice, 101 IOWA L. REV. 1263, 1274 (2016) (reviewing research showing that unequal access to resources affects the ability to obtain civil remedies). Even when wrongdoing is detected, entities with power to change policies and reform behavior may ignore that information, a phenomenon one of us has discussed in the context of suits against municipal law enforcement agencies. Joanna C. Schwartz, Myths and Mechanics of Deterrence: The Role of Lawsuits in Law Enforcement Decisionmaking, 57 UCLA L. REV. 1023, 1028 (2010) (reporting that many law enforcement agencies appear not to gather and analyze litigation information for personnel and policy implications). And individual officers may purchase insurance to limit their risk, which results in cost spreading that can mitigate specific deterrence— this is the familiar problem of moral hazard. See John Rappaport, How Private Insurers Regulate Public Police, 130 HARV. L. REV. 1539, 1545 (2017) (describing concerns about insurance and moral hazard). Finally, specific agreements to indemnify wrongdoers may also reduce the deterrent impact of individual liability. Theodore Eisenberg & Stewart Schwab, The Reality of Constitutional Tort Litigation, 72 CORNELL L. REV. 641, 686 (1987) (reporting in a study of § 1983 litigation that the authors observed no cases in which judgments or settlements were paid by individual officers, suggesting that “rampant official fear of personal liability may be an overreaction”). In areas well outside the context of constitutional litigation, indemnification is thought to undermine the deterrent impact of legal sanctions. See. e.g., Samuel W. Buell, Criminal Procedure Within the Firm, 59 STAN. L. REV. 1613, 1655-56 (2007) (discussing the relationship between deterrence and indemnification in corporate firms, suggesting that indemnification “causes managers to engage in desirable risk-taking once they begin employment”); Susan B. Heyman, Corporate Privilege and an Individual’s Right to Defend, 85 GEO. WASH. L. REV. 1112, 1160 (2017) (discussing the SEC’s position that footnote continued on next page 597 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) continue to debate the degree to which theories of deterrence apply to the specialized field of constitutional tort doctrine, where the government’s role in paying compensation has been thought by some to moderate the incentive effects of tort verdicts.148 Finally, scholars have disagreed about where the Court might best assign the incidence of tort liability to ensure that the relevant government actors take appropriate steps to reduce the likelihood of constitutional violations.149 Whatever the current state of scholarly disputation (and most assume that the liability rule matters a great deal150), the Supreme Court for its part has been quite clear in expressing its conception of the proper function of constitutional tort doctrine. Bivens has been justified from the outset as a means of deterring individual officers from violating the Constitution.151 As the story goes, if officers know that they face personal liability for constitutional violations, they will be less likely to violate those constitutional provisions.152 The Supreme Court not only accepts deterrence assumptions wholeheartedly in its Bivens decisions, but also sees the threat of overdeterrence as the fundamental justification for excusing liability through the doctrine of qualified immunity.153 The threat of personal liability has also served as a leading justification for narrowing Bivens doctrine and refusing to extend it further.154 148. 149. 150. 151. 152. 153. 154. indemnification for liabilities under the Securities Act of 1933 is against public policy because it undermines deterrence). Daryl Levinson has expressed skepticism about the ability of money damages to deter government officials, prompting a vigorous debate on the issue. See Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Allocation of Constitutional Costs, 67 U. CHI. L. REV. 345, 347 (2000); see also Myriam E. Gilles, Essay, In Defense of Making Government Pay: The Deterrent Effect of Constitutional Tort Remedies, 35 GA. L. REV. 845, 861 (2001) (responding to Levinson, supra). We have discussed this literature in other work. See Pfander & Hunt, supra note 9, at 1865; Reinert, supra note 9, at 848-49; Schwartz, supra note 147, at 1033-34. See supra note 8. See Reinert, supra note 9, at 815-17 (surveying literature); see also Richard H. Fallon, Jr., Bidding Farewell to Constitutional Torts, 107 CALIF. L. REV. 933, 979-80 (2019) (discussing the desirability of entity liability and the difficulty of achieving that goal in light of sovereign immunity from suit). See Corr. Servs. Corp. v. Malesko, 534 U.S. 61, 71 (2001) (“Bivens from its inception has been based . . . on the deterrence of individual officers who commit unconstitutional acts.”); supra Part I. See Malesko, 534 U.S. at 70 (“The purpose of Bivens is to deter individual federal officers from committing constitutional violations.”); Carlson v. Green, 446 U.S. 14, 21 (1980) (“It is almost axiomatic that the threat of damages has a deterrent effect . . . .”). See Harlow v. Fitzgerald, 457 U.S. 800, 814 (1982); see also supra note 62 and accompanying text (describing the Court’s justifications for qualified immunity). See supra Part I.C. 598 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Yet, our study of BOP payouts undercuts judicial and scholarly assumptions about the financial effects of Bivens actions on one group of federal employees. Individual officers contributed to settlements and judgments in less than 5% of the successful Bivens cases filed against employees of the BOP.155 When one considers the universe of all Bivens cases brought against BOP officers, the likelihood of officer contribution appears even more remote. One of us previously found that approximately 15.3% of Bivens actions filed against BOP officials result in a plaintiff’s settlement or judgment.156 If so, then over the ten-year period of this study there may have been 1100 or more Bivens cases filed against BOP officials.157 The eight cases in which officers contributed would amount to 0.7% of those cases filed. One final perspective bears mentioning. The Bureau of Prisons has 36,793 employees.158 Extrapolating from the study data, and assuming that all employees engage in wrongdoing at the same rate, less than 0.1% of BOP employees will contribute to a settlement or judgment during a twenty-year career.159 In short, the overriding purpose of Bivens liability—deterring the misconduct of individual officers by imposing monetary damages for constitutional violations—appears to have fallen out of the equation given these settlement practices. And the overriding purpose of protecting individual officers by limiting Bivens liability—through qualified immunity and contraction of the right to sue— appears to be unnecessary given the ways in which Bivens actions are litigated and resolved. The data on individual payments further reveal that, in the rare circumstance when defendants are obliged to use personal resources to resolve Bivens claims, the burden invariably falls on line employees of the BOP. Thus, 155. See supra Part II.A. 156. Reinert, supra note 9, at 836. 157. Indeed, this may underestimate the total number of Bivens cases filed against BOP employees. First, the prior research that one of us conducted regarding success rates in Bivens cases covered only five judicial districts. Id. at 832. Second, although we have searched exhaustively for all settlements of Bivens claims filed against BOP employees over the study period, some may have escaped our notice. If the success rate of all Bivens cases involving BOP employees were lower than 15%, and if there are additional Bivens settlements that we have not uncovered, then the denominator of total Bivens claims filed against BOP employees is likely well over 1100. 158. BOP Statistics: Staff Ethnicity/Race, FED. BUREAU PRISONS (last updated Jan. 11, 2020), https://perma.cc/S86N-2YE8. 159. If twelve BOP employees were required to contribute to settlements and judgments during the ten-year period of this study, then approximately twenty-four officers out of 36,793 would contribute to settlements and judgments over a period of twenty years, which is the length of time officers can serve before retirement. See Discover What Life Is Like Working for the BOP, FED. BUREAU PRISONS, https://perma.cc/X6BA-3FNM (archived Jan. 22, 2020) (explaining that officers are eligible to retire at age fifty with twenty years of service, and at any age with twenty-five years of service). 599 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) in the eight cases in which we identified payments from personal resources, all payments were made by employees who have direct contact with plaintiffs in the prison system. Five payments were from guards who committed sexual or physical assaults on prisoners; three payments were made (perhaps with the assistance of insurance coverage) after substandard medical care.160 In no case did we see evidence of any personal payment by the supervisors responsible for the prisons or practices in question. The absence of supervisory participation in personal liability payments raises questions about the degree to which the Supreme Court was well-advised to view personal supervisory liability as a threat grave enough to necessitate the elaborate precautions taken in such cases as Ashcroft v. Iqbal and Ziglar v. Abbasi.161 To be sure, individual liability does not stand alone as the only source of deterrence (or overdeterrence) in the context of constitutional litigation. Nonmonetary pressures of various sorts create incentives to comply with the law.162 Officers can theoretically be deterred (and overdeterred) by the threat of reputational harm occasioned by allegations of constitutional misconduct (just as the reputational rewards for successful law enforcement activity may encourage officers to take shortcuts in other contexts).163 Federal officers might overestimate the minuscule risk that they will have to pay a Bivens judgment,164 particularly if they attend carefully to the Department of Justice policy on indemnification.165 And the practical consequences of being a defendant (having to sit for a deposition or sign interrogatories) could function as a deterrent (or overdeterrent).166 Civil rights cases can put political pressure 160. See supra notes 82-83 and accompanying text. 161. See Ziglar v. Abbasi, 137 S. Ct. 1843, 1865 (2017) (calling for a renewed special-factors 162. 163. 164. 165. 166. analysis of the claims against a prison warden); Ashcroft v. Iqbal, 556 U.S. 662, 686 (2009) (narrowing supervisory liability in equal protection claims and imposing a plausibility pleading regime aimed at protecting high government officials from the threat of personal liability). See, e.g., Bernard Black et al., Outside Director Liability, 58 STAN. L. REV. 1055, 1139-41 (2006) (reporting that outside directors face little risk of being held personally accountable for breaches of duty to the corporation and concluding that the more substantial risks are the time, aggravation, and potential harm to reputation that a lawsuit can entail). Reinert, supra note 9, at 847-49; Schwartz, supra note 9, at 941. Schwartz, supra note 9, at 941. This may explain the anecdotal evidence we have heard that some federal officers purchase liability insurance. See supra note 87. Having found nothing in our data that sheds light on the practice, we view the frequency with which federal officers purchase such insurance as a promising topic for future research. See infra text accompanying notes 187-88. Daniel J. Meltzer, Deterring Constitutional Violations by Law Enforcement Officials: Plaintiffs and Defendants as Private Attorneys General, 88 COLUM. L. REV. 247, 283 (1988); Reinert, supra note 9, at 847-49. Criminal prosecution has the potential to deter, but such prosecutions are rarely initiated by the Department of Justice, in part because of footnote continued on next page 600 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) on government actors by exposing embarrassing details about official misconduct.167 Without questioning the importance of such factors (and without evaluating them separately168), we make the simple point that the data collected here reveal almost no support for the notion that individual officials face a genuine threat of personal liability, a central premise of many subsequent limitations on Bivens liability. B. Agency Incentives and the Role of Indemnification Direct personal liability is not, however, the only way for Bivens to achieve its deterrent purpose. The Supreme Court has long understood that federal government agencies may indemnify their employees, holding them harmless from constitutional tort and other personal liability. In the past, the Court seemed to accept indemnification as a potential reality in Bivens litigation, but not as a factor that argued for or against expanding personal capacity claims against federal officials.169 In Ziglar, the Court plainly assumed that indemnity might shift a portion of personal liability from individual Bivens defendants to the agencies for which they worked, and for the first time viewed the possibility of indemnification as a reason to defer to Congress and disfavor implied Bivens actions.170 Yet the Court did not articulate a view about the frequency with which the federal government indemnified its officers, and which government entity would absorb the costs. While the Court has not described the Bivens regime as a vehicle for inducing agencies to reduce the costs of constitutional violations,171 one can readily see how the combination of personal liability and agency indemnification might play that role. Indemnity shifts the payment obligation, thereby saddling the agency with liability, inducing the agency to take account of the risk of constitutional liability and encouraging the agency to institute policies 167. 