chevron-down Created with Sketch Beta.
Urban Lawyer

Good and Bad Ways to Address Police Violence

by Lawrence Rosenthal
Police Badge

Police Badge

Lawrence Rosenthal is a Professor of Law, Chapman University, Dale E. Fowler School of Law. The author formerly served as Deputy Corporation Counsel for Counseling, Appeals, and Legal Policy in the City of Chicago Department of Law. The author is grateful to Blanche Cook, Rachel Harmon, Kate Levine, Mario Mainero, Stephen Rushin, Kate Sachnoff, Lowell Sachnoff, Nirej Sekhon, Seth Stoughton, Jordan Woods, and the participants of the CrimFest conference held at Cardozo School of Law on July 10-11, 2016, for enormously helpful comments on prior drafts, and to Sherry Leysen and the staff of Chapman University’s Law Library for highly capable research assistance.  

[T]here is always a well-known solution to every human problem: neat, plausible, and wrong.

Henry Louis Mencken

Prejudices Second Series 158 (1921)

SCHOLARS HAVE LONG BEEN CONCERNED ABOUT THE PREVALENCE OF UNLAWFUL POLICE VIOLENCE AGAINST CIVILIANS.1 This has been coupled with a concern that police unfairly target racial minorities, particularly African Americans and black people,2 for official scrutiny.3

Consider police shootings of civilians. Statistics are scarce and inconsistent, a reality that inhibits reliable analysis.4 Even so, it appears that the overall rate of police homicide has dropped in recent decades.5 Far more troubling, however, is the mounting  evidence suggesting that those shot or killed by police are  disproportionately persons of color.6 The available data suggest a similar racial skew with respect to police use of nondeadly force.7

Police violence is generally explained in terms of sociological theories contending that coercive governmental authority is most likely to be deployed against individuals perceived by officers to be of inferior status; psychological theories contending that some officers are inclined to utilize violence when confronted with what they regard as oppositional behavior; and organizational theories contending that some police departments develop a culture that tolerates or even promotes violence.8 Whether the explanation is sociological,  psychological, or organizational, the precise mechanism by  which violence comes to be disproportionately used against  minorities is unclear. Perhaps the explanation most frequently  offered in legal scholarship is the prevalence of implicit bias, by which stereotypical perceptions of minorities cause officers to  perceive them as threatening — a phenomenon well documented in psychological literature.9

Racially-skewed policing not only has tangible consequences in terms of the disproportionate use of force against persons of color, but it also contributes to a racial skew in public perceptions: a poll commissioned by a presidential task force found that 83% of whites had a “great deal” or “fair amount” of confidence in the police, while only 63% of Hispanics and 52% of black people shared that view.10 This racial disparity has been reflected in polling data for some time.11

Reform advocates have advanced a wide variety of proposals to address police violence, usually involving more robust criminal, civil, or administrative sanctions against police. The striking thing about this reform agenda, however, is that it is all stick and no carrot. This approach is less likely to produce optimal performance than to reinforce the code of silence — frequently documented in the literature on policing — in which officers fail to acknowledge wrongdoing in order to insulate themselves and their peers from discipline. Equally important, there has  been no effort to assess proposed reforms in light of what we know about the sociology of policing and the ecology of high-crime communities. Utilizing the insights found in the literature on police science, sociology, and criminology and, occasionally,  the author’s experiences as a senior municipal official, this  Article assesses the merits and potential unintended  consequences of the leading reform proposals, and discusses some more promising reforms that have not received scholarly attention.

Part I surveys and assesses the leading proposals for addressing police violence, exploring a number of their potential weaknesses, especially their limited ability to make reliable determinations about whether officers have committed misconduct. The reform agenda, Part I concludes, is likely to fail.

Part II considers the unintended consequences of the reform agenda. In light of the powerful effects of poverty and instability in high-crime urban communities, Part II sets out the case for aggressive police intervention at criminogenic hot spots within those communities. It also considers the likely incentive effects of reforms that offer police all stick and no carrot. In a regime in which the risk of sanctions facing officers who undertake to prevent crime at hot spots increases, the incidence of police misconduct might fall, but there would be a corresponding risk of over-deterrence — often called de-policing — inhibiting effective policing where it is most needed. Such a regime may undermine deterrence of potential offenders as well, by leading them to believe that the police’s ability to intervene effectively in high-crime communities has been circumscribed. The reform agenda, in short, could make things worse.

