United States v. Louisville Jefferson County Metro Government

• Court: District Court, W.D. Kentucky
• Decided: January 18, 2025
• Docket: 3:24-cv-00722
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT WESTERN DISTRICT OF KENTUCKY LOUISVILLE DIVISION

UNITED STATES OF AMERICA PLAINTIFF

v. No. 3:24-cv-722-BJB

LOUISVILLE JEFFERSON COUNTY DEFENDANT METRO GOVERNMENT ORDER RE: PROPOSED CONSENT DECREE’S TRANSPARENCY & SUPPORT “Measure twice, cut once.” Muhammad-Ali v. Final Call, Inc., 832 F.3d 755, 757 (7th Cir. 2016). This carpenters’ adage is a good reminder for lawyers and judges considering the “extraordinary” step of a consent decree.1 Once entered as a court order, a decree may bind elected officials and their successors for years, involve judges and court-appointed administrators in the innerworkings of local government, and shift policymaking and accountability from City Hall to the U.S. Courthouse. “The point of discretion-limiting settlements,” most observers agree, “is to make policy that is difficult for a future administration to modify.” Simon Brewer, The Attorney General’s Settlement Authority and the Separation of Powers, 130 YALE L.J. 174, 217 (2020). And when the settlement takes the form of a consent decree, the judge’s direct involvement in that policymaking represents a further encroachment on the democratic process. Scholars and public officials disagree about whether and when such drastic interventions are necessary, wise, or even lawful.2 But few doubt their