168. 169. 170. 171. the high burden established decades ago by the Supreme Court. See Screws v. United States, 325 U.S. 91, 103 (1945) (plurality opinion) (holding that prosecutors charging officials with violations of 18 U.S.C. § 52 (now 18 U.S.C. § 242) must prove that the defendant had the “specific intent” to deprive a person of their constitutional rights). Gilles, supra note 148, at 860-61 (discussing the potential value of publicity in the context of municipal litigation); Alexander A. Reinert, Screening Out Innovation: The Merits of Meritless Litigation, 89 IND. L.J. 1191, 1229-30 (2014) (discussing the role of litigation in exposing problematic institutional dynamics). Schwartz, supra note 9, at 942-43 (reviewing literature). See supra note 69. Ziglar v. Abbasi, 137 S. Ct. 1843, 1856 (2017). Corr. Servs. Corp. v. Malesko, 534 U.S. 61, 70-71 (2001); FDIC v. Meyer, 510 U.S. 471, 485-86 (1994). 601 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) designed to prevent or reduce such risks.172 Some theories of tort liability presume that it is rational to impose liability on the entity best situated to make cost-effective changes that can influence the behavior of individual actors.173 This is one of the premises behind the Supreme Court’s decision to deny qualified immunity to municipalities in § 1983 litigation.174 To the extent agencies must pay for settlements or judgments in lawsuits against individual officers, agency administrators may have an incentive to minimize those payments.175 For this theory to apply in practice, however, the agency responsible for employing, training, and supervising employees engaged in misconduct must 172. See, e.g., SCHUCK, supra note 8, at 100-07 (seeking to determine the “cheapest cost avoider” within the limited set of potential “public defendants” in order to deter government misconduct); Gilles, supra note 148, at 859-67 (arguing in favor of the deterrent effect of governmental liability); Margo Schlanger, Inmate Litigation, 116 HARV. L. REV. 1555, 1672-90 (2003) (discussing ways in which indemnification can influence jails and prisons); cf. Michael C. Pollack, Taking Data, 86 U. CHI. L. REV. 77, 119-20 (2019) (arguing that imposing up-front costs on a government agency seeking private information held by internet service providers may provide more individual privacy protection than existing statutory and constitutional frameworks). Judge Jon Newman advocated in favor of abolishing absolute immunity for judges and prosecutors while simultaneously imposing liability on governmental agencies as a way of increasing deterrence through entity liability rather than individual liability. Jon O. Newman, Suing the Lawbreakers: Proposals to Strengthen the Section 1983 Damage Remedy for Law Enforcers’ Misconduct, 87 YALE L.J. 447, 463 (1978). There are reasons to be skeptical towards the claim that government absorption of officers’ liability will influence policymakers’ decisions. See Barbara E. Armacost, Organizational Culture and Police Misconduct, 72 GEO. WASH. L. REV. 453, 475 (2004) (identifying political reasons that police officials may not take measures to prevent unconstitutional policing even in the face of significant liability); Levinson, supra note 148, at 355-57; Schwartz, supra note 147, at 1028 (suggesting that law enforcement agencies lack sufficient information to guide policies in response to § 1983 lawsuits). 173. See, e.g., SCHUCK, supra note 8, at 17-18; Kramer & Sykes, supra note 8, at 285-86 (discussing the relative cost-effectiveness of imposing “strict vicarious liability” and “vicarious liability based on negligence” in the context of municipal torts); Anthony J. Sebok, Deterrence or Disgorgement? Reading Ciraolo After Campbell, 64 MD. L. REV. 541, 555-56 (2005) (summarizing and critiquing the “cheapest cost avoider” concept in the context of “socially compensatory damages”); see also Reinert, supra note 9, at 815-17. 174. See Owen v. City of Independence, 445 U.S. 622, 651-52 (1980). 175. In addition to scholarly ambivalence about the manner and extent of deterrence resulting from agency liability, see supra note 172, this claim is subject to the recognition that the deterrent signal of individual liability can be scrambled, see supra note 147. To be sure, when employees and their firms both face liability for the torts committed in the course of operations, they have incentives to bargain over the allocation of liability and may shift it among themselves by contract or otherwise. See generally Alan O. Sykes, The Boundaries of Vicarious Liability: An Economic Analysis of the Scope of Employment Rule and Related Legal Doctrines, 101 HARV. L. REV. 563, 566 (1988) (“[T]he choice between a rule of personal or vicarious liability may be unimportant.”). But if neither the employee nor the firm faces such liability, bargaining between employee and firm may not occur. 602 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) bear the incidence of liability.176 Our data show that, at least in the context of settlements involving federal prison employees, the BOP incurs few if any costs in the payment of settlements and judgments.177 The settlement practices we have observed and documented do not give any reason to believe that the BOP has litigation-related financial incentives to create policies and procedures that will reduce the risk that their officers commit constitutional violations. Other than the exceedingly rare settlement to which an individual employee contributed, every settlement we documented here appeared to have been paid directly by the United States Treasury from the standing Judgment Fund appropriation, not from the budget of the individual agency. Moreover, the BOP’s records, produced to us in response to our FOIA request, suggest that the BOP possesses only incomplete data about the nature of claims brought against its officers and the amounts paid to resolve these claims.178 Without more complete information about these cases, the BOP cannot take informed steps to reduce their incidence. We do not mean to overstate the point. Even if the BOP paid settlements directly, such payments would constitute a negligible portion of the BOP’s annual budget.179 We are not certain that such modest financial signals would have an appreciable impact on agency policy.180 But, in the analogous § 1983 176. The individuals responsible for instituting changes in policies and practices must also 177. 178. 179. 180. learn of the underlying misconduct. See generally Schwartz, supra note 147 (reporting that many law enforcement agencies fail to gather and analyze information from lawsuits brought against them); Joanna C. Schwartz, What Police Learn from Lawsuits, 33 CARDOZO L. REV. 841 (2012) (describing what several litigation-attentive law enforcement agencies have learned from lawsuits brought against them). For similar findings regarding the insulation of local law enforcement agencies from the costs of civil rights suits against their employees, see generally Joanna C. Schwartz, How Governments Pay: Lawsuits, Budgets, and Police Reform, 63 UCLA L. REV. 1144 (2016). See also Paul David Stern, Tort Justice Reform, 52 U. MICH. J.L. REFORM 649, 720-24 (2019) (emphasizing the importance of agency financial accountability for the torts of federal law enforcement officers). See supra notes 140-44. The BOP’s budget request for fiscal year 2018 was more than $7 billion. See FED. BUREAU OF PRISONS, U.S. DEP’T OF JUSTICE, FY 2018 BUDGET REQUEST AT A GLANCE 1 (n.d.), https://perma.cc/BP2N-SP2Y. Over the course of the ten years covered by our study, the United States appropriated some $64 billion to the Bureau of Prisons. See NATHAN JAMES, CONG. RESEARCH SERV., R42486, APPROPRIATIONS FOR THE BUREAU OF PRISONS (BOP): IN BRIEF 11 tbl.A-1 (2018), https://perma.cc/4T9D-LQBU. The total value of the Bivens claims resolved during the period in question was approximately $18.9 million, or approximately 0.03% of the prison budget. See infra Appendix B, Tables 1-2. We note, however, that there is empirical evidence from other contexts that suggests that decisionmakers within federal agencies are sensitive to funding streams. See, e.g., Michael D. Frakes & Melissa F. Wasserman, Does Agency Funding Affect Decisionmaking?: An Empirical Assessment of the PTO’s Granting Patterns, 66 VAND. L. REV. 67, 70 (2013) (reporting data that suggest that decisions by patent examiners are sensitive to fiscal implications for the Patent and Trademark Office). 603 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) context, one of us has found some evidence of agency deterrence on the relatively rare occasions when local law enforcement agencies are required to pay settlements and judgments from their own budgets.181 Even modest financial consequences appear to have some impact on agency willingness to learn from the liability message conveyed in the lawsuits brought against them. Judgment Fund payments convey no comparable message. As noted earlier, the sum total of Bivens payments represents only about 0.03% of the budget of the BOP for the relevant period. For a person earning $100,000 per year, that amounts to roughly $20. As a share of the entire budget of the government of the United States over the relevant period, the payments lose all signaling power; calibrated as a portion of the salary of our $100,000 earner, they would represent less than a single penny.182 Building on the intuition that such a modest payment obligation would escape the close attention of even the most budget-conscious members of Congress, one can ask further questions about the Court’s view of Bivens liability. As we have seen, the Court has assumed that the expansion of Bivens liability would pose a substantial threat to the federal budget, substantial enough in fact to warrant judicial caution in the recognition of rights to sue. Based on the total amount of BOP payments, however, one might doubt that even broadly expanded constitutional tort liability would lead to financial obligations sufficient to attract congressional attention, let alone to threaten the fisc. We thus question the Ziglar Court’s reliance in part on fiscal concerns for its posture of broad deference to Congress in the recognition of rights to sue under Bivens. The Court has made no effort to quantify or understand the threat to the fisc posed by the prospect of Bivens liability or to compare those costs with the financial burden imposed by defending suits for injunctive and habeas relief that the Court accepts as routine. Congress, for its part, has shown scant interest in the role, if any, that Bivens liability plays in its management of the nation’s finances. We know of no government effort to summarize the 181. See Schwartz, supra note 177, at 1195-96. Some evidence at the federal level suggests that agencies comply with injunctive decrees not because of any concrete threat of contempt sanctions but because such decrees send reputational signals that publicly shame responsible officials. See Nicholas R. Parrillo, The Endgame of Administrative Law: Governmental Disobedience and the Judicial Contempt Power, 131 HARV. L. REV. 685, 777-89 (2018) (exploring the role and limits of public shaming in the creation of incentives to comply with federal judicial decrees). Constitutional tort liability borne directly by the agency through indemnification may convey clearer reputational signals than liability passed along to the Judgment Fund. 182. According to White House data, the federal government spent $34.3 trillion over the ten-year period ending in 2016. See OFFICE OF MGMT. & BUDGET, HISTORICAL TABLES 332 tbl.14.2 (n.d.), https://perma.cc/4VSF-VVLE. One penny, as a share of the salary of our hypothetical $100,000 earner, would correspond to $34.3 million of the federal budget as a whole. 604 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) amount of resources devoted to the defense and settlement of constitutional tort claims, and no indication that those payments threaten federal financial stability.183 Given the relatively modest amounts reflected in the BOP data, we doubt that any such threat exists. C. The Department of Justice Narrative of Personal Liability Apart from revealing little threat of personal or agency liability, the payment practice we document here conflicts with the rhetorical position the government has long taken in representations made to the federal judiciary and to the legal profession in the course of defending Bivens claims. Since the 1980s, the Department of Justice has argued in court filings and public documents that the agency rarely if ever indemnifies individual Bivens defendants; this lack of assured agency indemnity sets up the government’s claim that suits brought under the Bivens doctrine expose individual defendants to potential financial ruin.184 One finds this narrative of personal liability— sometimes characterized as “devastating” or “ruinous” in the government’s briefs—reflected in a series of government submissions to the Court.185 In one brief, Department of Justice counsel explained that, in Bivens actions, “[a] public servant’s bank account, retirement savings, even his or her home is all in jeopardy, along with the fundamental prospect of providing for a family.”186 If one assesses the specific terms of the Department’s indemnity policy, one finds a superficially plausible basis for these representations. Although 183. The Department of Treasury provides annual Judgment Fund reports to Congress, but they do not purport to provide information regarding constitutional tort claims. See Judgment Fund, Annual Report to Congress, BUREAU FISCAL SERV. (last updated Dec. 12, 2019), https://perma.cc/3FHK-9FMV. 184. See, e.g., Brief for the Petitioners at 47-48, Schweiker v. Chilicky, 487 U.S. 412 (1988) (No. 86-1781), 1987 WL 880510 [hereinafter Schweiker Petitioners’ Brief] (arguing against a Bivens remedy because of “ruinous personal liability”). 