Part III contends that the most efficacious reforms give officers clear direction about when aggressive patrol is warranted and when officers are most likely to court discipline, using the concepts of safe harbors and their opposite — sure shipwrecks. Officers should receive a measure of deference at criminogenic hot spots, where aggressive tactics are most readily justified, but should face a heightened risk of sanctions when aggressive tactics are employed elsewhere. Disciplinary systems should also encourage candor by offering a measure of safe harbor — precluding the most severe sanctions — when officers provide truthful information, while requiring termination of officers who make false statements. Reforms of this character would also make enforcement priorities more transparent, enhancing the democratic accountability of the police.

I. The Reform Menu

A wide variety of reforms have been proposed to address the problem of police violence. Some rely on civil or criminal liability, others on managerial reform.12 There is, however, reason to doubt their efficacy.

A. Criminal Prosecution

One approach for addressing police misconduct is through the criminal law.13 Both state and federal law make criminal sanctions available for officers guilty of the use of excessive force.

Under state law, officers that utilize excessive force face potential criminal liability for assault or similar crimes, as well as homicide in the case of fatal encounters, though they enjoy a public-authority defense that generally utilizes the traditional criteria that govern justification defenses — the imminence of the danger, the necessity to use force, and the proportionality of the force used.14 Some have also advocated new statutes that would create criminal liability when officers violate internal policies governing the use of force, or otherwise engage in unprofessional behavior.15

Federal criminal liability confronts officers who, acting under color of law, willfully deprive an individual of constitutional rights.16 TheConstitution, moreover, is not silent on police violence. For example, in Tennessee v. Garner,17 the United States Supreme Court, after concluding that “apprehension [of a suspect] by the use of deadly force is a seizure subject to the reasonableness requirement of the Fourth Amendment,” held that “[t]he use of deadly force to prevent the escape of all felony suspects, whatever the circumstances, is constitutionally unreasonable.”18 Instead, deadly force is permitted to effect an arrest “when necessary to prevent [an] escape and the officer has probable cause to believe that the suspect poses a significant threat of death or serious physical injury to the officer or others.”19 More generally, the Fourth Amendment is understood to prohibit the use of unreasonable force, whether deadly or nondeadly, in the course of an arrest or other seizure of an individual.20 It follows that a police officer using constitutionally unreasonable force to effect an arrest faces potential federal criminal liability for the willful deprivation of Fourth Amendment rights.21 And, even absent a “seizure” that triggers Fourth Amendment protection, police violate the Due Process Clause when they undertake the type of “brutal” and “offensive” conduct regarded as “conscience shocking, in a constitutional sense,”22 and therefore could face federal criminal liability.23

Criminal liability for the use of excessive force by police may be available in theory, but in practice it seems to be a rarity. Although precise statistics are unavailable, there has long been widespread agreement that prosecutions and convictions of police for the use of excessive force are infrequent.24 One often-advanced explanation is that prosecutors must cooperate with police in a wide variety of law-enforcement matters, and therefore hesitate to introduce tensions in the prosecutorial-police relationship by bringing criminal cases against officers.25 This state of affairs, however, is not inevitable; as some have suggested, provision could be made for the use of special prosecutors not dependent on their relations with the police in other cases to investigate alleged police misconduct.26 Federal prosecution — under either federal law or by assigning federal prosecutors to handle state-law prosecutions — could also address this problem, to the extent that federal prosecutors are regarded as less dependent than their state counterparts on maintaining good relations with local police.27 To be sure, there is justifiable concern about the scope and clarity of the federal statute used to prosecute police misconduct, but this problem could be addressed by statutory amendment.28

Other problems with the use of criminal liability seem more intractable. Criminal liability, of course, requires proof beyond reasonable doubt.29 That standard poses formidable difficulties. Those who have studied criminal prosecutions for police violence consistently report that there are rarely independent witnesses in cases involving police violence, and because the victims are often criminal suspects or offenders that juries may well find unsympathetic, the available testimony often does not prove sufficiently credible to produce convictions.30 This problem is exacerbated by the consistent observation that officers, whether as suspects or witnesses, are rarely willing to testify in a fashion that will sustain an allegation of police misconduct, instead adhering to an informal code of silence.31 The net result, as Paul Chevigny once put it, is that it is difficult to reach reliable conclusions about allegations of police misconduct: “In the vast majority of cases, there will be no witnesses except the complainant and the police; even when there is more than one police witness present, the code of silence means that the case will be one of the complainant’s word against the police.”32

Beyond that, in the face of a colorable claim that a police officer was justified using force in the face of danger to themselves or others, juries experience an understandable reluctance to second-guess the officer’s judgment.33 As one leading scholar of the police observed: “Officers can always claim that they faced a threat to their own lives and were therefore justified in using force. Judges and juries are extremely deferential to these claims and to the police in general.”34 Juries will often be tempted to sympathize with an officer who undertakes to protect both himself and the public.

