A Proposal for Police Reform: Require Effective Accountability Measures in Police Union Contracts as a Condition of Tax-Exempt Status

In the wake of the May 25 police killing of George Floyd in Minneapolis, government leaders across the nation are urgently considering reforms that might prevent police brutality.1 Policy analysts have suggested changes to federal, state, and local laws to improve transparency for police departments and accountability for officers. Proposals have been made to outlaw chokeholds, grant public access to disciplinary files, end the defense of qualified immunity, increase federal oversight of troubled police departments, and strengthen civilian review boards. Some of these proposals have already been studied empirically by scholars and think tanks and enjoy support across the political spectrum.2

But there is reason to fear that some well-meaning legislative reforms will ultimately be frustrated unless the rules governing police accountability and discipline found in police contracts are overhauled.3 For example, some advocate for police department policies creating a duty to intervene, requiring officers to prevent the inappropriate use of force by other officers. But Minneapolis has a duty to intervene policy, yet the bystander video appears to show three officers failing to (effectively) intervene while a fourth used his knee to fatally pin down George Floyd’s neck. Many of the proposed police department policies are formulated as standards (such as requiring de-escalation or warning before using deadly force “when possible,” or requiring officers to exhaust all other “reasonable alternatives” before using deadly force), which police supervisors and disciplinary officers must interpret and apply in light of the circumstances.4 Many of the proposed policies have neither the force of law nor the clarity and concreteness of rules, nor do they address disciplinary decision-making processes.5 These policies might serve other functions, such as expressing norms. But if the problem is accountability, then we should not expect that modifying standards of conduct alone will change disciplinary outcomes or, for that matter, front-line behavior.

I. The Problem

Empirical studies of police reform efforts indicate that police unions, and the collective bargaining agreements (“CBAs”) they negotiate for police officers, often provide procedural protections for officers accused of misconduct that might be superficially (and politically) attractive, but that ultimately and tragically undermine police accountability.6 Some local politicians seem willing to reform these CBAs. Minneapolis Mayor Jacob Frey stated recently that “for years in Minneapolis, police chiefs and elected officials committed to change have been thwarted by police union protections and laws that severely limit accountability.” Unfortunately, too many city politicians have failed to show political courage when negotiating CBAs with police unions, and too many state politicians have, likewise, failed to build accountability into public-sector collective bargaining and dispute resolution laws or provide appropriate transparency for disciplinary records.7 There is a sad irony in city and state politicians decrying the lack of accountability of police departments they run and regulate.8

The discussion of reform is rightly focusing on the police union CBAs that provide seemingly insuperable procedural protections to officers accused of wrongdoing.9 Preliminary research from a team of economists at the University of Victoria suggests that the number of civilians killed by police in the United States increased about 60 to 70 per year after police officers gained collective bargaining rights. The lead researcher, Professor Rob Gillezeau, suggests this takeaway: “Police accountability matters and employers, in this case local and regional governments, have failed to bargain in a manner that protects public safety.” One recent review of 178 police union contracts found that they “commonly contain provisions that can insulate frontline officers from accountability and oversight.”10 The same author reviewed police discipline cases and concluded that “the complexity and formidability of the disciplinary appeals process may explain the inability of traditional external legal mechanisms to promote reform in American police departments.” Yet another study from researchers at the University of Chicago found that violent incidents increased approximately forty percent when sheriffs’ deputies in Florida obtained collective bargaining rights. These and other commentators identify a small number of provisions commonly found in police union CBAs that undermine police accountability and at times permit officer misconduct to go unchecked—even when federal consent decrees and settlements mandate reform.

II. The Legislative Proposal

This article proposes federal legislation intended to promptly eliminate such provisions from collective bargaining agreements. Federal legislation is appropriate here as a matter of policy because it appears that police unions will not give up these protections voluntarily, some state and local legislatures and executives will be unable or unwilling to tackle the issue politically, and the problem is widespread and implicates basic rights and liberties. In brief, I propose that police unions should receive tax-exempt status only if they implement certain indispensable accountability measures in their collective bargaining agreements.11 Tax-exempt status is important to unions: because union dues, donations, and investment income would otherwise be taxable income,12 losing tax-exempt status would dramatically shrink unions’ operating budgets.