185. See, e.g., Brief for Petitioners Dennis Hasty & James Sherman at 2, 20, 21, 27, Ziglar v. Abbasi, 137 S. Ct. 1843 (2017) (No. 15-1363), 2016 WL 6873021 (referring to “personal” liability or damages four times); Reply Brief for the Petitioners at 16, Schweiker, 487 U.S. 412 (No. 86-1781), 1988 WL 1026249 [hereinafter Schweiker Petitioners’ Reply Brief] (referring to “the devastating potential liability” facing Bivens defendants); see also Replacement Brief for John Ashcroft, the Official Capacity Defendants-Appellees & the United States at 43, Arar v. Ashcroft, 585 F.3d 559 (2d Cir. 2009) (No. 06-4216-cv) (en banc), 2008 WL 8132330 [hereinafter Arar Replacement Brief for Ashcroft] (arguing that recognizing a Bivens claim “would put at personal risk the officials involved in making the most sensitive and important decisions facing the nation” (emphasis added)); infra notes 195-99 and accompanying text. 186. Defendants’ Memorandum in Opposition to Plaintiff ’s Motion for Leave to Amend & Defendant’s Memorandum in Support of Cross-Motion to Stay Proceedings Pending Appeal at 7, Martin v. Naval Criminal Investigative Serv., No. 3:10-cv-01879 (S.D. Cal. Jan. 30, 2012), ECF No. 54. 605 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Department of Justice policies allow indemnification of its employees in appropriate cases, the indemnification policy is quite strict. For starters, the policy requires a finding by the Attorney General (or a designee) that indemnity is in the interest of the United States.187 In addition, the policy forecloses (except in exceptional circumstances) any request to “indemnify or to settle a personal damages claim before entry of an adverse verdict, judgment, or award.”188 So, while employees may request and receive representation by government lawyers during the pendency of Bivens litigation, they cannot request or secure any assurance as to indemnity for any personal liability until after they lose in court. Such indemnity rules surely complicate the settlement calculus for individual defendants in personal liability actions; they may prefer to settle on terms the Department’s lawyers specify rather than submit to a jury’s verdict as a prelude to presenting their applications for indemnity. Whatever its impact on settlement negotiations, the Department’s indemnity policy cannot sustain its continuing narrative of significant personal exposure to ruinous liability. Our study shows that Bivens defendants rarely contribute their own funds to resolve successful constitutional litigation brought against them. Even when they do, the amounts in question do not threaten financial devastation.189 Yet these facts have had little impact on the government’s narrative. Government attorneys persist in describing Bivens as potentially ruinous even though individual defendants almost never pay judgments or settlements in successful Bivens cases. What’s more, government attorneys play an active role in deliberately repackaging Bivens cases for settlement under the FTCA and Judgment Fund.190 Such repackaging belies any assertion that the Department harbors misconceptions about the ways its practices shift the ultimate incidence of Bivens liability to the U.S. Treasury.191 187. See 28 C.F.R. § 50.15(c)(1) (2019) (permitting indemnification upon request when “such 188. 189. 190. 191. indemnification is in the interest of the United States”). Id. § 50.15(c)(3); see also U.S. DEP’T OF JUSTICE, JUSTICE MANUAL § 4-5.412 (2018), https://perma.cc/5W5J-VXQ7 (noting that “there is no right to compel indemnification” and that “[p]re-judgment indemnification is disfavored . . . and is not available except in rare and extraordinarily compelling circumstances”). See Appendix B, Table 1. For examples of cases in which government attorneys repackaged claims for settlement under the Judgment Fund, see Part II.B.3 above. We have seen evidence that the Department of Justice actively seeks to conceal, rather than to disclose, the nature of its settlement practices. In response to a magistrate’s request for clarification of settlement authority and indemnification practices, the Department replied by objecting to the order on the ground that it invaded the discretion that the Department of Justice’s indemnity regulations had placed in the hands of the Attorney General. See The United States Department of Justice’s Objection to the Magistrate Judge’s Order Entered on June 22, 2011, at 3-7, Bolden v. Marberry, No. 2:09-cv-00312 (S.D. Ind. Aug. 5, 2011), ECF No. 93. Emphasizing that the Bureau of Prisons was not a party to the litigation, the government’s submission proceeded on footnote continued on next page 606 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) In looking for ways to square the practices we observe in the data with the Department of Justice narrative, we acknowledge some possible inwardfocused explanations for government attorneys’ behavior. First, government agencies and their lawyers may find it easier to secure a settlement through the FTCA and the Judgment Fund than through the indemnification procedures contemplated by current regulations. After all, relevant Department of Justice regulations explicitly disfavor prejudgment indemnification,192 and almost all settlements in our dataset occurred in the absence of a judgment. Second, maintaining a practice in which no employee is formally indemnified may serve a messaging purpose in the rare case in which the Department insists on a contribution from an individual employee. When that happens, the Department does not have to justify to its employees fine distinctions in cases in which it does and does not choose to indemnify under the regulations.193 Outward-focused reasons may also explain the Department’s insistence on maintaining the fiction of individual liability in Bivens cases. Put simply, the Department may find it strategically useful to convince judges that indemnification is not routine. Attorneys representing federal officials are advised from the outset of a case to reinforce to judges, “in both direct and subtle ways,” the significant difference between individual and official capacity suits and the practical consequences to the individual defendant.194 Most relevant to this Article, attorneys for federal defendants have pointed to the risk of personal liability and the uncertainty of indemnification to support various legal positions, including efforts to exclude certain evidence from the formal basis that the settlement of individual liability claims was a matter entirely between the official and the plaintiff. Id. at 5-6. Notably, the submission said nothing about the possible use of the Judgment Fund as a source of pre-trial settlement authority, despite the fact (as we have seen) that the Department often settles Bivens claims before trial by transforming them into claims against the government under the Judgment Fund. 192. 28 C.F.R. § 50.15(c)(1)-(4); U.S. DEP’T OF JUSTICE, supra note 188, § 4-5.412. 193. We note that this is not a concern about court challenges brought by disappointed federal employees denied indemnification. Courts have held that decisions to decline to represent Department of Justice employees (and implicitly to indemnify) are not reviewable. See, e.g., Fishman v. Washington-Adduci, No. 2:13-cv-04729, 2017 WL 3319107, at *4 (C.D. Cal. Aug. 3, 2017) (collecting cases holding that decisions under 28 C.F.R. § 50.15 are unreviewable and within the absolute discretion of the Department of Justice). 194. Mary Hampton Mason, You Mean I Can Be Sued? An Overview of Defending Federal Employees in Individual Capacity Civil Suits, U.S. ATTORNEYS’ BULL., July 2002, at 1, 4. 607 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) trial,195 to stay proceedings,196 to require personal service,197 to argue against personal jurisdiction,198 and, most pointedly, to deny a Bivens remedy altogether.199 195. See, e.g., Molina v. Perez, No. 6:13-cv-01025, 2015 WL 249024, at *4-5 (D. Kan. Jan. 20, 196. 197. 198. 199. 2015) (granting motion in limine on indemnification issue); Defendant DEA Special Agent Robert Cross’s Motion In Limine to Preclude Any Testimony or Reference at Trial to Indemnification or to His Defense Counsel or Defendant Cross as “The Government” at 3, Webb v. United States, No. 1:07-cv-03290 (N.D. Ohio Feb. 13, 2017), ECF No. 263; Notice of Motion in Limine to Exclude Reference to Possible Indemnification of Defendants by U.S. Department of Justice; Memorandum of Points & Authorities at 1, Bennett v. Dhaliwal, No. 2:14-cv-04697 (C.D. Cal. Sept. 18, 2015), ECF No. 57. See, e.g., Defendants’ Memorandum in Opposition to Plaintiff ’s Motion for Leave to Amend & Defendant’s Memorandum in Support of Cross-Motion to Stay Proceedings Pending Appeal at 7, Martin v. Naval Criminal Investigative Serv., No. 3:10-cv-01879 (S.D. Cal. Jan. 30, 2012), ECF No. 54. See, e.g., Cochran v. Barnes, No. 2:08-cv-00358, 2010 WL 455510, at *2 (N.D. Ind. Feb. 1, 2010). In Fiore v. Walden, the Ninth Circuit found personal jurisdiction over the Bivens defendant proper, in part because the U.S. Attorney’s Office was available to defend cases brought against federal officials. 688 F.3d 558, 583-84, 588 (9th Cir. 2012) (noting that the burden on the defendant was slight because he was represented by “the world’s largest law firm” (internal quotation marks omitted)), rev’d, 134 S. Ct. 1115 (2014). In the Supreme Court, Walden and his amici rebutted this by pointing to the personal nature of Bivens litigation and the uncertain scope of indemnification. Brief for Petitioner at 36-37, Walden v. Fiore, 134 S. Ct. 1115 (2014) (No. 12-574), 2013 WL 2390244 (noting that the suit threatened the petitioner’s “personal finances”); Brief for Federal Law Enforcement Officers Ass’n as Amicus Curiae Supporting Petitioner at 11-12, Walden, 134 S. Ct. 1115 (No. 12-574), 2013 WL 2445025 (noting that indemnification is uncertain); Brief for the United States as Amicus Curiae Supporting Petitioner at 19-20, Walden, 134 S. Ct. 115 (No. 12-574), 2013 WL 2445027 (noting that representation and indemnification in Bivens claims are “discretionary”). See, e.g., Brief for the Petitioners at 47, Ashcroft v. Iqbal, 556 U.S. 662 (2008) (No. 071015), 2008 WL 4063957 (relying on the prospect of personal liability to argue against supervisory Bivens claims based on constructive knowledge of wrongdoing); Brief for the Petitioners at 12, Wilkie v. Robbins, 551 U.S. 537 (2007) (No. 06-219), 2007 WL 128587 (referring to “threat of personal liability” from Bivens actions); Schweiker Petitioners’ Reply Brief, supra note 185, at 16 (referring to “the devastating potential liability” facing Bivens defendants); Schweiker Petitioners’ Brief, supra note 184, at 47-48 (arguing against a Bivens remedy because of “ruinous personal liability”); Supplemental Brief for the Appellees at 38-39, Koprowski v. Baker, 822 F.3d 248 (6th Cir. 2016) (No. 14-5451) (arguing against the creation of a Bivens remedy because the “threat of personal liability” would lead to difficulty in recruiting qualified candidates); Arar Replacement Brief for Ashcroft, supra note 185, at 43 (arguing that recognizing a Bivens claim “would put at personal risk the officials involved in making the most sensitive and important decisions facing the nation” (emphasis added)); Defendant Robert Buchan’s Motion to Dismiss & Memorandum in Support Thereof at 8, Engel v. Buchan, No. 1:10-cv-03288 (N.D. Ill. Aug. 16, 2010), ECF No. 39 (arguing against a Bivens remedy because of “over-deterrence” caused by the risk of substantial damages award “for which a government employee has no realistic ability to pay”). 608 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Similar strategic considerations might explain why government attorneys insist on settling some meritorious Bivens claims as FTCA claims, even when no viable FTCA claim has been or can be asserted.200 Seen from a system-wide perspective, such practices may obscure the existence of viable Bivens claims and prop up a commonly shared perception that Bivens claims rarely succeed. Judges make such skeptical claims about the merits, often citing statistics provided by Department of Justice attorneys as support.201 If they rarely succeed, Bivens claims may appear to do little more than burden federal dockets with insubstantial matters. Such perceptions may underscore judicial reluctance to extend the doctrine to new forms of constitutional litigation,202 a reluctance very much present in the Ziglar decision. At a more practical level, settling cases through the FTCA may enable individual defendants to claim in subsequent litigation that, although they have been sued before, they have never paid a judgment or a settlement, obscuring the extent to which the defendant may have engaged in past unconstitutional conduct. Whatever the reasons for the government’s descriptions of the financial threats faced by individual BOP employees, this study reveals those threats to be largely unfounded. To the extent the Department of Justice’s representations about financial risk encourage plaintiffs to settle cases at a discount or influence judicial perceptions of the Bivens doctrine in general or of its application to particular claims, the Bivens remedy may be undermined for reasons that have little relationship to reality. D. Payment Practices, Transparency, and Congressional Oversight of the Judgment Fund The payment practices revealed in our data also pose questions for Congress, as the institution responsible for proper stewardship of the public fisc and for oversight of payments from the Judgment Fund. Congress has not accepted liability on behalf of the government for constitutional torts committed by federal employees. The FTCA applies to garden-variety torts defined as such in the law of the state where the injury occurred, but does not 200. See supra Part II.B.3. Notably, the right to trial by jury on a Bivens claim may encourage government attorneys to discuss settlement. Given the Department’s indemnity policy, which prohibits the pre-verdict indemnification of personal capacity defendants, the FTCA provides the only available source of funds for pretrial resolution of the claims. 