The reluctance of juries to second-guess police decisions is likely exacerbated by the very factor thought to produce excessive force disproportionately directed at persons of color — implicit bias. As we have seen, the pervasive influence of racial stereotypes may cause officers to perceive racial minorities as threatening.35 The same phenomenon may also influence juries. The limited empirical evidence available suggests that implicit bias importantly influences jury decision-making.36

One proposal advanced to address the difficulty in assembling proof of police misconduct is to require that officers utilize body cameras to improve the ability to monitor police and assess allegations of misconduct.37 Critics, however, observe that video evidence frequently provides an incomplete depiction of events and is prone to be interpreted in light of the preexisting beliefs and biases of the viewer.38 The available empirical evidence suggests that viewers of video evidence form conclusions about its significance consistent with their preexisting biases, including, in particular, their preexisting beliefs about the police.39 All this suggests that the problem of implicit bias is unlikely to be solved by body cameras. Increasing video surveillance may ameliorate problems of proof, but even a video will often leave room for what jurors may regard, rightly or wrongly, as a reasonable doubt.

One leading scholarly treatment concluded: “[T]he criminal law remains so broad and presents so many enforcement difficulties that it cannot serve meaningfully as the parameters for any professional’s discretion.”40 There is good reason to doubt the likelihood that criminal liability can play a significant role in curbing unlawful police violence.

B. Civil Liability

Since it does not require proof beyond a reasonable doubt, civil liability may hold more promise for combating unlawful police violence. Indeed, exposing those culpable for losses to state-law tort liability for the damages they cause is ordinarily justified as producing an economic incentive to undertake cost-justified measures to avoid inflicting injury on others.41 In addition to potential state-law tort liability, as we have seen, police violence can also violate the United States Constitution,42 and many scholars argue that the damages liability that officials face when they commit constitutional torts deters misconduct in the same manner as ordinary tort liability.43

Nevertheless, there is widespread agreement that civil liability has not proven effective in addressing police misconduct.44 Part of the explanation may be that even with a more favorable burden of proof, plaintiffs still face formidable problems in assembling credible evidence and convincing judges and juries to second-guess the decisions of officers on patrol.45 Even putting this problem aside, however, a variety of others remain. Consider, for example, the limitations the law imposes on the liability of governments and their officials.

As for civil liability under federal law, sovereign immunity bars actions against the United States for constitutional torts committed by federal officials,46 and the Federal Tort Claims Act bars tort claims against the federal government arising out of assault, battery, false imprisonment, and false arrest.47 Individual federal officials can be held liable for their constitutional torts absent “special factors counseling hesitation” based upon “the absence of affirmative action by Congress,”48 but they are entitled to qualified immunity unless their conduct contravened clearly established constitutional law.49 Under this qualified immunity defense, officials are frequently immunized on the view that when they acted, it was not clearly established that their use of force violated the Constitution.50

States are shielded from liability under federal law for constitutional torts by virtue of the Eleventh Amendment, coupled with Congress’s failure to abrogate that immunity by authorizing suits against states.51 State officials may be sued for their constitutional torts,52 but they are also entitled to the defense of qualified immunity.53

Municipalities and other units of local governments can be held liable for the constitutional torts of police officers only when they are committed pursuant to local custom, policy, or the decisions of municipal policymakers,54 and they enjoy absolute immunity from awards of punitive damages.55 Local officials can be sued for constitutional torts, but like federal and state officials, they are entitled to qualified immunity.56

Under state law, state and local governments can be sued for torts committed by their employees, as can the individual employees, but that liability is limited by a wide variety of statutory immunities, which are also generally available to state and local officials sued in tort based on alleged conduct within the scope of their employment.57

Beyond that, whether under state or local law, collective bargaining agreement, or as a matter of employment policy, police officers are usually indemnified for the costs of judgments against them by their employers, whether issuing under federal or state law.58 Labor economics explains the ubiquity of indemnification; employers must offer sufficient compensation to employees to account for the risk of liability that employees face, and indemnification is the most efficient vehicle to minimize the risk that exposure to liability will reduce employees’ productivity.59 For this reason, individual officers generally do not experience financial consequences for their misconduct.