The last time Congress limited eligibility for tax-exempt status to achieve a civil rights objective was (to my knowledge) 1976, when Congress adopted § 501(i) of the Internal Revenue Code to prohibit racial discrimination by § 501(c)(7) social clubs.13 The Senate report said: “In view of national policy, it is believed that it is inappropriate for a social club or similar organization described in section 501(c)(7) to be exempt from income taxation if its written policy is to discriminate on account of race, color or religion.”14 Similarly, it is inappropriate for a union serving public employee members to be exempt from income taxation if it contracts with public entities in a manner that impedes accountability and frustrates effective supervision, prompt and binding discipline, and transparency.15 Countless federal court cases have affirmed this national policy favoring effective police discipline. For example, courts routinely reject challenges by officers to departmental regulations or disciplinary decisions in deference to the police department’s duty to maintain discipline. Courts also have held municipalities civilly liable for failing to maintain discipline in their police forces.16

Tax policy also justifies my proposal. The rationale for providing tax-exempt status to labor unions is that unions increase utility by overcoming collective action problems, enhance democracy by giving less powerful interests a more equal voice, and help secure public goods (nonexcludable and nonrivalrous goods), such as improved working conditions and standard wages, for the mutual benefit of their members. But CBAs that undermine effective police discipline harm the public and union members alike. And while there may be reasons to subsidize police unions with tax-exempt status, there is no reason to subsidize unions that undermine accountability, discipline, and transparency in policing.

Thus, Congress should adopt legislation providing that no police union shall be exempt from taxation under § 501(c) of the Internal Revenue Code unless any collective bargaining agreement to which the union is a party satisfies certain statutorily-defined core requirements of accountability and transparency, such as the following:17

1. Officers may be interviewed by police department supervisors and prosecutors at any time following any use of force.
2. Officers may be charged with knowingly making a false statement without regard to whether the officer was provided an opportunity to review video or audio evidence before making the statement.
3. Records of complaints and discipline may not be expunged from officers’ personnel files unless found meritless on the basis of competent evidence and all retained records must be made available to the public upon a showing that access is reasonably necessary to serve a lawful purpose (and subject to appropriate restrictions in the interest of the privacy and safety of the officer and any victims, witnesses, or other third parties).18
4. Complaints of misconduct may not be dismissed or disregarded solely on account of the passage of time or because they were made anonymously or without a signed affidavit, and the complainant’s name need not be revealed to the officer before charges are brought.
5. The union will not accept as members or knowingly provide any services or benefits to any officers who have been terminated for gross misconduct by any law enforcement agency.19

The legislation would authorize the U.S. attorney general to add or remove requirements, pursuant to the Administrative Procedure Act, if he or she finds, after gathering and considering relevant data and recommendations from police departments, police unions, and independent police experts (and other interested persons), that the CBA provision inhibits (or does not inhibit) the prompt, effective, and fair administration of screening, training, investigation, discipline, and termination of law enforcement officers in a manner that upholds public order and safety.20

Furthermore, any police union found by the attorney general, after considering relevant evidence, to have knowingly interfered (by act or omission, including by means otherwise permitted by the CBA) with any provision of any settlement agreement or consent decree relating to any alleged pattern or practice of violating federal law would have its tax-exempt status revoked. This change gives police unions an incentive to voluntarily amend or waive provisions of collective bargaining agreements that conflict with or impede the requirements of civil rights settlements or consent decrees.21

These proposed requirements on tax-exempt status might be enough to convince police unions to reform their own CBAs by striking or waiving some key procedural barriers to accountability. But the police unions will need to do it themselves, in coordination with the stakeholders they serve.22 Lasting reform requires police union (and officer) buy-in; the new provisions cannot be seen as yet another “bad idea dreamed up by civilians.”23 This proposal does not endeavor to micromanage the disciplinary process, but it also does not leave disciplinary reform solely to the local political process, which is sometimes unable to overcome police union opposition.24

Any covered union seeking tax-exempt status must file, along with its Form 1024 application for recognition as tax-exempt, and annually thereafter with its Form 990, a certification that (1) the union’s current contract complies with the requirements and (2) the union’s current contract is consistent with any judicial settlement agreement or consent decree relating to any alleged pattern or practice of violating federal law (a) entered prior to the effective date of the union contract and (b) in effect during any part of the period of the union contract. Because the IRS might not have the requisite expertise to determine whether a police union’s contract satisfies the proposed new requirements, the new legislation would grant concurrent investigative and enforcement authority with respect to these matters to the attorney general. The legislation would also direct the attorney general to issue guidance on CBAs and best practices for cooperating with settlements and consent decrees—and perhaps work with representatives of local governments, police departments, and police unions to develop a set of model CBA provisions, adoption of which might result in an administrative presumption of compliance.