201. One of us has written extensively about this attitude. See Reinert, supra note 9, at 82728 & nn.94-99. 202. As one example, see Crawford-El v. Britton, 93 F.3d 813, 838 (D.C. Cir. 1996) (en banc) (Silberman, J., concurring) (citing Department of Justice statistics regarding the success of Bivens claims and concluding that “[o]bviously, the vast majority of these suits are meritless”), vacated, 523 U.S. 574 (1998). 609 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) apply to claims based on the Constitution.203 Yet we find that, in cases where the plaintiff brought meritorious claims implicating the Bivens doctrine, government attorneys regularly arranged to have those claims resolved— explicitly or informally—as claims brought under the FTCA. As Part II demonstrates, cases in our dataset were resolved as FTCA claims even when there were clear barriers to proceeding with FTCA settlements.204 The Supreme Court held in 1993 that a claimant’s failure to exhaust administrative remedies will bar an FTCA claim.205 Along similar lines, courts have held that the failure to file an administrative claim with the appropriate agency bars a federal court from exercising subject matter jurisdiction over an FTCA claim.206 Yet in at least thirteen Bivens cases in our dataset, plaintiffs 203. While the FTCA provides for the imposition of vicarious tort liability on the federal government for the torts of its officers and employees, the FTCA makes no provision for the assertion of claims under federal constitutional or statutory law. See FDIC v. Meyer, 510 U.S. 471, 478 (1994) (noting that, under the FTCA, state law provides the source of substantive liability and explaining that the FTCA does not apply where federal law provides the rule of decision). Indeed, when Congress expanded the FTCA to accept liability for intentional torts committed by law enforcement officers in 1974, and immunized federal officials for common law torts in 1988, it acted to preserve rather than displace the Bivens suit for constitutional tort claims. See Pfander & Baltmanis, supra note 10, at 132-38. 204. See supra notes 120-37 and accompanying text 205. McNeil v. United States, 508 U.S. 106, 113 (1993). The FTCA framework at issue in McNeil requires exhaustion through submission to an agency and then requires claimants to file in court within six months of the conclusion of the administrative process. See 28 U.S.C. § 2401(b) (2018) (barring suit unless filed within six months of the agency’s denial); id. § 2675(a) (requiring exhaustion). For doubts as to the jurisdictionality of the litigation bar in McNeil, see SISK, supra note 10, at 112-13. Whether McNeil ’s holding is better understood as making exhaustion jurisdictional or treating it as a claimprocessing rule, lower courts have generally treated McNeil as a holding bearing on subject matter jurisdiction. See, e.g., George v. E. Orange Hous. Auth., 687 F. App’x 122, 124 (3d Cir. 2017) (per curiam) (stating that a “post-suit attempt to pursue administrative remedies did not give the District Court subject matter jurisdiction to hear [plaintiff’s] FTCA claim”); Mader v. United States, 654 F.3d 794, 807 (8th Cir. 2011) (en banc) (acknowledging that the McNeil Court did not use the word “jurisdiction,” but reading the exhaustion requirement as necessary to invoke district court jurisdiction); Rasul v. Myers, 563 F.3d 527, 528 n.1 (D.C. Cir. 2009) (per curiam) (treating the failure to exhaust as a jurisdictional bar after McNeil); Turner ex rel. Turner v. United States, 514 F.3d 1194, 1202 n.5 (11th Cir. 2008) (treating McNeil as a jurisdictional holding); Siemientkowski v. Moreland Homes, Inc., 25 F. App’x 415, 416 (6th Cir. 2002) (same); Brady v. United States, 211 F.3d 499, 502 (9th Cir. 2000) (same); Duplan v. Harper, 188 F.3d 1195, 1199 (10th Cir. 1999) (same); cf. Acosta v. U.S. Marshals Serv., 445 F.3d 509, 513 (1st Cir. 2006) (citing First Circuit precedent for the same proposition); Kokotis v. U.S. Postal Serv., 223 F.3d 275, 278-79 (4th Cir. 2000) (requiring exhaustion for district court jurisdiction). But see Glade ex rel. Lundskow v. United States, 692 F.3d 718, 723 (7th Cir. 2012) (treating the requirement as a claims-processing rule). 206. Roma v. United States, 344 F.3d 352, 362 (3d Cir. 2003) (“[T]he requirement that the appropriate federal agency act on a claim before suit can be brought is jurisdictional and cannot be waived.”); Millares Guiraldes de Tineo v. United States, 137 F.3d 715, 720 footnote continued on next page 610 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) amended their complaints to add FTCA claims without any indication that they had administratively exhausted their claims.207 In another nineteen cases in our dataset, settlements in Bivens cases were styled as made under the FTCA even though the plaintiff had never alleged an FTCA claim and there was no indication that an administrative claim ever was filed with any agency, let alone with the BOP.208 And in at least five Bivens cases, the settlements were styled as made under the FTCA even though the plaintiff’s FTCA claim had previously been dismissed.209 Additionally, in as many as thirty-two of the cases in which evidence of administrative exhaustion was lacking, plaintiffs also appear to have amended their Bivens case to add an FTCA claim—or settled their Bivens claim as an FTCA claim—outside the applicable statutes of limitations.210 The Supreme 207. 208. 209. 210. (2d Cir. 1998) (stating that the claim must have been presented to the appropriate agency in writing and for a sum certain). This exhaustion requirement “has been viewed as ‘a non-waivable jurisdictional requirement’ limiting the suit to claims fairly made to the agency.” Acosta, 445 F.3d at 513 (quoting Santiago-Ramirez v. Sec’y of Dep’t of Def., 984 F.2d 16, 18, 19-20 (1993)); see also Brooks v. Silva, No. 13-6539, 2015 WL 12762112, at *2 (6th Cir. Apr. 6, 2015) (relying on First and Fourth Circuit cases to conclude that exhaustion was jurisdictional); Ali v. Rumsfeld, 649 F.3d 762, 775 (D.C. Cir. 2011) (providing that failure to exhaust is jurisdictional); Daniels v. United States, 135 F. App’x 900, 901 (8th Cir. 2005) (per curiam) (providing that presentment of a claim to an agency is a jurisdictional prerequisite); Joelson v. United States, 86 F.3d 1413, 1422 (6th Cir. 1996) (“Because Joelson does not allege that he has filed an administrative claim, he has not satisfied the jurisdictional prerequisite to obtaining judicial review under the Federal Tort Claims Act, and the district court properly dismissed this claim.”). Of all of the circuits, only the Seventh Circuit has questioned the near-unanimous view that satisfaction of the FTCA’s exhaustion requirement is jurisdictional. See Lundskow, 692 F.3d at 723 (holding that the exhaustion requirement is not jurisdictional and may be waived). Numerous courts of appeals have held that the exhaustion requirement cannot be waived. See, e.g., D.L. ex rel. Junio v. Vassilev, 858 F.3d 1242, 1244 (9th Cir. 2017); Estate of Cummings v. United States, 651 F. App’x 822, 828 (10th Cir. 2016) (distinguishing between time of filing requirements, which after United States v. Kwai Fun Wong, 135 S. Ct. 1625 (2015), may be waived, and the exhaustion requirement itself, which cannot); Shelton v. Bledsoe, 775 F.3d 554, 569 (3d Cir. 2015); Acosta, 445 F.3d at 513; Celestine v. Mount Vernon Neighborhood Health Ctr., 403 F.3d 76, 82 (2d Cir. 2005) (noting that the FTCA exhaustion requirement “is jurisdictional and cannot be waived”). See supra notes 120-33 and accompanying text. See supra notes 135-37 and accompanying text. See supra notes 127-33 and accompanying text. The FTCA has a two-year statute of limitations for presenting a claim to the relevant agency, and a six-month deadline for filing in federal court after rejection of a claim filed with the agency. See 28 U.S.C. § 2401(b) (barring suit unless the claim is presented to the relevant agency within two years after accrual and the suit is filed within six months of the agency’s denial). These limitations are subject to equitable tolling, and in at least one case in our dataset the court allowed equitable tolling for the plaintiff ’s FTCA claim. See Hoslett v. Dhaliwal, No. 3:11-cv-00674, 2013 WL 5947253, at *3 (D. Or. footnote continued on next page 611 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Court recently held that the FTCA’s statutes of limitations were not jurisdictional and were subject to equitable tolling (and therefore were presumably subject to waiver by the government).211 Before 2015, however, the courts of appeals were divided with some holding these time limitations were jurisdictional and therefore not subject to waiver.212 Some lower courts in our dataset nonetheless routinely accepted complaints that were filed without satisfying FTCA’s exhaustion requirement in a timely fashion (or at all). Some such filings occurred in districts at a time when controlling circuit authority held that the plaintiff’s failure to satisfy these preconditions to suit deprived the district court of subject matter jurisdiction.213 To summarize, in almost 19% of the Bivens cases in our dataset, FTCA claims were added during the course of litigation and settlement without any finding that exhaustion and/or statute of limitations requirements had been satisfied.214 We can readily imagine why plaintiffs would be willing to overlook these deficiencies—presumably they are eager to be compensated and are less particular about the pot of government money from which the payment comes. But we find it more difficult to understand why attorneys for the government would arrange,215 and why courts would agree to approve, settlements in which the parties have amended their lawsuits to invoke the FTCA and to thereby trigger the payment of Bivens claims through the Judgment Fund. 211. 212. 213. 214. 215. Nov. 6, 2013). It appears that these statute of limitations requirements were not followed in the cases described above in notes 120-33. Kwai Fun Wong, 135 S. Ct. at 1638. Compare, e.g., Alexander v. United States, 646 F.3d 185, 190-91 (5th Cir. 2011) (per curiam) (stating that equitable tolling is not available), abrogated by Kwai Fun Wong, 135 S. Ct. 1638, and Leonhard v. United States, 633 F.2d 599, 624 (2d Cir. 1980) (stating that courts lack jurisdiction where claims were presented to an agency more than two years after accrual), with, e.g., Arteaga v. United States, 711 F.3d 828, 832-33 (7th Cir. 2013) (stating that equitable tolling is allowed). For example, the Third Circuit permitted equitable tolling of the FTCA’s limitations period in extremely narrow circumstances, see Santos ex rel. Beato v. United States, 559 F.3d 189, 197 (3d Cir. 2009), but several cases were settled as FTCA claims in the Third Circuit that would not have met these requirements, see Williams v. Warmerdorf, No. 3:07-cv-01283 (M.D. Pa. 2007); supra notes 121-23 and accompanying text (discussing Johnson v. Martinez, No. 2:04-cv-01967 (E.D. Pa.)). The Tenth Circuit did not allow any equitable tolling and treated the FTCA’s time limitations as jurisdictional, see Barnes v. United States, 776 F.3d 1134, 1148 (10th Cir. 2015), yet cases in our dataset were settled as FTCA claims in the circuit even though they did not satisfy these jurisdictional requirements, see supra notes 124-26 and accompanying text (discussing Stine v. Allred, No. 1:11-cv-00109 (D. Colo.)); supra note 110 and accompanying text (discussing Shannon v. Federal Bureau of Prisons, No. 1:03-cv-00352 (D. Colo.)). See supra notes 120-37 and accompanying text. For discussion of the Department of Justice and BOP attorneys’ possible thought processes in these matters, see Part III.C above. 612 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Satisfying Bivens claims through Judgment Fund payments—regardless of whether they are restyled as FTCA claims—also appears contrary to the congressionally created remedial regime. Under existing law, no statute authorizes payment of Bivens claims against individual officers from the Judgment Fund. If the Bivens claim results in an assessment of liability or a settlement, there are two options contemplated by existing law: either the individual officer pays personally (perhaps with the assistance of liability insurance) or the employee seeks indemnification directly from the agency for which she works.216 Each agency has different regulations regarding indemnification—the Department of Justice (where the BOP is found) provides for indemnification of a settlement or judgment when the conduct is within the scope of the officer’s employment and when indemnification is “in the interest of the United States.”217 If an agency agrees to indemnify the employee, then the funds are paid from the agency’s budget, not from the Judgment Fund.218 But in none of the settlements that we document here is there any indication that the BOP paid settlement funds out of its own budget.219 Members of Congress and scholars alike have worried that some executive branch agencies shift costs to the Judgment Fund that might more properly be paid from the agency’s own appropriation.220 For example, Congress took a dim view of the efforts of agencies to arrange for the Judgment Fund to pay the costs associated with an award of attorney’s fees under the Equal Access to Justice Act (EAJA).221 Such fees are payable when agencies are adjudged to have taken indefensible positions in litigation; Congress felt that the agencies should pay any such fees to force them to internalize the costs associated with their litigation posture.222 Congress might similarly worry that agencies with access 216. See SISK, supra note 10, at 398-99. 217. 28 C.F.R. § 50.15(c)(1), (2) (2019). 