To be sure, indemnification of law-enforcement officials could be curbed or abolished, as some have proposed.60 But, a regime that imposed liability on individual officers would come with its own problems.

If only the limited assets of police officers were available to satisfy judgments, potential plaintiffs and their counsel might have inadequate incentives to bring damages actions for fear that a substantial judgment could not be satisfied from the limited assets of the typical police officer.61 Even more important, imposing damages liability on individual officers would likely produce over-deterrence. Police officers internalize few, if any, of the benefits of effective policing, which are instead externalized to the public at large.62 Thus, if officers are made to internalize the full costs of policing through damages liability, rather than promoting efficient investments in loss prevention, individual liability is likely to induce officers to conclude that it is in their interest to avoid liability-creating behavior.63 Indeed, this explains why indemnification has emerged in order to minimize the risk of over-deterrence and the political costs that it would impose on officials if police became  overly timid.

If sufficient political will existed, however, federal and state statutes could be amended to expand the scope of governmental liability.64  That outcome would effectively shift the full costs of police misconduct onto public employers, avoiding the problematic incentive effects of imposing liability on individual public employees. But, a more robust regime of governmental damages liability would also have problematic incentive effects.

Even under the current liability regime, though many municipalities incur substantial liability costs stemming from police misconduct, there is scant evidence that these costs have incentivized them to undertake reforms.65 The explanation may lie in the difference between the incentives of private and governmental actors. In the private sector, potential defendants have an economic incentive to reduce liability in order to maximize profits or revenues, but Daryl Levinson has argued that because government is not organized to maximize profits or revenues, we cannot be confident that exposure to damages liability will incentivize the government to undertake similar measures.66 The likelihood that the government will undertake cost-justified measures to reduce expected  liabilities is particularly low, he added, when it comes to aggressive policing that may pay political  dividends, since government responds to political and not economic incentives.67 Damages liability may come to be regarded as simply the cost that must be paid for effective policing and its political benefits.68

The argument against the efficacy of governmental damages liability may be overstated. Damages liability likely imposes political costs on government by diverting scarce public resources from what policymakers are likely to regard as their politically optimal uses.69 Liability may also impose political costs on government as litigation brings governmental misconduct to the public’s attention and fixes fault for that misconduct.70 Yet, when, as with policing, the political costs and benefits of  utilizing aggressive or more cautious tactics may be unpredictable, it is difficult to reach reliable conclusions about the incentive effects of governmental exposure to damages.

Equally important, a regime of governmental damages liability sufficiently robust to offer a likelihood of deterrence is also likely to impose significant costs on third parties. Damages awards are generally not paid out of the police budget, but rather out of the larger pool of funds available to meet the expenses of the pertinent unit of local government.71 Some jurisdictions require police budgets to absorb at least some of the cost of litigation payouts, but even then, it is common to budget in ways that limits the financial impact of litigation costs on police operations.72 This should be unsurprising: regardless of the budgeting scheme utilized, it is likely to be politically  unacceptable to deny police departments the resources that police executives deem essential to  protect the public.

Thus, as George Thomas has observed, when it comes to budgeting for police-related liabilities, “the incentives there are mostly perverse.”73 And, in the intense political competition for available public resources, the programs most likely to suffer when resources must be diverted to the payment of judgments are the programs serving those with the least political influence — most likely the poor.74 Thus, not only does governmental liability for police misconduct generate uncertain incentive effects, but rather than forcing a wrongdoer to internalize the costs of its misconduct, governmental damages liability is more likely to impose costs on innocent third parties.

Accordingly, damages liability represents a highly problematic approach to police misconduct. The political costs flowing from the financial burden of civil liability, coupled with the adverse publicity that civil litigation alleging police misconduct can generate, may produce some pressure on government to reduce rates of police misconduct, but that is likely the most that can be said of damages liability.

Another civil remedy remains available: actions for injunctive relief. 75 Injunctive relief effectively shifts some measure of control over the training, supervision, and discipline of police officers to a court. The availability of injunctive relief is circumscribed, however, because a plaintiff seeking to enjoin police policies or practices that produce unlawful violence is generally required to establish a credible and nonspeculative threat that he will be subjected to such unlawful violence in the  future to establish standing to seek injunctive relief.76  This poses a serious obstacle to plaintiffs seeking injunctive relief against police practices.77