Finally, the proposed legislation is not intended to be retroactive. Rather, it would provide police unions a one-year grace period, during which time they can seek to renegotiate CBAs with local governments and submit their certifications of compliance. Of course, police unions might choose not to renegotiate their contracts and forfeit their 501(c) tax-exempt status. Those unions would likely be unable to survive the resulting blow to their operating budgets and might be replaced by different unions—perhaps introducing a new era of competition in union representation for police officers, and possibly starting a race to the top.25 Officers and the public interest alike would be better served by unions willing to embrace accountability.

III. The Policy

This proposal has two particular advantages. First, it encourages reasonably prompt police union alignment with widely accepted police reforms that may help protect civil rights. Currently, police unions have a natural incentive to secure maximal procedural protections for their members; yet such protections unfortunately appear to have become major impediments to police accountability. With these new requirements in place, however, police unions would have a compelling incentive to institute reform in order to obtain and maintain their tax-exempt status. Viewed objectively, and in comparison to due process rights existing elsewhere in our legal system, officers would continue to enjoy robust means of defending themselves against allegations of misconduct—a necessary premise of fair police reform, as the president of the Fraternal Order of Police has insisted. Second, this proposal gives the Department of Justice new ways to (a) promote national standards and best practices in the domain of police accountability and (b) monitor whether existing CBAs promote police accountability and whether union activities align with the civil rights commitments of the communities served by their officer-members.26

This proposal does not involve excessive federal intervention into local police departments or burdensome protocol or bureaucracy for officers on the street. And this proposal (for better or worse) is not premised upon any of the popular ideologies or metanarratives concerning police violence in America. Rather, the premise is an observation about the role of CBAs in the breakdown of police accountability, which has been voiced on the right and the left. Heather Mac Donald of the Manhattan Institute gave testimony on June 10, 2020, to the House Committee on the Judiciary, observing that “[s]ome powerful unions place too many roadblocks in the way of firing incompetent or abusive cops.”27 Former President Barack Obama identified “collective bargaining agreements with police unions” as one of the important targets of local police reform. Hopefully we have not lost the capacity to recognize well-founded agreement and act on it.

Lawmakers at the local, state, and federal levels have many options for addressing the role that CBAs play in impeding accountability and transparency. Local and state legislators could consider whether to amend or repeal so-called “Law Enforcement Officer Bills of Rights,” many of which create (among other things) the hearing and appeals processes available to officers found by departments to have engaged in misconduct.28 State legislators could consider whether to curtail collective bargaining rights for law enforcement employees with respect to matters of discipline.29 Congress could adopt requirements for any federal funding of local police departments, conditioning eligibility for federal funds on procedural reforms complementary to those listed above—as House Democrats proposed in the Justice in Policing Act of 2020 and as Senate Republicans proposed in the JUSTICE Act.30 However, such measures appear less likely to garner prompt bipartisan support and a timely roll-out, and some of them may face lengthy court challenges on vagueness and federalism grounds.31 The Department of Justice can consider ramping up pattern-or-practice investigations and the Collaborative Reform Initiative of the Office of Community Oriented Policing Services, but such efforts will be more effective if police unions have an incentive to take an active part in the most consequential, structural reforms.

IV. Objections and Responses

My proposal is subject to the following objections, each of which I believe can be satisfactorily answered:32

1. The proposal would force police unions to choose between tax-exempt status and procedural protections in CBAs that have admittedly been criticized, but that are no different than protections afforded by the CBAs of other public-sector unions. Why single out police unions?

The integrity of policing is under attack throughout America today. Certain procedural protections have been identified as likely undermining effective discipline and accountability; this proposal might encourage police unions to embrace reform. And if the proposal is adopted, and the reform is successful, then perhaps the attention focused on police union CBAs today might be turned to other public-sector CBAs tomorrow.33

2. The proposal is supposed to garner bipartisan support in Congress. But why would Republicans support increased federal agency intrusion into local, private organizations? Doesn’t this type of proposal invite Congress and executive branch agencies to propose other politicized requirements masquerading as “public safety” regulations to suffocate unpopular but politically effective organizations?

There are four reasons this proposal might be palatable to Republicans, notwithstanding the federal mechanisms. First, this proposal does not touch First Amendment rights—it has nothing to do with religion, speech, press, assembly, or lobbying. Second, there is no question here of federal executive agencies or courts legislating public policy (here, the national interest in effective police discipline); Congress would specifically address the political question implicating eligibility for federal tax-exempt status; the policy would not be extrapolated from the supposed import of terms such as “charitable purpose,” “public benefit,” or the “betterment” of those engaged in labor. Third, it is well established that the government may decline to subsidize an activity that the Constitution prevents the government from penalizing or coercing.34 And, fourth, although federal action may be justified by the widespread failure of state and local politics to address issues implicating basic rights and liberties, this proposal does not police the internal management or activities of the unions; and it leaves the negotiation of new procedures to local governments and police unions. (And, for additional comfort, Congress could also consider a sunset provision on the legislation, allowing, say, ten years for the parties to reset contractual relationships, practices, and expectations.)