218. See 31 U.S.C. § 1304(a)(3) (2018) (appropriating funds to pay judgments, awards, and 219. 220. 221. 222. settlements under a series of specified statutes and omitting any reference to Bivens claims). It may be that some amount of money was paid through the Judgment Fund and then repaid by the BOP. See Figley, supra note 17, at 171-73. But there is no indication in the agreements we were provided that this is occurring. See id. at 167-75; see also VIVIAN S. CHU & BRIAN T. YEH, CONG. RESEARCH SERV., R42835, THE JUDGMENT FUND: HISTORY, ADMINISTRATION, AND COMMON USAGE 12-15 (2013), https://perma.cc/W87R-ABEG (recounting recent legislation that would improve Judgment Fund transparency and provide for agency reimbursement of the Fund in appropriate situations). Congress has been considering legislation that would require the Department of the Treasury to extend its practice of posting Judgment Fund payment information on a publicly accessible website. See Judgment Fund Transparency Act of 2017, H.R. 1096, 115th Cong. (2017). See Figley, supra note 17, at 171-75. See id. 613 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) to Judgment Fund payments may take too few steps to lessen the magnitude of Bivens liability. Indeed, the government has argued against extending Bivens to suits against federal agencies in part because the relevant statutory scheme does not authorize expenditures of federal funds to satisfy Bivens claims.223 Congress might sensibly share this concern. The Bureau of Prisons’ response to our FOIA request makes clear that the agency collects only piecemeal information about FTCA and Bivens cases alleging wrongdoing by their employees.224 Without complete information about these cases— including the allegations in the cases and the amount paid—the Bureau of Prisons cannot make informed decisions about how to prevent similar constitutional violations in the future.225 Perhaps if the BOP or its employees were financially responsible in more of these cases, they would collect more comprehensive information and take more care in the future.226 The need for more comprehensive information-gathering brings us to a larger concern with government transparency. The payment practices revealed in our data might seem surprising given the narrative of personal liability that has dominated Bivens literature and jurisprudence. That sense of 223. See Brief for the Federal Deposit Insurance Co. at 25-26, 26 n.20, FDIC v. Meyer, 510 U.S. 471 (1993) (No. 92-741), 1993 WL 348895 (citing 31 U.S.C. § 1304). 224. See supra Part II.C. 225. See generally Schwartz, supra note 147 (arguing that law enforcement agencies that gather and examine data about lawsuits act to deter misconduct, but not studying the Bureau of Prisons specifically). One might usefully contrast reporting on the incidence of Bivens litigation with the elaborate reporting and reimbursement obligations that Congress imposed on federal agencies in the so-called “No FEAR Act,” the Notification and Federal Employee Antidiscrimination and Retaliation Act of 2002, Pub. L. No. 107174, 116 Stat. 566 (codified at 25 U.S.C. § 2301 note at 205-08 (2018)). The Act requires agencies to file annual reports on workplace discrimination of all sorts and to reimburse the Judgment Fund when whistleblower litigation results in judgments payable by the United States. Id. §§ 201, 203, 116 Stat. at 569-70. For an account of the Act’s provisions, see Sarah Wood Borak, Comment, The Legacy of “Deep Throat”: The Disclosure Process of the Whistleblower Protection Act Amendments of 1994 and the No FEAR Act of 2002, 59 U. MIAMI L. REV. 617, 650-57 (2005). For a discussion of other statutes that resemble the reimbursement features of the No FEAR Act, see Stern, supra note 177, at 656-57 (describing the Contract Disputes Act, the Stored Communications Act, and the Foreign Intelligence Surveillance Act as including provisions for agency reimbursement of the Judgment Fund). 226. See Schwartz, supra note 177 at 1195-96 (arguing that increased financial effects of lawsuits on agencies would encourage accountability within the agencies). We find it ironic, in light of the practices disclosed, that the Supreme Court has repeatedly invoked the need for deference to Congress as a basis for refusing to vindicate constitutional rights in Bivens-type actions. In many of the settlements we document in this study, the government has seemingly paid little heed to congressional expectations about the relationship between Bivens, FTCA, and the Judgment Fund. See supra Part III.B; see also Reinert & Mulligan, supra note 34, at 1501-04 (arguing that the remedial framework governing § 1983 actions and the FTCA demonstrates that Bivens claims should be allowed against both private and public employees). 614 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) surprise no doubt reflects a failure on the part of responsible government institutions to report on the nature and extent of Bivens liability. We know of no report that informs the public of the genuine success rates for Bivens litigation, the frequency with which particular agencies commit actionable constitutional violations, the amount of and sources from which agencies draw the money needed to pay off meritorious claims, and the steps agencies have taken to lessen the likelihood of constitutional violations in the future. As a result, both Congress and the courts lack the information they need to evaluate the efficacy of constitutional tort litigation. E. On the Need for Future Research and the Uncertain Future of the Bivens Action Partly due to these transparency problems and partly due to the narrow collection of litigation files we obtained through our FOIA requests, we cannot provide a complete picture of Bivens payment practices. To be sure, BOP employees are responsible for the lion’s share of Bivens claims against federal government actors.227 But because our data come from a single agency within the Department of Justice, they limit our ability to speak more broadly. We know that other agencies within the Department defend Bivens claims and suspect that they follow the practice we identify here of repackaging the viable claims for settlement and payment under the FTCA.228 Still, our data do not shed definitive light on those practices. Similarly, while we know that officials working in agencies of the federal government outside the Department of Justice have been named in important Bivens actions, we have yet to secure any information about the payment and indemnification practices within those agencies.229 In some instances, the officers sued in such proceedings do not qualify as investigative and law enforcement officers within the meaning of the FTCA.230 That, in turn, may 227. Reinert, supra note 9, at 837 tbl.2 (reporting that prison claims made up almost half of all Bivens filings in a five-district study conducted from 2001 to 2003). 228. The Judgment Fund website lists a series of payments made on behalf of other law enforcement agencies within the Department of Justice, such as the FBI, the Bureau of Alcohol, Tobacco, and Firearms, and the DEA. Payments from the Judgment Fund can be searched by visiting https://jfund.fiscal.treasury.gov/jfradSearchWeb/ JFPymtSearchAction.do and entering a date range and federal agency. 229. See, e.g., Wilkie v. Robbins, 551 U.S. 537, 541 (2007) (naming officials in the Bureau of Land Management); Harlow v. Fitzgerald, 457 U.S. 800, 802 (1982) (naming White House aides); Butz v. Economou, 438 U.S. 478, 480-82 (1978) (naming officials in the Department of Agriculture). 230. None of the defendants named in the Bivens suits in note 229 above would qualify as law enforcement officers, as defined in 28 U.S.C. § 2680(h) (2018). See supra note 42. But agencies housed in departments outside the Department of Justice do employ a number of officers who meet that definition. See Pellegrino v. U.S. TSA, 937 F.3d 164, 168 (3d footnote continued on next page 615 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) make it far more difficult for the agencies in question to restyle claims for payment through the Judgment Fund. We encourage additional research into payment practices as policymakers reflect on how best to structure a complex system of government accountability. Along with our call for more research, we caution against drawing overly broad policy conclusions from the data collected here. Some might argue, based on the payment practices described in Part II, that Bivens liability no longer plays a distinctive role in our system of accountability, having been displaced by settlements and payments that almost always occur within the context of the FTCA. Building on that conclusion, some might encourage the Court to overturn Bivens, arguing that doing so would pose no grave threat to constitutional remedies. Such an argument might assume that, so long as the FTCA remained in place, it would assure adequate remediation for any wrongdoing on the part of federal officials. Another version of the argument might emphasize the degree to which common law tort remedies substitute for viable Bivens claims; so long as the common law framework remains intact, Bivens may not contribute much on the margins to a system of government accountability. Yet these arguments closely resemble those that the government made, and the Court rejected, in Carlson v. Green, itself a case involving employees of the BOP.231 We think the reasons Carlson identified in rejecting the argument that tort liability under the FTCA displaced the Bivens remedy remain sound today: Constitutional violations deserve special remedial attention, and common law tort claims brought against the government cannot alone adequately ensure the protection of constitutional values.232 Moreover, it is worth noting that the data reflect litigation practice after a Bivens claim has been resolved—in that light, they may reveal less about the value of Bivens claims as a mode of accountability than about the strategic decisions of the parties to such litigation. Our data support additional bases to conclude the Bivens doctrine plays an important role. First and foremost, nothing in our data undercuts the conclusion that Congress has explicitly designed the FTCA to provide a remedy that supplements, but does not displace, Bivens actions. One should not confuse the observation that the government has settled cases under the FTCA Cir. 2019) (en banc) (TSA screeners); Nurse v. United States, 226 F.3d 996, 1002-03 (9th Cir. 2000) (customs agents); Celestine v. United States, 841 F.2d 851, 852-53 (8th Cir. 1988) (per curiam) (VA hospital security guards); Caban v. United States, 671 F.2d 1230, 1234 (2d Cir. 1982) (INS agents). 231. See supra notes 45-50 and accompanying text (describing Carlson, 446 U.S. 14 (1980)). 232. Cf. Fallon, supra note 150, at 988-89 (reading the Westfall Act as contemplating the continued viability of Bivens claims, notwithstanding the existence of a remedy under the FTCA). 616 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) for its own convenience with the conclusion that, absent the Bivens doctrine, the FTCA would have provided the same remedy or any remedy at all. Even within the Bureau of Prisons, our data suggest that Bivens continues to do some independent work in fostering the assertion and settlement of viable claims; it is otherwise difficult to explain why the United States would agree to settle cases through the FTCA (despite substantial barriers to recovery) in exchange for the plaintiff’s agreement to dismiss a Bivens claim. And we do not know how the litigation of these cases was influenced, if at all, by the federal government’s stated position that it rarely if ever provides indemnification for individual officers in Bivens litigation.233 What’s more, our understanding of the Department of Justice’s approach to the practice of indemnification suggests that the Department itself maintains an institutional interest in preserving the prospect of personal liability (however theoretical), perhaps as a way to encourage its employees to respect constitutional boundaries.234 Instead of offering a promise or assurance of indemnity, the Department of Justice offers employees only the opportunity to petition for indemnification after an adverse verdict or award has been entered. That enables government attorneys to retain negotiating leverage with employee defendants who have engaged in particularly egregious forms of misconduct: Assistant United States Attorneys might plausibly demand a monetary contribution from the employee during settlement negotiations by threatening to deny indemnity in the event a Bivens claim were to proceed to trial.235 We raise these as theoretical possibilities to explain why the Department might be invested in maintaining the myth of individual liability—our data cannot evaluate the extent to which individual officers adjust their behavior based on their belief that indemnification is not guaranteed. That the Department of Justice might see value in preserving a (modest) threat of Bivens personal liability through its rule of delayed and uncertain indemnification tells us much about the relative competence of courts and executive agencies in determining where the ultimate burden of government 233. See 28 C.F.R. § 50.15(c)(1)-(4) (2019) (permitting indemnification upon request, but stating that “[a]bsent exceptional circumstances,” the Department of Justice will not “entertain a request either to agree to indemnify or to settle a personal damages claim before entry of an adverse verdict, judgment, or award”); U.S. DEP’T OF JUSTICE, supra note 188, § 4-5.412 (noting that “there is no right to compel indemnification” and that “[p]rejudgment indemnification is disfavored . . . and is not available except in rare and extraordinarily compelling circumstances”). We note that in all of the cases we reviewed in which a Bivens claim remained in the case, the plaintiffs prevailed through settlements and the settlements occurred in the absence of an adverse judgment. 