1 See OWEN M. FISS, THE CIVIL RIGHTS INJUNCTION 92 (1978) (“What is extraordinary about the structural injunction is the nature of the enterprise and what that does to the judicial office.”). 2 See, e.g., id. at 93 (“The judge is brought perilously close, so it is claimed, to breaching the line between administration and law.”); Horne v. Flores, 557 U.S. 433, 447–48 (2009) (Breyer, J., dissenting) (describing two “side[s] of a scholarly debate about how courts should properly handle decrees in ‘institutional reform litigation’”); Michael W. McConnell, Why Hold Elections? Using Consent Decrees to Insulate Policies from Political Change, 1987 U. CHI. LEGAL F. 295, 297–98 (“[C]onsent decrees frequently require the executive to surrender a portion of its authority to the courts; this raises the usual set of problems of reconciling aggressive judicial supervision of executive performance with the structural principles set out in the Constitution.”). In New Orleans, the Mayor, Governor, and Attorney General have tried, thus far unsuccessfully, to fully end the Police Department’s consent decree after 12 years. See Joint Statement from Attorney General Liz Murrill and Governor Jeff Landry on Partnership with City of New Orleans in Effort to End NOPD Consent Decree, Louisiana Off. of the Att’y Gen., Jan. 13, 2025, available at https://www.ag.state.la.us/Article/218. significance, cost, and duration. See, e.g., Horne v. Flores, 557 U.S. 433, 447–48 (2009). Certainly no one should doubt the significance, cost, and potential duration of these legal proceedings. After a 2-year investigation examining 5 years of policing, followed by an additional year of negotiations, the U.S. Department of Justice and Louisville Metro Government have proposed 5 or more years of judicial oversight of the Louisville Metro Police Department. Funding the court-appointed monitor alone could run more than $7million dollars, to say nothing of additional personnel, operational, and compliance costs. And before the reform efforts end, the judge and monitor—not the Mayor, Police Chief, or Metro Council—would be responsible for determining whether the Police Department had achieved “substantial compliance” with federal law. Proposed Consent Decree (DN 4-1) ¶ 701. The Police Department, according to the United States, is engaging in an ongoing pattern of incredibly serious legal violations: using excessive force—guns, tasers, and police dogs, for example— against the citizens of Louisville; stopping them on the streets without a lawful reason; policing black people more harshly than others on the basis of their skin color; infringing protesters’ First Amendment rights; and denying equal emergency services to persons with behavioral disabilities. Complaint (DN 1) ¶¶ 20–78. The City admits its officers have violated the rights of its citizens from time to time—as this Court is all too familiar. See, e.g., United States v. Wilson et al., No. 3:22-cr-47-BJB, United States v. Crews, No. 3:22-cr-20-BJB; cf. Napper v. Hankison, et al., No. 3:20-cv-764-BJB. But the City does not admit that the Police Department has engaged in a pattern of such violations. See generally Answer (DN 22). The parties could’ve publicly offered evidence and aired their competing positions at a trial to resolve this dispute. Instead, the City has agreed to settle rather than fight the allegations. Generally, the law favors consensual agreement over contested adjudication. See United States v. Lexington-Fayette Urban County Government, 591 F.3d 484, 490 (6th Cir. 2010). And given the lengthy and difficult reckoning over policing in Louisville, the desire to turn from past disagreements to future reforms is entirely understandable. To do so, the City has agreed to hundreds of changes in its laws, policies, and personnel, followed by a robust process of review and assessment. Joint Motion for Entry (DN 4). Those detailed reforms constitute a 240-page decree the parties have submitted for the Court’s approval. This would mark the end of the matter if the City and Justice Department had structured this resolution as a bilateral agreement. They could’ve adopted their own timeframe, benchmarks, outside reviewer, and enforcement mechanisms—just as they did in the proposed consent decree—but without the need for judicial approval and ongoing involvement. See Brewer, 130 YALE L.J. at 187–89 (challenging perceived gap between enforceability of settlements relative to consent decrees). The Supreme Court has made clear that the parties—without involving a judge in the implementation of a consent decree—could’ve asked the Court to retain jurisdiction to enforce any breaches. See Kokkonen v. Guardian Life. Ins., 511 U.S. 375, 381–82 (1994) (court may dismiss case and retain jurisdiction over agreement). And relief surely could’ve been swift if the Justice Department showed the police had relapsed to or persisted in ongoing serious constitutional violations. But here the parties took a different and less conventional path. They didn’t condition the end of this lawsuit on a settlement, but instead began this suit with an agreement that the Court would begin overseeing the LMPD consent decree. As this Court has previously explained, a “consent decree is a strange hybrid in the law, both a voluntary settlement agreement which could be fully effective without judicial intervention and a final judicial order placing the power and prestige of the court behind the compromise struck by the parties—in effect, a settlement agreement subject to continued judicial policing.” Lexington Insurance Co. v. Ambassador Group LLC, 581 F. Supp. 3d 863, 865 (W.D. Ky. 2021) (cleaned up) (quoting Brown v. Neeb, 644 F.2d 551, 557 (6th Cir. 1981), and Williams v. Vukovich, 720 F.2d 909, 920 (6th Cir. 1983)). This unusual remedy pulls judges—trained to apply law to historical facts—outside their bailiwick into the unfamiliar territory of prospective operations and oversight. See FISS, above, at 92–93. Contrary to some suggestions, judicial approval of proposed decrees is not automatic. Before approving such an arrangement, judges must determine whether it is a “fair, adequate, and reasonable” resolution of the alleged legal violations and otherwise “consistent with the public interest.” Lexington-Fayette, 591 F.3d at 489 (citations omitted). Caselaw interpreting this standard is relatively limited.3 But at least a few principles guide this exercise of judicial discretion. Courts must undertake an “independent evaluation” of the decree and avoid “rubber stamp approval.” United States v. Akzo Coatings of America, Inc., 949 F.2d 1409, 1435 (6th Cir. 1991) (quoting City of Detroit v. Grinnell Corp., 495 F.2d 448, 463 (2d Cir. 1974)). But they may not “substitute [their] own judgment for that of the parties to the decree.” Id. (citing United States v. Jones & Laughlin Steel Corp., 804 F.2d 348 (6th Cir.