3. Why can’t police unions circumvent the proposed regulations by lobbying state and local legislatures to adopt the problematic CBA provisions as part of Law Enforcement Officer Bills of Rights, such that police unions could sign acceptable CBAs and still protect “bad apples” through state and local procedural guarantees?

As briefly noted above, the proposal would require unions that wish to retain their tax-exempt status to waive, in their CBAs, any state or local statutory protections that confer the same procedural protections that would cause their tax-exempt status to be forfeited if contained in their CBAs. The legislation should expressly state the national policy of promoting police accountability by eliminating such procedural protections from CBAs and authorizing unions to waive such state or local protections even if those state or local protections contain anti-waiver provisions. This should suffice for preemption. Unions may waive state statutory rights of their members in a CBA, provided the wavier is clear and unmistakable.35

4. Why can’t police unions set up for-profit affiliates to act as bargaining agents for public-sector contracts and thereby moot the (perhaps too indirect) attempt to achieve substantive CBA reform by regulating the unions’ tax-exempt status?

The answer to this question will depend on state law, but in New York, for example, this is not possible because the Taylor Law provides that union dues (i.e., the union’s revenues that are exempt from federal income taxation) may only be collected on behalf of and paid to the recognized/certified organization serving as the collective bargaining agent. So police unions could set up for-profit affiliates, but those affiliates would not be able to rely on membership dues deduction and transmission, and no other nonprofit union affiliate would have the right to deduction and transmission by the employer because the nonprofit affiliate would not be the recognized/certified organization. It would be possible for unions to set up for-profit bargaining agents and ask union members to voluntarily contribute dues to affiliated nonprofits. But this reintroduces the collective action problems that labor law tried to solve by creating a system of mandatory dues deduction and transmission to the bargaining agents.36 It ignores “the need for coercion implicit in attempts to provide collective goods to large groups.”37 So a for-profit bargaining agent is not a realistic way to sidestep the proposed legislation.

Conclusion

Congress and the Department of Justice should consider this targeted proposal, which builds on empirical scholarship regarding the mechanisms for promoting police accountability and might garner prompt bipartisan support. A new Presidential Commission on Law Enforcement and the Administration of Justice was launched in January 2020. The Republican-sponsored JUSTICE Act, which failed a cloture vote on June 24, 2020, called for establishing a National Criminal Justice Commission, which would, among other things, “develop recommended best practices guidelines to ensure fair and effective policing tactics and procedures that encourage equitable justice, community trust, and law enforcement officer safety,” including “best practices for the hiring, firing, suspension, and discipline of law enforcement officers.” Our government leaders must urgently consider ways to bring about changes in police union CBAs. This proposal is a concrete and attainable way to reach that goal.