234. Economic theory predicts that employers and employees will negotiate over the incidence of liability for on-the-job torts. See Sykes, supra note 175, at 565-66. 235. Although we cannot know for sure, such threats may have played a role in officers’ contributions to settlements in our dataset. See supra Part II.A. 617 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) liability should fall. For much of the nineteenth century, federal courts took no part in such calculations; the judicial task was limited to the determination as to whether the official defendant had violated the law and, if so, what sort of recompense was appropriate.236 Courts eschewed the business of protecting government officials, however well-meaning, from financial liability for their tort-based wrongs and refused to adopt doctrines of qualified immunity.237 That left government officials personally accountable, but with networks of supportive superior officers and coworkers who could help them navigate the process of seeking indemnification.238 A similar internal process may accompany the resolution of substantial Bivens claims today. Importantly, though, that internal process has been deliberately structured by the Department of Justice, perhaps as a tool of internal employee control and supervision for use in only the most egregious cases.239 One can understand why the Department, having for its own administrative reasons retained an incomplete assurance of indemnification, would turn around and highlight the threat of personal liability as a rhetorical justification for narrowing access to Bivens-type remedies. But one has greater difficulty in understanding why the federal courts would continue to attend to such rhetorical claims given the practices we have identified here. Apart from the distinctive contributions that Bivens claims make to the settlement process within the Department of Justice, a decision to abandon Bivens would leave many government agencies outside the Department subject to little prospect of liability for their intentional wrongs. As we have seen, the FTCA limits the government’s liability to intentional torts committed by 236. For an account of the nineteenth-century model of personal liability and indemnity through the adoption of a private bill in Congress, see generally Pfander & Hunt, supra note 9. 237. See, e.g., Mitchell v. Harmony, 54 U.S. (13 How.) 115, 135-37 (1851) (imposing personal liability on an army officer, despite the Court’s recognition that the officer had played an important role in a military operation that was “boldly planned and gallantly executed,” and confirming that courts had no business fashioning rules to protect officials from liability); Little v. Barreme, 6 U.S. (2 Cranch) 170, 178-79 (1804) (refusing to protect a naval officer from personal liability, despite the officer’s good faith effort to carry out the orders of his superior officer). 238. For an account of the process of indemnity and the role of the committee on claims, see Pfander & Hunt, supra note 9, at 1890-93. 239. As we observed earlier in Part II.A, the conduct in cases in which employees contributed personal resources to settle viable Bivens claims, though constitutionally significant, was not necessarily more culpable than that in cases in which employees were held harmless. Perhaps the decision to insist on employee financial contributions reflects less the severity of the wrong than the supervisor’s perception of the need to discipline a repeat offender through the imposition of what operates in effect as a monetary sanction. In other words, internal BOP personnel practices may inform the decision whether to hold a particular employee harmless or to insist on personal payments. 618 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) investigative and law enforcement officials; many of the officers in other agencies fail to qualify as such.240 Similarly, many law enforcement officers work either along the border or outside the United States; any extraterritorial injury they inflict would fall outside the FTCA’s coverage.241 Finally, liability for the torture or unlawful detention of people held in military confinement might well fall within the FTCA exclusion for the combatant activities of the armed forces.242 It would make very little sense to argue from our data that the Court should deny access to a Bivens action in those areas of government activity where the FTCA does not apply and where official liability for constitutional torts likely serves as the only mode of redress for the victims of wrongful conduct. Yet in its most recent Bivens decision, the Supreme Court took precisely that approach. In Hernandez v. Mesa, the Court refused to recognize a Bivens remedy for the cross-border shooting of an unarmed Mexican teenager by a Border Patrol officer.243 Had the death occurred in the United States, it would have fallen within the original search-and-seizure Bivens context. But because the teenager was across the border in Mexico, this new context led the Court to apply its restrictive special factors analysis in assessing the claim’s viability.244 The Court acknowledged that the FTCA and other potentially applicable remedies did not apply to the claim in question, which arose from injuries sustained outside the formal boundaries of the United States and thus triggered the FTCA’s foreign country exception.245 But rather than treating the FTCA’s inapplicability as a gap in the system of remedies that a Bivens action might well help to fill, the Court viewed congressional silence as part of a consistent refusal to “authorize the award of damages for injury inflicted outside our borders.”246 240. For an account of the virtual immunity of the federal government and its officers from 241. 242. 243. 244. 245. 246. tort or other liability in cases of assault and battery that arise outside the law enforcement context, see SISK, supra note 10, at 165-66, 170-72 (describing the intentional tort exception and the narrow law-enforcement proviso to that exception). See 28 U.S.C. § 2680(k) (2018). For the FTCA exclusion for “combatant activities of the military or naval forces,” see 28 U.S.C. § 2680(j). Courts have adopted a relatively broad interpretation of the term combatant activities. See, e.g., In re KBR, Inc., Burn Pit Litig., 744 F.3d 326, 351 (4th Cir. 2014) (concluding that a contractor who provided water treatment and trash disposal at military bases in Iraq and Afghanistan was engaged in combatant activities). 140 S. Ct. 735, 739 (2020). Id. at 743-47. Id. at 748 & n.10; see also 28 U.S.C. § 2680(k); Sosa v. Alvarez-Machain, 542 U.S. 692, 712 (2004) (“[T]he FTCA’s foreign country exception bars all claims based on any injury suffered in a foreign country . . . .”). Hernandez, 140 S. Ct. at 747. 619 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Similar conclusions would seemingly apply to cases arising out of the war on terror in which egregious misconduct has been documented on the part of U.S. officials. In Arar v. Ashcroft, for example, U.S. officials subjected a Canadian citizen to extraordinary rendition and torture based on suspect (at best) evidence of complicity in terrorist activity.247 The Second Circuit refused to recognize a Bivens claim, because it presented a new Bivens context and special factors counseled hesitation.248 Nor could the FTCA provide a remedy (again due to the foreign country exception). Other courts agree that the overseas mistreatment of U.S. citizens at the hands of federal officers fails to give rise to a viable Bivens claim.249 In short, we worry about both the FTCA’s displacement of Bivens-style deterrence in the areas where the remedies overlap and about the absence of any remedy at all in new contexts that lie outside the FTCA. In areas that the FTCA does not cover, a Bivens claim will likely provide the only viable remedy, but it will be these areas that will be most likely to flunk Ziglar’s new context test and have to overcome the significant hostility that currently exists towards extending Bivens claims into new areas (hostility supported in part by the fiction of the meritless Bivens claim that is undermined by the settlement patterns we report here).250 For claimants in this category of actions, it truly will be Bivens or nothing.251 Yet in Ziglar and Hernandez, the Court seems to have substantially discounted the adequacy of alternative remedies as a factor in the recognition of new Bivens remedies. We add one final note to highlight an important issue of system design that does not necessarily turn on ultimate incidence of Bivens liability. We discuss here only the comparatively narrow question of who pays when victims of government wrongdoing allege Bivens claims. While our findings have important implications for the design of constitutional remedies, they do not exhaust the factors that inform remedial and system design choices. Even if remedies were to run entirely against the government, holding individuals harmless, Bivens litigation could be justified as a way to bolster the lawannouncing capacity of the federal courts.252 For well understood reasons, 247. 585 F.3d 559, 563-66 (2d Cir. 2009) (en banc). 248. Id. at 574-81. 249. See Pfander, supra note 72, at 758-61 (discussing such cases as Vance v. Rumsfeld, 701 F.3d 193 (7th Cir. 2012), and Doe v. Rumsfeld, 683 F.3d 390, 396 (D.C. Cir. 2012), both of which involved the overseas mistreatment of U.S. citizens at the hands of military officers). 250. For a discussion of this dynamic in one prominent case, see notes 243-46 above. 251. See Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S. 388, 410 (1971) (Harlan, J., concurring in the judgment). 252. On the importance of enhancing the ability of the federal courts to give voice to constitutional norms in the context of Bivens-style litigation, see PFANDER, supra note 51, at 57-69 (discussing the failure of the federal courts to assess the constitutionality of footnote continued on next page 620 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) federal courts lack the power to address many questions of constitutional law in the context of suits for declaratory and injunctive relief. Such law-saying disabilities seem particularly acute in the context of litigation seeking redress for overseas constitutional torts, including extraordinary rendition, extended detention, and cruel and inhumane forms of interrogation.253 Conclusion Much has changed doctrinally since the Supreme Court first recognized that individual federal officers were personally liable for their constitutional torts. While initially welcoming the deterrent effects of such a threat of personal liability, the Court has lately expressed greater concern with the problem of overdeterrence. That concern led first to the judicial creation of immunity doctrines and then to the judicial reluctance to allow victims to seek redress for constitutional violations. In Ziglar v. Abbasi, the Court prominently featured the threat of personal liability and a related concern with the burden of indemnification in its sharply restrictive approach to the recognition of a federal right to sue. Openly criticizing its earlier decisions as the product of a benighted “ancien regime,”254 the Court now views the threat of new Bivens liability as raising concerns of fiscal management that properly call into play the judgment of the political (rather than the judicial) branches. In the course of taking this doctrinal turn, the Court has displayed little interest in knowing who actually pays when constitutional tort claims succeed. Indeed, the Court has been content to assume that the burden of Bivens liability falls primarily on individual officers and their employing agencies. The Department of Justice has actively encouraged those static assumptions, adopting a restrictive indemnity policy that preserves a nominal threat of personal exposure and enables Department attorneys to trumpet the threat of personal liability in opposing new Bivens claims. Nor has the Court displayed any interest in ascertaining the magnitude of Bivens liability and the nature of the threat it poses to fiscal stability. Testing them in this study of successful litigation against the BOP, we find that the Court’s assumptions about the government’s payment practices do not hold. Congressional recognition of intentional tort liability under the FTCA has enabled the Department of Justice to provide individual employees and such war-on-terror claims as those for extraordinary rendition and enhanced interrogation directed at U.S. citizens and others). 253. See id. at 61-69 (describing the failure of the federal courts to clarify the law governing such war-on-terror issues as torture, extraordinary rendition, and military detention of U.S. citizens). 254. Ziglar v. Abbasi, 137 S. Ct. 1843, 1855 (2017) (quoting Alexander v. Sandoval, 532 U.S. 275, 287 (2001)). 