3 Beyond the oft-repeated standard of “fair, adequate, reasonable, and consistent with the public interest,” the Court has found little in the way of specific guidance for district courts deciding whether to enter a proposed consent decree. In one sense this is unsurprising: “[t]he death of caselaw is presumably correlated with the lack of an incentive for a consenting party to take an appeal.” Lexington Insurance Co. v. Ambassador Group LLC, 581 F. Supp. 3d 863, 868 n.2 (W.D. Ky. 2021). That’s nevertheless concerning: judicial review rests on the Court’s ability to apply a “neutral principle”—some “intellectually coherent statement of the reason for a result which in like cases will produce a like result.” ALEXANDER M. BICKEL, THE LEAST DANGEROUS BRANCH 59 (2d ed. 1986). And the only principles advanced by the parties— deference and institutional competence—are hardly neutral ones. 1986) (judge may approve or reject—but not alter—consent decree)). As counsel for the Justice Department noted during the recent hearing on this proposed decree, this evaluation “must not be turned into a trial or a rehearsal of the trial.” Grinnell, 495 F.2d at 462; see Hearing Tr. (DN 52) at 118:2–3. Yet only by understanding “sufficient facts” about an agreement may a judge “intelligently” decide whether to approve it. Grinnell, 495 F.2d at 462–63. Although the standard is forgiving, not every consent decree passes muster. See, e.g., United States v. New York City Housing Auth., 347 F. Supp. 3d 182, 197–98 (S.D.N.Y. 2018) (Although deferential, “a district court is not merely a ‘rubber stamp’ for a government agency seeking the entry of a consent decree.”) (quoting SEC v. Citigroup Global Markets, Inc., 752 F.3d 293 (2d Cir. 2014)). Some are rejected by city officials.4 And some demand judicial modification or termination in response to unjustified extensions or other changed circumstances. See Horne, 557 U.S. at 447–48; cf. Rizzo v. Goode, 423 U.S. 362, 378–80 (1976) (reversing district court that “injected itself by injunctive decree into the internal disciplinary affairs” of the Philadelphia Police Department). Curiously, the Justice Department seems to disagree with this approach—as this case’s brief history reveals. The parties filed the lawsuit and proposed decree on December 12, 2024, and the Court promptly asked for the parties’ “proposed timeline” for judicial review of the decree. DN 7. In response, the parties skipped past any judicial scrutiny, assumed a signed decree, and explained the process for selecting an independent monitor. DN 8-1 at 1 (beginning proposed schedule with “Monitor Selection”). Then the Court issued a preliminary order setting a hearing and outlining areas for the parties’ attention. See DN 18. But the Department showed up with no witnesses and no evidence—only the 2023 Findings Report (DN 1-1) that it incorporated into the Complaint. See n.9, below. The Justice Department implicitly and at times explicitly asked the Court to simply defer to its expertise and judgment. The settlement here “is a consent decree,” counsel explained, and “the courts are deferring to the parties on these sorts of agreements.” Hearing Tr. at 152:7–12. Deference to litigation risk and policy judgment is one thing, but blind deference—without any insight into the facts and law on which the parties rely—is another. And the latter sort of deference plainly differs from the “independent

4 The City of Memphis recently rejected the Justice Department’s request to jointly seek a consent decree. See Megan Fayard & April Thompson, Memphis Hires Former Federal Judge to Monitor MPD, WREG MEMPHIS, https://wreg.com/news/local/memphis-hires- former-federal-judge-to-monitor-mpd/ (last visited Jan. 15, 2025). Instead they named a former judge—not an active one overseeing the litigation—to supervise implementation. According to the City Attorney, “there are better ways to reimagine policing [than a consent decree] that do not slow the process or cost the taxpayers millions of dollars.” Letter from Tannera G. Gibson, Memphis City Attorney, to DOJ Civil Rights Division Officials Re: Department of Justice Pattern or Practice Investigation of City of Memphis, at 2 (Dec. 4, 2024). evaluation” contemplated by the Sixth Circuit. Akzo Coatings, 949 F.2d at 1435 (quotation omitted). So the Court aimed to evaluate the reasonableness and adequacy of the proposed consent decree by asking counsel several rather basic questions about the factual allegations underlying the Justice Department’s claimed violations. These concerned the Justice Department’s investigatory findings, their relationship to the proposed remedies, and the baseline they set for the Court’s eventual evaluation of the Police Department’s progress toward “substantial compliance” with the Consent Decree (and presumably the Constitution). • How many times over the past decade does the United States believe LMPD officers shot civilians without justification? Hearing Tr. at 45:20–22. • How often have officers unlawfully tased, punched, or kicked suspects? Id. at 48:18–20. • When is the last time the Police Department violated the First Amendment rights of protestors through mass arrests or retaliation for video recordings or critical speech? Id. at 107:1–3, 111:8–10. • How did the Justice Department determine that LMPD policing affects black people disproportionately relative to other races? Id. at 112:23–25. • How should 911 dispatchers determine whether to send unarmed behavioral-health workers instead of armed police in response to a behavioral-health incident? Id. at 127:20–22. • Where in the Complaint or 2023 Findings does the United States allege the LMPD’s sexual-assault protocols violate federal law, thus justifying the robust reform process for those protocols set forth in the decree? Id. at 129:16–21, 130:8–12.5