• 1Democratic representatives have introduced the Justice in Policing Act of 2020 and have proposed House Resolution 988 (116th Congress, 2019–2020), which condemns “all acts of brutality, racial profiling, and the use of excessive force by law enforcement officers” and calls for various law enforcement policies and reforms. Legislative proposals are discussed in Grace Segers & Caitlin Huey-Burns, Congress crafts police reform legislation in response to police violence, CBS News (June 5, 2020), and Conor Friedersdorf, Police Reform is Popular. Rioting Is Not., Atlantic (June 3, 2020).
• 2For example, the Cato Institute found that “79% of Americans would prefer that an ‘outside law enforcement agency take over the investigation’ when an officer is suspected of criminal wrongdoing.”
• 3See U.S. Commission on Civil Rights, Police Use of Force: An Examination of Modern Policing Practices 58 (Nov. 15, 2018) (“Police union contract provisions can pose another challenge deterring law enforcement officials from holding their officers accountable.”); Lucy Morrow Caldwell, It’s Time to Take On Police Unions, National Review (Dec. 16, 2014); Paige Fernandez, Police Unions Should Never Undermine Constitutional Policing, ACLU (May 15, 2019) (“Time and again, we witness transformative advances on use of force and biased police and civilian oversight, just to have them undermined behind the closed doors of collective bargaining with police unions.”).
• 4This does not mean that better use of force policies should not be adopted.  Some have pointed to Camden, New Jersey, as a success story.
• 5Some of the problems of implementation, and securing lasting change, are discussed in Simone Weichselbaum, Policing the Police: As the Justice Department pushes reform, some changes don’t last, Marshall Project (May 26, 2015).
• 6See, e.g., Stephen Rushin, Police Union Contracts, 66 Duke L.J. 1191, 1198 (2017) (analyzing 178 police union collective bargaining agreements and finding that “a substantial number of these contracts unreasonably interfere with or otherwise limit the effectiveness of mechanisms designed to hold police officers accountable for their actions”), and the other articles cited herein.
• 7See James Surowiecki, Why Are Police Unions Blocking Reform?, New Yorker (Sept. 12, 2016) (“Cities don’t have to concede so much power to police unions. So why do they? Big-city unions have large membership bases and are generous when it comes to campaign contributions. Neither liberals nor conservatives have been keen to challenge the unions’ power. Liberals are generally supportive of public-sector unions; some of the worst police departments in the country are in cities, like Baltimore and Oakland, run by liberal mayors. And though conservatives regularly castigate public-sector unions as parasites, they typically exempt the police.”); Conor Friedersdorf, How Police Unions and Arbitrators Keep Abusive Cops on the Street, Atlantic (Dec. 2, 2014); Ken Girardin, New York lawmakers have sought to weaken police discipline, Empire Center for Public Policy (June 1, 2020).
• 8This dynamic has been observed by, e.g., the Wall Street Journal’s Editorial Board. See The Problem with Police Unions, Wall St. J. (June 10, 2020). See also R. Rainey & H. Otterbein, Local unions defy AFL-CIO in push to oust police unions, Politico (June 30, 2020) (“‘No contract is rammed down the throat of a city or jurisdiction. They signed it, they negotiated it, they agreed to it,’ said Jim Pasco, executive director of the National Fraternal Order of Police.”); R.T. Rybak, I Was Mayor of Minneapolis. I Know Why Police Reforms Fail., Atlantic (June 18, 2020) (“The lack of accountability seems incongruous because the mayors and city councils that negotiate with police unions include some of the country’s most progressive elected officials and represent some of the country’s most progressive constituencies.”).
• 9See D. Belkin, K. Maher & D. Paul, Clout of Minneapolis Police Union Boss Reflects National Trend, Wall St. J. (July 7, 2020); N. Scheiber, F. Stockman & J. Goodman, How Police Unions Became Such Powerful Opponents to Reform Efforts, N.Y. Times (June 6, 2020); Walter Olson, Police Misconduct and ‘Law Enforcement Officers’ Bill of Rights’ Laws, Cato Institute (Apr. 24, 2015) (“Union contracts often add further layers of insulation from discipline.”); Samuel Walker, Institutionalizing Police Accountability Reforms: The Problem of Making Police Reforms Endure, 32 St. Louis U. Pub. L. Rev. 57, 76–77 (2012) (“Collective bargaining agreements, for example, contain provisions related to the investigation of alleged officer misconduct (whether on the basis of a citizen complaint or an internally generated complaint) that impede a timely and thorough investigation.”); Kevin M. Keenan & Samuel Walker, An Impediment to Police Accountability? An Analysis of Statutory Law Enforcement Officers’ Bills of Rights, 186 Pub. Interest L.J. 185, 188 (2005) (“Collective bargaining agreements, for example, contain provisions related to the investigation of alleged officer misconduct (whether on the basis of a citizen complaint or an internally generated complaint) that impede a timely and thorough investigation.”); Samuel Walker, The Neglect of Police Unions: Exploring One of the Most Important Areas of American Policing, 9 Police Practice & Research 95 (2008) (“[T]here are specific provisions in many collective bargaining agreements that inhibit investigations.”); Catherine Fisk & L. Song Richardson, Police Unions, 85 Geo. Wash. L. Rev. 712, 750 (2017) (“The power of police unions to negotiate over terms of employment and disciplinary processes, which is at the core of the collective bargaining process in any unionized workplace, will be essential to consider in any serious approach to police reform.”).