621 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) responsible agencies a virtually airtight assurance against any constitutional tort liability. The Department has done so by treating viable Bivens claims as payable under the FTCA and the Judgment Fund. Our study reveals that more than 95% of the successful Bivens claims brought against employees of the Bureau of Prisons over a ten-year period were resolved in such a manner. The BOP itself, apparently, was never called upon to pay successful claims (although internal Department of Justice protocols might conceivably lead to other modes of agency accountability). The Court’s hostility to Bivens litigation rests on the perception that individual officers pay successful claims, a practice that may have been common during the “ancien regime” when Bivens was decided, but one that has been superseded by the Judgment Fund payment practices we document here. Aside from casting doubt on bedrock judicial assumptions about the incidence of liability, our study has important implications for the future of constitutional tort litigation. The data reported here call into serious doubt the Court’s reliance on an almost nonexistent threat of personal liability as a justification for constricting the scope of Bivens remedies. This study similarly undermines the Court’s associated reluctance to extend Bivens remedies to new categories, where the need for constitutional redress may be especially acute. Finally, this Article calls into question whether the sums involved in the cases we have reviewed pose a threat to fiscal stability sufficiently grave to justify the Court’s apparent judgment that extending Bivens liability will henceforth almost always require explicit congressional approval. Awards of the kind reflected in our data can provide important redress to the victims of constitutional violations without posing anything more than a negligible threat to the public fisc. 622 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Appendix A: Methodology This Appendix provides additional information about our study methodology. We first submitted a Freedom of Information Act (FOIA) request to the Federal Bureau of Prisons (BOP), seeking records reflecting all lawsuits brought against the BOP and/or its officers and agents that resulted in a settlement or plaintiff’s judgment between January 1, 2009 and December 31, 2014, the amount of the payments in each case, and information about whether individual federal employees or the BOP contributed to any of these settlements and judgments.255 The BOP responded to our FOIA request by providing what it described as “the litigation records” of cases that resulted in a payment of $1000 or more over a ten-year period.256 Included in their first production of records were 9277 pages, 7796 of which they released to us with minimal or no redaction. After subsequent discussions, we were provided with an additional 920 pages of litigation documents, 856 pages of which had minimal redactions.257 The BOP provided us with 216 separate case files, but we later concluded—with agreement by the BOP—that some of those files were not responsive to our request. By our count, the BOP provided us with files regarding 209 cases with payments of $1000 or more to plaintiffs that were resolved between 2007 and 2017.258 The BOP’s response, while helpful, was incomplete.259 While the BOP provided us with documents regarding each of these 209 cases, the records did 255. We submitted similar requests to the FBI and the Department of Justice, but they did not agree to provide us with information in response to our requests. 256. Letter from Ian M. Guy, Supervisory Attorney, U.S. Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Apr. 19, 2017) (on file with authors). The parameters of their production were different from the parameters of our request because these records had been compiled and produced in response to another, similar FOIA request. See Email from Ronald L. Rodgers, Senior Counsel, Information & Remedies Processing Branch, Office of Gen. Counsel, Fed. Bureau of Prisons, to Joanna C. Schwartz, Professor of Law, UCLA Sch. of Law (Oct. 22, 2018, 11:12 AM) (on file with authors). 257. In this subsequent production, the BOP additionally provided us with files of “tort” and “labor” cases that were resolved without litigation. Because these claims were resolved administratively, and without naming individual officers, we have not included them in this study. 258. In the course of our research, we found another 31 cases that relate in some way to the 216 case files produced by the BOP. When two cases were treated by the court and parties as related and resolved through one disposition, we coded them as a single case. 259. When we followed up to ask about these gaps in the case files, our contact at the Bureau of Prisons explained that BOP attorneys do not represent the agency in litigation or even in negotiating a settlement of a claim—such representation is provided by the Civil Division of the appropriate U.S. Attorney’s Office, and it seems that often a final copy of a settlement agreement is not returned to the BOP attorneys with whom the various Assistant United States Attorneys liaise. Letter from Ian M. Guy to Joanna Schwartz, supra note 143. 623 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) not always answer the empirical questions we have posed. For example, the BOP did not provide settlement agreements for many of the cases, making it impossible to confirm whether individual officers contributed to the settlements. The BOP also did not provide us with the complaints filed or other information about the underlying causes of actions in several cases, making it difficult to discern whether the plaintiffs brought Bivens claims in each of these cases.260 And the BOP’s files contained some—but not all—motions and opinions in these cases, offering an incomplete picture of the course of litigation in these cases. To secure missing information on the cases, we looked to other sources. We began by reviewing the dockets of each of these 209 cases and by pulling relevant complaints, motions, settlement agreements, and court orders.261 We also contacted plaintiffs’ and defense attorneys who entered appearances in these cases when the FOIA materials and electronic dockets provided insufficient information about the dynamics of litigation or terms of settlement in these cases. Having supplemented the BOP’s records, we coded the 209 cases for multiple variables including whether the plaintiff proceeded pro se or with counsel, the date and district in which the case was filed, the nature of the claim and injury, the date the case was closed, and the amount paid to plaintiffs. In an effort to ensure coding accuracy and consistency, one of us undertook the initial coding of all 209 files. Another of us reviewed approximately 20% of those files, focusing on files where the coding decisions were less obvious. We paid particular attention to what claims were included in the initial complaint—claims brought under the Federal Tort Claims Act (FTCA), FOIA, Bivens, or some other theory of liability—what claims remained in the case at the time it was resolved, and the mechanisms by which the causes of action shifted during the course of litigation. After coding the 209 cases in the BOP’s FOIA production, we focused our study on the 108 matters in which the plaintiff included Bivens claims at some point during the course of litigation.262 We based our decision to include or exclude specific cases on a careful examination of the allegations in the initial and amended complaints as the case evolved from initial filing to resolution. Most of these excluded claims (97 of the 101 non-Bivens cases) were filed and resolved under the FTCA. In these cases, the plaintiff did not seek to impose liability on individual officers under the Bivens doctrine at any point during 260. Incarcerated people can bring other types of claims, including claims under the FTCA and FOIA, directly against the U.S. government for conduct by BOP officials. 261. We relied on the federal courts’ Public Access to Court Electronic Records (PACER) system as well as Bloomberg Law to review dockets. 262. We report more fully on those cases in the main body of the Article. 624 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) the course of litigation. The remaining four excluded non-Bivens cases were brought under FOIA, which does not make damages available to successful litigants. In an effort to confirm that the Bivens case information produced by the BOP was comprehensive and representative, we conducted a follow-up assessment using electronic docket records aggregated by Bloomberg Law. We searched the files of cases initiated between 2005 and 2014, seeking to identify all cases asserting Bivens claims against BOP employees in which a settlement or plaintiff’s judgment was reached. Having identified successful claims, we compared our list to the list of cases disclosed by the BOP. We found sixtythree additional Bivens cases that were not included in the BOP’s FOIA files (all but one involved a settlement). We then submitted additional FOIA requests to the BOP to gather additional information about the sixty-three cases we uncovered. There remain some gaps in our information about the 171 cases in the dataset. The BOP was unable to provide us with information about the amount of the settlements in eleven cases, and could only approximate the settlement amounts in another eight cases. Accordingly, we do not know the total amount plaintiffs recovered in these 171 cases, but know that they received more than $18,975,629. The BOP was also unable to locate and/or produce settlement agreements in sixty of the cases in the dataset, so we do not always know exactly how the parties framed their agreements—relevant particularly when the plaintiff had agreed to substitute FTCA claims for the Bivens claims in some manner. Despite these limitations, the BOP production and our own compilation provide an apparently comprehensive and reliable portrait of successful Bivens cases brought against the BOP.263 The BOP production reveals the nature of the claims asserted, the manner in which the claims were resolved, and the source of funds that were paid to plaintiffs. Similar claim and payment information appears in claims we identified in our separate review of files from the specified ten-year period. Thus, while the BOP limited its production to settlements of $1000 or more, we found similar practices in the cases where less than $1000 was paid. Similarly, the BOP’s production was limited to cases in which government attorneys represented the defendants. One might worry that payment practices in such represented cases could systematically differ from the practices in cases in which the federal official defendant was not represented by government counsel.264 By examining all filings in the specified 263. We of course cannot make claims about the settlement and indemnification practices of other federal agencies. 264. We discuss indemnity policies in Part III, but we do not know what factors inform the Department of Justice’s discretionary decisions to deny legal representation. We understand that it may happen in circumstances in which the government agrees with footnote continued on next page 625 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) ten-year period, our review included all Bivens cases, not just those in which the government agreed to represent the defendant. That we found a similarity between the payment practices in the BOP production and our own compilation gives us some confidence that the pattern we identify was not simply an artifact of the government’s decision to provide legal counsel. the victim and has criminally prosecuted the official for the misconduct at issue. See Email from Paul Figley, Prof. of Legal Rhetoric, Wash. Coll. of Law, Am. Univ., to James E. Pfander, Professor of Law, Nw. Pritzker Sch. of Law (July 21, 2018, 3:24 PM) (on file with authors). Disclaimers of personal liability representation may be accompanied by disclaimers of FTCA coverage, perhaps on the theory that the FTCA does not accept liability for assault and battery or on the theory that particularly egregious conduct falls outside the scope of employment. See Gregory C. Sisk, Holding the Federal Government Accountable for Sexual Assault, 104 IOWA L. REV. 731, 777-81 (2019) (noting the omission of assault and battery claims from FTCA coverage, except when committed by law enforcement officers, and describing trend in state courts to broaden the scope of vicarious liability by treating intentional torts as within the scope of employment). 626 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Appendix B: Data Contained within this Appendix are the data upon which our discussion is based. The four tables reflect the four main ways in which Bivens cases in the dataset were resolved. Table 1 sets out the eight cases in the dataset in which there was some contribution by individual officers. Tables 2-4 set out the cases in which individual officers did not contribute, and the ways those Bivens claims were resolved—through dismissal during litigation (Table 2); through settlement of Bivens and FTCA claims with the United States assuming the entire settlement (Table 3); and through formal or informal substitution of the Bivens claim with an FTCA claim (Table 4). The data are largely selfexplanatory, with one exception: The primary allegations in each case are denoted through the following abbreviations: ACCOM (failure to accommodate disability); COC (conditions of confinement); COPF (correction officer use of force); COSA (correction officer sexual assault); DP (due process violation) FOOD (food contamination); FTPPF (failure to protect prisoner from physical force); FTPSA (failure to protect prisoner from sexual assault); MED (medical malpractice); PI (personal injury); REL (religious discrimination); and SPEECH (free speech). 