5 The lack of allegations regarding the LMPD sexual-assault investigations is troubling— given that a large portion of the proposed consent decree would override current policies. See ¶¶ 351–96. Counsel for the Justice Department conceded that the Complaint didn’t allege any such violation of the law, Hearing Tr. at 129:22–25, but suggested that the 2023 Findings contained such an allegation. Id. at 130:4–15. They do not. The Findings state explicitly that the Department “do[es] not find reasonable cause … to believe that LMPD’s practices result in gender bias in violation of federal law.” 2023 Findings at 69. Pressed further, counsel suggested that its allegation of racial discrimination under the Safe Streets Act Remarkably, the United States refused to offer any answer and information whatsoever during the hearing. This remained true even after the Court invited counsel to identify even a rough estimate—caveated however counsel chose. Like Bartleby the Scrivener, however, counsel apparently preferred not to say. See Butte County v. Hogen, 613 F.3d 190, 195 (D.C. Cir. 2010) (quoting Herman Melville, BARTLEBY, THE SCRIVENER: A STORY OF WALL STREET 10 (Dover 1990) (1853)). At the hearing’s conclusion, the Court again invited the parties to confer and report whether they’d be submitting any additional information in support of the proposed decree’s fairness, reasonableness, adequacy, and public-interestedness. As the Court made clear, its unanswered questions are important to its own review of the proposed decree, the Police Department’s progress toward reform,6 and the public’s understanding and trust of this process. The very next day, however, counsel emailed chambers that they wouldn’t provide any further submissions. This is a curious approach for parties that bear the burden of persuasion. Even curiouser when those parties are public agencies—accountable to the people and responsible for enforcing the laws. Curiouser still when those agencies are asking the Court itself to get into the law-enforcement business by overseeing Departmental reform. The topics in question, moreover, haven’t exactly been kept under lock and key: they are covered (at least anecdotally) in a published report that the Justice Department has touted extensively to the community. See 2023 Findings. Indeed, the Department has repeatedly underscored “transparency” as a major justification for pursuing a Court-implemented consent decree rather than another resolution. See Joint Motion for Entry at 6; Initial Hearing (DN 7). What possible justification would the Department have for withholding this information? Plaintiff’s counsel expressed concern that responding would risk “retrying the case at settlement,” Hearing Tr. at 118:2–3, and coming “dangerously close to” trying the case on its merits, id. at 119:1–3. This is hard to take seriously: the Court made crystal clear that allegations and estimates would suffice—and that it merely sought to understand the relationship between the alleged violations and agreed remedies, not to find facts to that effect. The second reason is equally unsatisfying: counsel argued that disclosing the basis for its allegations would undermine the Justice Department’s litigation “leverage.” Id. at 45:23–46:3, 56:14–17. The primary

somehow embraced an allegation of sex discrimination under the same statute. Hearing Tr. at 130:19–24, 131:15–18. This adventurous argument hardly merits a response. 6 The Department has also declined to share relevant information with its ostensible partner in reforming the LMPD—the Metro Government. See Hearing Tr. at 55:5–8 (metro counsel agreeing that it would “be useful to the city in pursuing” reform with additional information the DOJ refuses to provide). concern, it seems, is to preserve its litigation leverage against the City should current settlement attempts fail and the case proceed to trial.7 Id. at 44:7–10. This appears wholly unsupported in any legal rule or judicial decision known to the Court. And counsel hasn’t supplied any support for it. If the safety and civil rights of Louisvillians are at stake, and if the City has already agreed to work toward reform of the underlying policies and practices, what would justify shielding this information from the City, Court, and public? As explained above, the Court’s role in approving and overseeing a consent decree necessarily envisions a reasoned consideration of the basis for the agreement. Consent decrees must be “limited to reasonable and necessary implementations of federal law.” Horne, 557 U.S. at 450 (citation omitted). At a minimum, that means the remedies the parties ask the Court to issue must bear some connection to the rights violations they seek to cure. See Hearing Tr. at 132:14–20 (DOJ counsel agreeing with this point). Indeed, as the Supreme Court’s canonical consent-decree opinion states plainly, “the nature of the violation determines the scope of the remedy.” Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 16 (1971). Only “[o]nce a constitutional violation is found,” even if the alleged violation is resolved in a consent decree, may “a federal court … tailor ‘the scope of the remedy’ to fit ‘the nature and extent of the constitutional violation.’” Hills v. Gautreaux, 425 U.S. 284, 293–94 (1976) (quoting Milliken v. Bradley, 418 U.S. 717, 744 (1974)). Consent decrees, moreover, are injunctions—and therefore subject even more than normal to a judge’s remedial discretion. The decision to issue injunctive relief is and always has been one “resting in the sound discretion of the Court.” 2 JOSEPH STORY, COMMENTARIES ON EQUITY JURISPRUDENCE 156 (Boston, Hilliard Gray & Co. 1836). The kind of structural injunction proposed in the parties’ consent decree demands an uncommon level of judicial supervision. See FISS, above, at 92. It follows that, in the exercise of its equitable discretion, the Court must scrutinize the consent decree with more than a cursory glance, ensuring that the scope of relief sought is consistent with the violations alleged. In other words, no rubber stamps. See Akzo Coatings, 949 F.2d at 1435 (applying the same standard—fairness, adequacy, reasonableness, and public interest) in the CERCLA context); New York City Housing Auth., 347 F. Supp. 3d at 197–98 (applying practically identical standard in Fair Housing Act context); id. at 188 (rejecting proposed consent decree because the “record reveal[ed] a substantial basis to conclude” that it was not “fair, reasonable, or consistent with the public interest”). That a consent decree may provide for “broader relief than the court could have awarded after a trial,” Firefighters v. City of Cleveland, 478 U.S. 501, 525 (1986),