• 10See also Seth Stoughton, The Incidental Regulation of Policing, 98 Minn. L. Rev. 2179, 2212 (2014) (“A more recent investigative report by a newspaper in Florida, a state that lacks a strong civil service regime, found that thousands of officers from agencies across the state retained their jobs even after being arrested or implicated in crimes due to ‘a disciplinary system that has been reshaped in officers’ favor by the state’s politically influential police unions.’ Grievance procedures can undermine supervisory efforts to discipline officers even in light of a clear violation of law or policy.”) (internal brackets and footnote omitted).
• 11On the basis of a survey using the Tax Exempt Organization Search tool on the IRS website, and other publicly-available information, it appears that the collective bargaining representatives of police officers in the twenty-five largest cities that permit collective bargaining (identified in Rushin, 66 Duke L.J. at 1222–23) filed for tax-exempt status either as social welfare associations, under § 501(c)(4) of the Internal Revenue Code, labor organizations, under § 501(c)(5), fraternal organizations, under § 501(c)(8), or voluntary employees’ beneficiary associations, under § 501(c)(9). For this reason, the proposed amendment would be a new section of 26 U.S.C. § 501, rather than an amendment of any of the specific exempt classifications.
• 12See 9 Mertens Law of Fed. Income Tax’n § 34:187 (“An organization that loses its exemption will generally be treated as a corporation for tax purposes and, thus, required to file Form 1120 and taxable on any net income from business or investment activity.”).
• 13See David A. Brennen, Race and Equality Across the Law School Curriculum: The Law of Tax Exemption, 54 J. Legal Educ. 336, 346 (2004) (“From a black perspective, one might wonder why this provision applies only to social clubs. Does this mean that racial discrimination by other tax-exempt organizations is permissible for tax law purposes? Although the public policy doctrine announced by the Supreme Court in Bob Jones University v. United States clearly prohibits racial discrimination by tax-exempt charities, that doctrine applies only to tax-exempt charities. The Bob Jones University public policy doctrine does not apply to the many other tax-exempt organizations described in the tax exemption statute.”).
• 14Id. at 346 n.35 (quoting S. Rep. No. 94-1318, at 8, 1976 U.S.C.C.A.N. 6051, 6058).
• 15Cf. Bob Jones Univ. v. United States, 461 U.S. 574, 591 (1983) (“When the Government grants exemptions or allows deductions all taxpayers are affected; the very fact of the exemption or deduction for the donor means that other taxpayers can be said to be indirect and vicarious ‘donors.’”).
• 16See, e.g., Tindle v. Caudell, 56 F.3d 966, 971 (8th Cir. 1995) (“Because police departments function as paramilitary organizations charged with maintaining public safety and order, they are given more latitude in their decisions regarding discipline and personnel regulations than an ordinary government employer. . . .  The regulations at issue in this case are rationally related to the department’s legitimate interest in developing ‘discipline, esprit de corps, and uniformity’ within its ranks.”) (citation omitted); Locurto v. Giuliani, 447 F.3d 159, 183 (2d Cir. 2006) (“The First Amendment does not require a Government employer to sit idly by while its employees insult those they are hired to serve and protect. Under the circumstances, ‘an individual police officer’s [or firefighter’s] right to express his personal opinions must yield to the public good.’”) (internal citations omitted); O’Donnell v. Barry, 148 F.3d 1126, 1135 (D.C. Cir. 1998) (“First, the District rightly argues, and the district court correctly agreed, that because of the special degree of trust and discipline required in a police force there may be a stronger governmental interest in regulating the speech of police officers than in regulating the speech of other governmental employees.”); Briggs v. Malley, 748 F.2d 715, 719–20 (1st Cir. 1984), aff’d and remanded, 475 U.S. 335 (1986) (“If we cannot demand of our police officers that they recognize when they do not have the authority to make a search or effect an arrest, then we have given up the very idea of a rule of law. The exercise of police power within the law is the very foundation of the social contract.”); Malley v. Briggs, 475 U.S. 335, 353 n.9 (1986) (Powell, J., concurring) (“It is of course true that actions by police must comport with the Constitution. Police departments and prosecutors have an obligation to instill this understanding in officers, and to discipline those found to have violated the Constitution.”); see also Estate of Roman v. City of Newark, 914 F.3d 789, 800 (3d Cir.), cert. denied, 140 S. Ct. 82 (2019), and cert. denied, 140 S. Ct. 97 (2019); Bordanaro v. McLeod, 871 F.2d 1151, 1162 (1st Cir. 1989).
• 17This list draws from but is not the same as the Collective Bargaining Agreement Checklist included in Appendix 6 of the Chicago Police Accountability Task Force Report. See also Leadership Conference on Civil and Human Rights, New Era of Public Safety: A Guide to Fair, Safe, and Effective Community Policing (2019), Recommendation 7.14 (Oppose Provisions that Weaken Accountability Systems When Negotiating Collective Bargaining Agreements).