627 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Table 1 Cases in Which Officers Contributed to the Payment of Settlements and Judgments Case Name Ortiz v. Bezy 2:05-cv-00246 Court S.D. Ind. Allegations MED Claims at Filing Bivens Claims Immediately Before Resolution Total Owed by Individual Defendants Total Paid by U.S. Bivens $10,000 (possibly paid by insurer) 0 0 MonclovaChavez v. McEachern 1:08-cv-00076 E.D. Cal. COPF Bivens Bivens $10,000 (though defendants defaulted) Doe v. United States 1:08-cv-00517 D. Haw. COSA Bivens & FTCA Bivens & FTCA $3000 $67,500 Bolden v. Marberry 2:09-cv-00312 S.D. Ind. MED Bivens Bivens Unknown 0 Doe v. United States 1:12-cv-00640 D. Haw. COSA Bivens & FTCA Bivens & FTCA $25,000 $15,000 Harrison v. Jackson 1:12-cv-04459 N.D. Ga. COSA Bivens & FTCA Bivens $11,000 0 Shirley v. Manning 3:13-cv-00236 D. Or. PI Bivens & FTCA Bivens & FTCA $1500 $6000 Jones v. Caraway 2:14-cv-00319 S.D. Ind. COC Bivens Bivens $662.95 0 $61,162.95 $88,500 Total paid: 628 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Table 2 Cases in Which Bivens Claims Were Dismissed During Litigation Case Name Court Allegations Reason Bivens Claims Dismissed Total Paid by U.S. Harmon v. United States 5:00-cv-01072 S.D. W. Va. PI/MED Dismissed sua sponte by court $2500 Jones v. Reno 5:01-cv-03094 D. Kan. PI/MED Dismissed sua sponte by court $7500 $40,000 Northington v. Hawk-Sawyer 2:04-cv-01032 C.D. Cal. MED Jury found in favor of individual defendants on Bivens claim; in favor of plaintiff on FTCA claim Manning v. U.S. Dep’t of Justice 8:04-cv-01486 D. Md. PI Voluntarily dismissed $3000 Baker v. United States 1:05-cv-00147 W.D. Pa. FTPPF/ MED Dismissed on MTD/MSJ $90,000 Gonzalez v. Sanders 2:05-cv-00269 E.D. Ark. PI/MED Dismissed on MSJ $813,000 $1000 Fernandez v. Fed. Bureau of Prisons 1:05-cv-03288 N.D. Ill. PI Defendants’ motion to substitute United States for all parties granted Clark v. United States 3:06-cv-00016 S.D. Ill. COPF Voluntarily dismissed $10,049.51 Almashleh v. United States 1:06-cv-00106 W.D. Pa. MED Dismissed on MSJ Unknown Vandersteen v. Wessberg 0:06-cv-02251 D. Minn. PI Dismissed on MTD $10,000 Murillo v. United States 2:06-cv-06699 C.D. Cal. MED Dismissed sua sponte by court $75,000 Chess v. United States 1:07-cv-05333 N.D. Ill. FTPPF Dismissed on MSJ $25,000 Aviles v. Levi 2:08-cv-02440 E.D. Pa. REL/MED Voluntarily dismissed $32,500 629 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Reason Bivens Claims Dismissed Total Paid by U.S. Vincent v. Fed. Bureau of Prisons 2:08-cv-03286 C.D. Cal. MED Voluntarily dismissed $250,000 McIntosh v. Glen 1:09-cv-00295 W.D. Pa. MED Dismissed on MSJ $7000 Martinez v. United States 5:09-cv-00375 C.D. Cal. COPF/ MED Dismissed on MTD $1500 $3000 Edenfield v. United States 1:09-cv-01384 N.D. Ga. PI Three claims dismissed on MTD; two claims dismissed sua sponte by court Ford-Sholebo v. United States 1:09-cv-02287 N.D. Ill. MED Voluntarily dismissed $700,000 Warrender v. Lappon 1:09-cv-02697 E.D.N.Y. MED Dismissed on MTD/MSJ $1000 Rockett v. United States 1:09-cv-03036 E.D.N.Y. PI Dismissed on MSJ $4000 West v. Peoples 1:09-cv-03328 N.D. Ga. COPF Dismissed on MTD $4500 Duran v. Lindsay 1:09-cv-05238 E.D.N.Y. MED Dismissed on MTD/MSJ Unknown Zidell v. Kanan 4:10-cv-00106 N.D. Tex. MED Dismissed sua sponte by court Unknown $1500 Barker v. McPherson 2:10-cv-00314 S.D. Ind. COPF Bench trial decision in favor of individual defendants on Bivens claim; in favor of plaintiff on FTCA claim Weathington v. United States 1:10-cv-00359 W.D. La. FTPPF/ COPF/MED Dismissed on MSJ $10,000 Irvin v. Owens 9:10-cv-01336 D.S.C. FTPPF Dismissed on MSJ $465,000 Ford v. Mitchell 1:10-cv-01517 D.D.C. OTHER (sentencing error) Dismissed on MTD $350,000 630 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Reason Bivens Claims Dismissed Total Paid by U.S. Banks v. United States 1:10-cv-05308 E.D.N.Y. MED Voluntarily dismissed $100,000 Navarro-Morales v. Lockett 5:11-cv-00245 M.D. Fla. MED Voluntarily dismissed $400,000 Renner v. Cole 3:11-cv-00419 W.D. Wisc. MED Dismissed on MSJ $150,000 Shuster v. Cabanas 1:11-cv-01764 D.N.J. MED Dismissed on MTD/MSJ $6250 Fontanez v. Lopez 1:11-cv-02573 D.N.J. MED Dismissed sua sponte by court $20,000 Williams v. United States 2:11-cv-05612 E.D. Pa. MED Voluntarily dismissed $575,000 Mohsen v. United States 4:12-cv-00045 D. Ariz. MED Dismissed on MSJ Unknown Lee v. United States 4:12-cv-00197 N.D. Tex. COC Voluntarily dismissed $60,000 $2000 Penick v. United States 2:12-cv-00341 S.D. Ind. MED Bivens claims against nineteen named defendants dismissed sua sponte by court; Bivens claims against two named defendants dismissed on MTD/MSJ Spells v. Fenstermaker 3:12-cv-00455 M.D. Pa. FOOD Dismissed sua sponte by court Approximately $12,000 Ruiz v. United States 1:12-cv-00521 D.N.M. MED Dismissed on MSJ $132,500 Legrand v. United States 3:12-cv-00743 M.D. Pa. FOOD Dismissed on MTD/MSJ $2500 Woods v. Holt 1:12-cv-00900 M.D. Pa. FOOD Voluntarily dismissed Approximately $2000 Brown v. Holt 3:12-cv-00956 M.D. Pa. FOOD Voluntarily dismissed Approximately $2000 Love v. U.S.P. Canaan 1:12-cv-01030 M.D. Pa. FOOD Voluntarily dismissed Approximately $2000 631 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Reason Bivens Claims Dismissed Total Paid by U.S. Strong v. Holt 3:12-cv-01036 M.D. Pa. FOOD Voluntarily dismissed Approximately $2000 Connelly v. Holt 1:12-cv-01187 M.D. Pa. FOOD Voluntarily dismissed Approximately $2000 Cruz v. United States 5:12-cv-02149 D.S.C. COPF/PI Dismissed on MTD $45,000 Harris v. United States 1:12-cv-02392 M.D. Pa. FOOD Dismissed on MTD Approximately $1500 M.G. v. United States 3:12-cv-02956 S.D. Cal. FTPSA Dismissed on interlocutory appeal $50,000 Smith v. United States 3:13-cv-00337 W.D. Tex. PI/MED Dismissed on MTD/MSJ $150,000 Wakefield v. United States 4:13-cv-00339 N.D. Fla. FTPPF Dismissed on MSJ $20,000 Johnson v. Merritt 2:13-cv-00441 S.D. Ind. COPF Voluntarily dismissed Unknown Jackson v. Welliver 3:13-cv-00641 M.D. Pa. PI Voluntarily dismissed $2850 Hildebrand v. United States 1:13-cv-01233 C.D. Ill. MED One claim dismissed on MTD; one claim voluntarily dismissed $57,000 Kontos v. United States 5:13-cv-01398 C.D. Cal. MED Voluntarily dismissed $199,000 Thornton v. United States 1:14-cv-00447 W.D. La. MED Dismissed sua sponte by court $7800 Ruffin v. United States 1:14-cv-00761 D.D.C. MED Voluntarily dismissed $3000 Clemmons v. United States 2:14-cv-00885 N.D. Ala. MED Dismissed sua sponte by court $2550 Dunbar v. United States 1:14-cv-01838 D. Colo. FTPPF/ SPEECH Dismissed on MTD $30,000 Zepeda-Zalaberry v. Smith 2:14-cv-02738 D. Ariz. MED Dismissed sua sponte by court $1100 Davis v. United States 2:15-cv-00121 E.D. Ark. MED Dismissed sua sponte by court $600,000 Total paid: $5,547,100 632 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Table 3 Cases in Which Bivens and FTCA Claims Remained at Settlement, with Payment Only by the U.S. Government Case Name Court Allegations Settlement Contingent on Dismissal of Bivens Claims Separate from Dismissal of Action Gil v. Reed 3:00-cv-00724 W.D. Wisc. MED Yes $20,000 Muhannad v. United States 3:00-cv-00864 S.D. Ill. COPF/REL Yes $1205 Kikumura v. Osagie 1:03-cv-00236 D. Colo. MED No $30,000 Fritts v. Zych 5:03-cv-03377 D. Kan. MED No $125,000 Teague v. United States 1:04-cv-01800 D. Colo. FTPPF No $3,000 Oleson v. United States 3:05-cv-00033 W.D. Wisc. PI No $3500 Haas v. Prince 2:05-cv-00112 E.D. Ark. MED No $2360 Ahart v. Willingham 3:05-cv-01016 D. Conn. MED No $400,000 Bramwell v. Murray 1:05-cv-07504 S.D.N.Y. COPF/MED No $50,000 Johnson v. United States 2:06-cv-00006 N.D. W. Va. MED No $30,000 Zepeda v. United States 1:06-cv-00676 D. Haw. COSA No $17,500 Custard v. Turner 1:06-cv-01036 D. Colo. ACCOM No $1400 Burnette v. Fed. Bureau of Prisons 1:06-cv-01396 W.D. La. FTPPF No $35,000 Total Paid by U.S. 633 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Settlement Contingent on Dismissal of Bivens Claims Separate from Dismissal of Action Yates v. United States 2:06-cv-01876 E.D. Pa. COSA No $246,250 Donaldson v. Samuels 1:06-cv-05627 D.N.J. COPF No $19,000 Dantone v. Winn 4:06-cv-40022 D. Mass. PI/MED No $3000 Ricketts v. Assoc. Warden of Unicor 1:07-cv-00049 M.D. Pa. FTPPF No $400,000 Lindsey v. Fed. Bureau of Prisons 4:07-cv-00461 N.D. Tex. MED No $1750 Shaheed v. Nalley 3:07-cv-00679 S.D. Ill. COPF No $48,000 Nolan v. Hamidullah 4:07-cv-01141 D.S.C. ACCOM Yes $14,500 Castaneda v. United States 2:07-cv-07241 C.D. Cal. MED No $1,950,000 Wormley v. United States 1:08-cv-00449 D.D.C. DP No $50,000 Houston v. United States 2:08-cv-01076 C.D. Cal. COSA No $235,000 Albin v. United States 1:08-cv-01271 S.D. W. Va. MED No $985,000 Morris v. Jones 2:08-cv-03842 E.D. Pa. MED Yes $75,000 Magassouba v. United States 1:08-cv-04560 S.D.N.Y. FTPPF/ COPF/MED No $20,000 Howard v. United States 1:09-cv-00096 E.D.N.Y. COPF/MED No $30,000 Ngerntongdee v. United States 2:09-cv-00213 W.D. Wash. MED No $880,000 Total Paid by U.S. 634 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Settlement Contingent on Dismissal of Bivens Claims Separate from Dismissal of Action Richardson v. Fed. Bureau of Prisons 1:09-cv-02082 E.D. Cal. FTPSA/ MED No $2500 Manswell v. United States 1:09-cv-04102 S.D.N.Y. PI/MED Yes $12,500 Lisonbee v. United States 1:10-cv-00058 D. Utah MED No $412,500 Bell v. Zuercher 7:10-cv-00072 E.D. Ky. MED No $975,000 Louis v. United States 5:10-cv-00075 N.D. W. Va. FTPSA/ FTPPF No $15,000 Rigdon v. Carey 5:10-cv-00130 M.D. Fla. MED Yes $3500 Cottini v. United States 2:10-cv-00294 C.D. Cal. MED Yes $200,000 Hensarling v. United States 5:10-cv-00344 M.D. Fla. FTPPF No $62,500 Lichtenberg v. United States 1:10-cv-00353 D. Haw. MED No Unknown Rappe v. Harvey 1:10-cv-04636 N.D. Ill. MED No $25,000 Fitts v. Malatinsky 2:10-cv-11100 E.D. Mich. MED Yes $15,000 Satterwhite v. Dy 2:11-cv-00528 W.D. Wash. MED No $1,400,000 Brewer v. Fed. Bureau of Prisons 1:11-cv-00605 E.D.N.Y. DP No $43,000 Hoslett v. Dhaliwal 3:11-cv-00674 D. Or. MED No $15,000 Furtney v. United States 3:11-cv-01090 D. Or. PI No $80,000 Total Paid by U.S. 635 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Settlement Contingent on Dismissal of Bivens Claims Separate from Dismissal of Action Shelton v. Bledsoe 3:11-cv-01618 M.D. Pa. FTPPF/ COPF No $11,000 Sheridan v. Rios 0:11-cv-02487 D. Minn. PI/MED No $15,000 Vailette v. Lindsay 1:11-cv-03610 E.D.N.Y. MED No $12,000 Words v. United States 2:11-cv-14261 E.D. Mich. MED No $15,000 Brooks v. Bledsoe 3:12-cv-00067 M.D. Pa. SPEECH No $500 Cooper v. United States 5:12-cv-00162 M.D. Fla. COPF No $10,000 Carter v. Moon 5:12-cv-00269 M.D. Fla. OTHER (fraud) No $5000 Spotts v. Lindsey 3:12-cv-00583 M.D. Pa. FOOD No Approximately $8000 Lee v. Pfister 5:12-cv-00794 C.D. Cal. COPF Yes $6500 Riopedre v. United States 8:12-cv-02806 D.S.C. MED No $375,000 Clarke v. Fed. Transfer Ctr. 2:13-cv-00026 E.D. Ark. MED No $800,000 Mack v. United States 1:13-cv-00280 M.D. Pa. FOOD No $2000 Hirano v. Williams 5:13-cv-02371 C.D. Cal. COC No $50,000 Hill v. United States 1:13-cv-03404 D. Colo. FTPSA Yes $70,000 Morales v. United States 1:14-cv-00485 E.D.N.Y. MED Yes $52,500 Total Paid by U.S. 636 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Case Name Court Allegations Settlement Contingent on Dismissal of Bivens Claims Separate from Dismissal of Action Total Paid by U.S. $98,300 Am. Humanist Ass’n v. Fed. Bureau of Prisons 3:14-cv-00565 D. Or. REL No (injunctive relief with attorneys’ fees paid by United States) Luna v. Jordan 1:14-cv-02028 M.D. Pa. FTPPF/ COPF Yes $5350 Ingram v. United States 1:14-cv-02091 S.D.N.Y. MED Yes $850,000 Chicarielli v. United States 1:14-cv-06765 S.D.N.Y. FTPPF/ MED Yes $36,500 Dawkins v. Greenspan 2:14-cv-07269 C.D. Cal. MED No $150,000 Total paid: $11,510,615 637 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Table 4 Cases in Which Only Bivens Claims Remained at Settlement, with Payment by U.S. Government Docket Shannon v. Fed. Bureau of Prisons 1:03-cv-00352 Johnson v. Martinez 2:04-cv-01967 Merriweather v. Zamora 2:04-cv-71706 Montgomery v. Johnson 7:05-cv-00131 Hammond v. Sherman 1:05-cv-00339 Sloan v. Pugh 1:05-cv-00527 Nunez v. Lindsay 3:05-cv-01763 Hill v. Laird 2:06-cv-00126 Brown v. LaManna 2:06-cv-00390 Buckley v. Harding 1:06-cv-00413 Al-Kidd v. Sugrue 5:06-cv-01133 Green v. Wiley 1:07-cv-01011 Williams v. Warmerdorf 3:07-cv-01283 Williams v. Smith 1:07-cv-01382 Ellis v. United States 1:08-cv-00160 Rodriguez v. Wiley 1:08-cv-02505 Court Allegations Formal Substitution of FTCA Claims for Bivens Claims D. Colo. SPEECH Yes $6000 E.D. Pa. MED Yes $40,000 E.D. Mich. OTHER (privacy) No $27,000 W.D. Va. COPF/ SPEECH No $115,000 W.D. Pa. COPF Yes $18,000 D. Colo. OTHER (sentence miscalculated) No $30,000 M.D. Pa. SPEECH Yes $3000 E.D.N.Y. SPEECH No $10,000 D.S.C. ACCOM No $15,000 D. Colo. FTPPF Yes $18,500 W.D. Okla. DP Yes $28,500 D. Colo. MED No $2100 M.D. Pa. FTPPF/ MED Yes $1500 M.D. Pa. FTPPF Yes Unknown W.D. Pa. REL Yes Unknown D. Colo. MED No $1150 Total Paid by U.S. 638 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Docket Garcia v. Hicks 1:08-cv-07778 Taylor v. Miller 7:09-cv-00145 Patel v. Fed. Bureau of Prisons 1:09-cv-00200 McCarroll v. Matteau 9:09-cv-00355 Brown v. Laing 6:09-cv-00392 Brown v. Blocker 2:09-cv-00434 Willis v. Lappin 1:09-cv-01703 Counts v. Hollingsworth 3:10-cv-00229 Anderson v. Drew 1:10-cv-00996 De Anda v. Smith 1:10-cv-01094 Stine v. Allred 1:11-cv-00109 Montoya v. Wall 1:11-cv-01414 Tuttamore v. Allred 1:11-cv-01522 Freeman v. Woolston 1:11-cv-01756 Brizard v. Terrell 1:11-cv-02274 Bennett v. Watts 5:12-cv-00016 Mundo v. Shaw 1:12-cv-00184 Skurdal v. Fed. Det. Ctr. 2:12-cv-00706 Gillings v. Lepe 1:12-cv-01533 Court Allegations Formal Substitution of FTCA Claims for Bivens Claims S.D.N.Y. SPEECH Yes $7500 E.D. Ky. COPF No $25,000 D.D.C. REL No $50,000 N.D.N.Y. SPEECH Yes $4250 E.D. Ky. COPF/ MED No $9000 D.S.C. MED No $15,000 E.D. Cal. FTPPF No $3000 S.D. Ill. MED No Unknown D.S.C. OTHER (right to marry) No $5000 C.D. Ill. FTPPF/ MED Yes $3000 D. Colo. MED Yes $2000 C.D. Ill. MED Yes $475,000 D. Colo. MED No $5000 D. Colo. COPF Yes $1500 E.D.N.Y. FTPPF Yes $4000 S.D. Miss. MED No $10,350 N.D. W. Va. REL No $20,000 W.D. Wash. REL No $45,000 E.D. Cal. COPF Yes $2750 Total Paid by U.S. 639 Electronic copy available at: https://ssrn.com/abstract=3343800 The Myth of Personal Liability 72 STAN. L. REV. 561 (2020) Docket Holmes v. Lepe 1:12-cv-01649 Caballero v. Mejia 4:13-cv-00630 Shepherd v. Palmer 1:14-cv-02992 Laurent v. Castellanos 1:14-cv-03340 Bolden v. Beaudouin 1:14-cv-05470 Court Allegations Formal Substitution of FTCA Claims for Bivens Claims E.D. Cal. COPF No $2501 N.D. Fla. MED Yes $500,000 D.N.J. MED Yes $100,000 E.D.N.Y. FTPPF Yes $5500 E.D.N.Y. MED Yes Unknown Total Paid by U.S. $1,611,101 Total paid: 640 Electronic copy available at: https://ssrn.com/abstract=3343800