Yet, this barrier to injunctive relief can be lowered. Congress has broadened the availability of injunctive relief by authorizing the Attorney General to seek “appropriate equitable and declaratory relief ” against “a pattern or practice of conduct by law enforcement officers . . . that deprives persons of rights, privileges, or immunities secured or protected by the Constitution or laws of the United States.”78 Some have championed this provision as a promising vehicle to achieve reform.79 Some have, moreover, advocated the enactment of parallel state statutes authorizing injunctive relief,80 or the enactment of statutes permitting private plaintiffs to seek similar relief.81 To be sure, caution is warranted; scholars who have studied the existing federal remedy have found that it has produced, at best, mixed results.82

Injunctive relief permits courts to impose reforms that could reduce rates of police violence. To the extent that liability for damages creates pressure on policymakers, it may produce pressure for reform as well. Accordingly, both damages liability and injunctive actions offer some opportunity to induce police managers to reform. It is therefore the subject of managerial reform to which we next turn.

C. Managerial Reform

Most scholars view efforts by police departments to reform from within in order to reduce the rate of unlawful violence as largely ineffective.83 For one thing, efforts to discipline officers are constrained by collective bargaining agreements and statutory provisions that afford police special procedural protections that make disciplinary investigations difficult to undertake, such as a right to representation during investigative interviews and liberal rights to discovery of the evidence to be used against them prior to any disciplinary hearing.84 For another, there is reason to question whether incumbent elected officials and police executives have the will to reform; incumbents may fear that a program of reform represents a tacit acknowledgement of the prevalence of misconduct on their watch.85

Yet, the difficulties in convincing incumbents to embrace managerial reform may not be insurmountable. The prevalence of police misconduct might produce the political will to reform. Beyond that, adopting some form of civilian review or oversight of police departments might also create an independent check on incumbent officials that would enhance accountability.86 Still, in practice, there is little evidence that civilian review has had a meaningful impact on police violence.87 In any event, whether initiated by police executives, civilian review authorities, elected officials, or the courts, there is no shortage of proposals available to the reform-minded.

Some proposals focus on public perceptions in an effort to reduce friction between the police and the community. Along these lines, some have advocated the use of policing strategies and tactics that reflect respect for community members and inclusive decision-making, resting on research suggesting that incorporating norms of procedural justice into the development of policing strategy would enhance the perceived legitimacy of policing.88 One approach that is often embraced involves community-policing models in which policing strategy is developed in consultation with community members.89

In the most disadvantaged communities, however, the difficulties in involving the community in policing strategy are likely to be greatest.90 There is, moreover, little data suggesting that police violence is driven by community residents’ lack of respect for police or policing.91 Beyond that, the extent to which courteous and respectful officers solicitous of the community they police can defuse potential confrontations or enhance the community’s view of the police is unclear.92 Without altering the tactics that give rise to friction between police and community, there is reason to doubt whether incorporating norms of procedural justice is likely to alter the community’s perceptions. One study found, for example, that even when officers were trained to focus on procedural justice values during traffic stops by exhibiting courtesy and respect, black drivers were nevertheless far more likely to view traffic stops as illegitimate than white drivers.93

Whether police violence is a product of sociological, psychological, or organizational factors, if it is driven by implicit or even explicit bias that renders officers all too willing to perceive a threat from persons of color — or the real threats that officers may accurately perceive in high-crime  communities — then increasing the community’s perception of the legitimacy of policing strategy is unlikely to reduce the rate at which officers on patrol perceive threats and respond with violence.94 Indeed, there is no data indicating that incorporating procedural-justice norms into police strategy reduces the rate of police violence.95 And, if disproportionate rates of police violence against persons of color persist, that will erode the community’s perception of legitimacy no matter how much attention police departments pay to procedural justice.

Some advocates of reform call for greater transparency regarding police tactics, such as the collection and dissemination of data on the use of stop-and-frisk tactics,96 or police violence more generally.97 But again, although greater transparency in policing may increase the political pressure to reduce police violence, absent some alteration in strategy and tactics that reduces the likelihood of violent confrontations, it is unclear that any reform focused on the public’s perceptions will reduce the frequency of police violence. And even if reforms aimed at promoting transparency temporarily produce improved public perceptions, the improvement in public confidence will likely prove fleeting if these reforms are ultimately unable to reduce rates of police violence.98

Most scholars of policing focus on reducing police violence through improved training and supervision of officers,99 as well as developing systems that facilitate the identification of at-risk officers.100 Others call for comprehensive systems of licensing for police officers.101 Some advocate de-biasing training to combat the prevalence of implicit bias,102 and others champion the use of de-escalation tactics that call for the use of the least force necessary in any situation.103 Some argue that police should reduce their use of proactive tactics that produce potentially violent confrontations, such as reducing reliance on stop-and-frisk tactics in high-crime areas.104