• 18For nuanced discussion of the legitimate privacy and transparency interests involved, see Kate Levine, Discipline and Policing, 68 Duke L.J. 839 (2019).
• 19Something similar was discussed in Ben Grunwald & John Rappaport, The Wandering Officer, 129 Yale L.J. 1676, 1770–71 (2020) (noting that a version of this requirement was adopted by Connecticut in 2015).
• 20As a matter of legislative drafting and administrative rulemaking, the legislative text and procedure envisioned would be similar to the law concerning the classification of controlled substances, under 21 U.S.C. § 811.
• 21See, e.g., Settlement Agreement, U.S. v. City of Albuquerque, 14 Civ. 1025 (JB) (D.N.M. July 30, 2019) (ECF No. 465, incorporating ECF No. 465-1 at ¶¶ 56, 74) (imposing certain obligations “consistent with any existing collective bargaining agreements”).
• 22There is some reason to think that police officers themselves might support such reforms. As Jonathan M. Smith (former Chief of the Special Litigation Section of the DOJ Civil Rights Division) has argued: “Reform is good for union members — in fact, the overreach of law enforcement bills of rights and some union contracts have harmed the very officers the contract rules are intended to protect. The obstacles to correcting police misconduct have not only undermined confidence in the police, especially among minorities, but have actually placed officers at greater risk by damaging relations between police departments and communities.” Police Unions Must Not Block Reform, N.Y. Times (May 29, 2015).
• 23The proposal described herein is intended to respect the need for local buy-in, while encouraging unions to attend to problematic CBA provisions more urgently than they have to date. See also Stephen Rushin & Allison Garnett, State Labor Law and Federal Police Reform, 51 Ga. L. Rev. 1209, 1226 (2017) (“All of this suggests that unilaterally forcing major revisions to the collective bargaining agreement—a document designed through collaboration—on an unwilling police union may prove unreasonably disruptive and hamper the overall reform process. In many ways, the approach by the DOJ during the Obama Administration made sense. It did not shy away from calling out potentially problematic language in collective bargaining agreements. Nevertheless, it did not seek to use the federal courts to overturn these collective bargaining agreements. Instead, it pressured municipalities to renegotiate these problematic terms when the collective bargaining agreements expired. This seems to strike a reasonable balance.”).
• 24See Rachel Greszler, Confronting Police Abuse Requires Shifting Power From Police Unions, Heritage Foundation (June 9, 2020) (“At a minimum, local and city officials should renegotiate collective bargaining provisions in police contracts to remove provisions that hamstring departments’ ability to enforce accountability and discipline.”). However, Greszler cites a 2017 Reuters report on police union contracts that begins with the story of the unsuccessful attempt in San Antonio to negotiate such provisions out of the police union contract.
• 25There are, of course, barriers and defensive tactics to union competition, but this prospect deserves further consideration.
• 26This proposal falls squarely within the spirit of Executive Order 13896, which recognized that “the Department of Justice has a historically important role in helping to develop, identify, and establish best practices for law enforcement and supporting a range of programs related to the administration of justice.” Congress would need to provide funding for sufficient staff at the Department of Justice to study, promulgate, and enforce the new rules.
• 27See also Rafael A. Mangual, The Limits of Police Reform, Wall St. J. (June 11, 2020); Steven Greenhut, Saving Bad Apples,City J. (Nov. 7, 2013); Daniel DiSalvo, Not Public-Spirited, City J. (June 8, 2020).
• 28See Stephen Rushin, Police Disciplinary Appeals, 167 U. Pa. L. Rev. 545, 553–54 (2019) (suggesting, e.g., that “policymakers could replace arbitrators with democratically accountable actors” or consider “giving arbitrators narrower standards of review or limiting their ability to reduce punishment if the evidence supports the alleged violation.”) (internal citations omitted). See generally Keenan & Walker, supra note 10.
• 29See, e.g., S. Stoughton, J. Noble, & G. Alpert, How to Actually Fix America’s Police, Atlantic (June 3, 2020) (“[U]nions have leveraged the collective-bargaining process to create labyrinthine procedural protections that can make it exceptionally difficult to investigate, discipline, or terminate officers.”). The most widely-reported limitation of collective bargaining rights for public employee unions is Wisconsin’s Act 10, which largely exempts public safety employees. One author recently made a connection between Act 10 and police union reform. Rushin does not reference Act 10, but does argue “that states should amend labor laws to increase transparency and community participation in the development of police disciplinary procedures,” and mentions the possibility that “states could amend labor laws to remove police disciplinary procedures from the list of appropriate subjects for collective bargaining.” Governor Walker also considered extending Act 10 to police and fire unions. In New York City, because there is local “legislation [that] specifically commits police discipline to the discretion of local officials[,] . . . collective bargaining over disciplinary matters is prohibited.” Patrolmen’s Benevolent Ass’n of City of New York, Inc. v. New York State Pub. Employment Relations Bd., 6 N.Y.3d 563, 571–72 (2006); see also id. at 576 (stating that “the public interest in preserving official authority over the police remains powerful.”). But elsewhere in New York, police unions have a right to collectively bargain over disciplinary matters. See, e.g., City of Syracuse v. Syracuse Police Benevolent Ass’n, Inc., No. 6869/2019, 2020 WL 2462111 (Sup. Ct., Onondaga Cty. May 11, 2020).