7 According to the DOJ, in the last decade, only two cases under 34 U.S.C. § 12601 have actually proceeded to trial. Hearing Tr. at 46:19–20. bears only on the “scope of the remedy,” Swann, 402 U.S. at 16. It says nothing about “the nature of the violation.” Id. The Court must have a firm grasp of the latter before it can consider the former.8 Otherwise how could it assess the baseline of the Police Department’s past violations, the relationship between those violations and the proposed reforms, and future progress toward eliminating those practices and ending the decree? See Proposed Consent Decree ¶ 701. Having chosen the “extraordinary” device of a consent decree, the United States and the City of Louisville owe it to this Court and the people to explain the facts that led them here. The DOJ’s reticence is puzzling. If it so badly wanted to avoid scrutiny of its findings, it had an alternative path forward. It could have offered, for example, a regular, private settlement agreement. That agreement could look identical to the proposed consent decree here: policy reforms, monitor supervision, and the rest. And as described above, it could’ve been enforceable in this same Court. This option, it bears repeating, wouldn’t have required the Justice Department to surrender its “leverage” and share the full extent of its findings with the Court or the public. Having made the Court central to its reforms, however, the Justice Department now declines to share the factual basis for those efforts. * * * In response to the proposed agreement, this Court must comply with the law and exercise reasoned discretion in determining whether the proposed consent decree is a fair, adequate, and reasonable response to the Police Department’s alleged violations that would serve the public interest. To make that determination, the Court now orders—rather than invites—the parties to address (within a month) the questions initially raised during its hearing earlier this week. The public’s interest in the transparent, accountable, and durable reforms the Government has promised should mean the parties have little reason to avoid sharing this and related information.

8 Counsel for the Justice Department repeatedly referred to the 2023 Findings, stating that they contain “sufficient evidence … for the Court to proceed and make a finding of fair, reasonable, and adequate.” Hearing Tr. at 118:7–11. While the Findings contain some numerical data, they are primarily anecdotal, unattributed, undated, and silent on the frequency and recency of the alleged violations. Compare, e.g., 2023 Findings at 12 (listing raw use-of-force reports—both those considered excessive and those considered lawful), with id. at 12–19 (listing anecdotal allegations of excessive force). The parties’ supporting declarations, moreover, are even more abstract. See DN 36. 1. Ongoing Pattern and Practice a. Given that the parties agree the reforms contemplated in the decree are already underway, why is an injunction necessary to compel them?? b. What authority authorizes the Justice Department to withhold, based on “litigation leverage,” information relevant to the judicial decision the parties sought? 2. Fourth Amendment a. How many instances of excessive lethal force did the 2023 Findings reveal? b. How many instances of excessive non-lethal force did the 2023 Findings reveal? And what is the ratio of excessive force to lawful force during the same time period? 3. First Amendment a. To what extent do the parties maintain the Police Department has violated citizens’ First Amendment rights since the year 2021? 4, Americans with Disabilities Act a. How will dispatchers and officers determine whether a 911 call requires a police response, a mental-health response, or both? b. What legal standard determines whether “deflection” of 911 calls to a behavioral-health response team are “reasonable accommodations” required by the ADA? 5. Safe Streets Act a. Why has the Justice Department refused to share its disparate- impact assessment with the Court or the City ? b. On what legal basis could the Court impose injunctive relief to remedy a sex-discrimination allegation that the Complaint and Report have not even alleged? See 2023 Findings at 69; Complaint q 17.

Benjamin Beaton, District Judge United States District Court January 18, 2025

9 See, e.g., Hearing Tr. at 150:12-13 (Chief Humphrey: “we are going to proceed on the path that we’re on one way or the other.”); id. at 63:2-8 (Counsel for the Mayor explaining that LMPD has “already changed its policy on nonlethal force in line with what the consent decree requires”); id. at 64:3-8 (same for disproportionate strikes).