• 30See generally Congressional Research Service, What Role Might the Federal Government Play in Law Enforcement Reform? (June 1, 2020). For an overview of federal assistance programs, see Executive Office of the President, Review: Federal Support for Local Law Enforcement Equipment Acquisition(Dec. 2014).
• 31A lengthy court battle over this proposed legislative and administrative action is unlikely here, even though it subjects certain organizations to different requirements for tax exempt status, because “statutory classifications are valid if they bear a rational relation to a legitimate governmental purpose.” Regan v. Taxation With Representation of Washington, 461 U.S. 540, 547 (1983) (holding that 501(c)(3) tax-exempt status may be denied an organization using tax-deductible contributions for substantial lobbying activities without violating the First Amendment or Equal Protection Clause).
• 32I am grateful to readers of earlier drafts of this article, who raised different versions of these objections.
• 33In a recent defense of police officer disciplinary rights under CBAs, the President of the Police Benevolent Association of the City of New York stated that, “[i]n the disciplinary realm too, our only demand has been for the due process rights afforded to other civil servants,” but this begs the question as to what procedural rights for civil servants are consistent with public integrity and the common good.
• 34See, e.g., Ysursa v. Pocatello Educ. Ass’n, 555 U.S. 353, 358 (2009) (“[A] legislature’s decision not to subsidize the exercise of a fundamental right does not infringe the right, and thus is not subject to strict scrutiny.”) (internal quotation marks and citation omitted); Regan v. Taxation With Representation of Washington, 461 U.S. 540, 549 (1983) (observing that “tax exemptions and deductions . . . are also a matter of grace that Congress can, of course, disallow as it chooses.”) (internal quotation marks, modifications, and citation omitted); Harris v. McRae, 448 U.S. 297, 318 (1980) (“Whether freedom of choice that is constitutionally protected warrants federal subsidization is a question for Congress to answer, not a matter of constitutional entitlement.”).
• 35See, e.g., Metro. Edison Co. v. NLRB, 460 U.S. 693, 705–06 (1983) (“This Court long has recognized that a union may waive a member’s statutorily protected rights, including his right to strike during the contract term, and his right to refuse to cross a lawful picket line. Such waivers are valid because they rest on the premise of fair representation and presuppose that the selection of the bargaining representative remains free. Waiver should not undermine these premises. Thus a union may bargain away its members’ economic rights, but it may not surrender rights that impair the employees’ choice of their bargaining representative.”) (internal quotation marks and citations omitted); New York City Transit Auth. v. New York State Pub. Employment Relations Bd., 8 N.Y.3d 226, 234 (2007) (holding that “the Weingarten right created by section 75(2) [i.e., the right to have a union representative present with the employee at an investigatory interview, if the employee reasonably believes that the interview might result in disciplinary action], unlike the right given by the Taylor Law to ‘participate in . . . employee organization[s],’ may be surrendered in collective bargaining.”); cf. Spencer v. New York City Transit Auth., No. 95-CV-4779 (JG), 1999 WL 51814, at *10 (E.D.N.Y. Jan. 14, 1999) (“It is firmly established in New York that unions may waive individual employees’ constitutional rights, including due process requirements, through collective bargaining agreements.”).
• 36Mancur Olson, The Logic of Collective Action:  Public Goods and the Theory of Groups 88 (1971) (“A rational worker will not voluntarily contribute to a (large) union providing a collective benefit since he alone would not perceptibly strengthen the union, and since he would get the benefits of any union achievement whether or not he supported the union.”).
• 37Olson acknowledged that collective action problems might be solved, at least in the short term, by strong emotions or ideological motives, but that “some measure of coercion” would generally be necessary. Id. at 87. Alternative solutions to the collective action problems are explored in C. Fisk & M. Malin, After Janus, 107 Cal. L. Rev. 1821 (2019). It would be naïve to suppose that creative lawyers and legislators could not find ways for police unions to both solve the collective action problems and refuse to adhere to the proposed CBA reforms. But these efforts would take time and cost money. And, in the interim, local government officials might find the courage to bargain in the interests of their constituents, and better CBAs might gain acceptance.