Some managerial reforms are easier to implement than others. There is little reason to doubt that reforms that merely ask police officers to do less work — such as reducing the rate of stop-and-frisk in high-crime neighborhoods — can be readily implemented. There is, after all, little reason to believe that officers will refuse to abstain from difficult and potentially dangerous work. One study,  for example, found that in response to supervisory directives, New York police officers dramatically reduced the rate at which they stopped suspects.105

When it comes to implementing effective systems to supervise and discipline officers when they intervene in the streetscape in ways that could produce violent confrontations, however, difficulties multiply. De-biasing training sounds promising, for example, but the evidence of its efficacy is, at best, mixed.106 Moreover, psychological research involving simulated shootings has reached no clear conclusion about the extent to which implicit bias actually infects the decisions of officers to shoot in self-defense.107 Indeed, the available data indicates that a firearm is recovered from the victim of police shootings in the clear majority of cases,108 and as Nirej Sekhon has pointed out, the concept of implicit bias “is less helpful in making sense of those cases — the majority — where the victim actually had a firearm.”109

There is, however, an even larger problem looming over reforms that focus on police training and supervision. Scholars have long questioned the extent of supervisory control over police officers on patrol in light of the substantial discretion and necessarily limited scrutiny that officers experience in the field.110 When it comes to the use force against potential threats to officer safety, this problem is likely to be particularly acute. After all, as they perform their duties, officers are unlikely to be indifferent between taking too much and too little effort to protect themselves.111 Officers are likely to resist managerial directives that ask them to refrain from the use of force under circumstances in which they regard force as warranted to protect themselves, even if, as a consequence of implicit bias or other factors, they may be overly inclined to perceive danger.

In the final analysis, accordingly, no matter what policies are adopted, and no matter how officers are trained, there seems to be no substitute for imposing accountability on officers; unenforced policies are likely to become dead letters.112 As David Rudovsky once wrote: “All other remedies will ultimately fail if not accompanied by a system of training, supervision and discipline that is structured to ensure that departmental policies relative to use of force and other restrictions on the arbitrary use of power are implemented and enforced.”113

There are, however, formidable obstacles to creating effective systems of officer accountability. As we have seen, reliable independent witnesses are rarely available to corroborate a charge of misconduct, and police frequently adhere to a code of silence.114 In the roughly fifteen years that I occupied a senior position in the City of Chicago’s Department of Law, I never encountered a case in which a police officer testified in disciplinary proceedings against another officer accused of misconduct.115 The only reliable “tell” was when an officer, asked whether he had witnessed misconduct on the part of another officer, denied seeing anything. That told us two things: the police suspect was guilty, and the police suspect was regarded by his colleagues as a jerk. Even so, these insights supplied no evidence usable in disciplinary proceedings.

All this makes reliable fact-finding enormously difficult. The increasing use of body-mounted cameras and other forms of video surveillance may ameliorate problems of proof, but we have seen that cameras are no panacea.116 Indeed, the evidence of their effect on police violence to date is mixed.117 Nothing in the reform menu, however, involves a mechanism to improve ability of adjudicators in the disciplinary process to reliably resolve disputes of fact between police officers and complainants.118 Without a reliable means available to determine whether an officer has engaged in misconduct, any effort to hold officers accountable for deviations from use-of-force policy is likely to fail.

One might try to counter these difficulties by adopting a more skeptical attitude toward the accounts of police officers in disciplinary investigations and adjudications.119 It is unclear, however, whether this approach would reduce the overall error rate, or merely increase the proportion of errors that represent false findings of guilt as opposed to false exonerations. Even worse, when officers are asked to continue to intervene in the streetscape in contexts that can lead to violent confrontations, but with a heightened risk that those interventions will lead to sanctions, the law of unintended consequences may rear its ugly head. It is to that thorny problem that we turn next.

II. The Perils of Reform

One could imagine a strengthened regime of criminal or civil liability, or of administrative discipline, sufficiently robust to represent a credible deterrent to the use of excessive force.120 For example, despite the difficulties of obtaining a conviction by proof beyond reasonable doubt, aggressive prosecutors who credibly threaten to bring criminal charges might act as a real deterrent, even if acquittal rates are high. After all, no one likes to be subjected to a criminal prosecution, even if it proves ultimately unsuccessful. A more robust regime of civil liability that narrows defenses and limits immunity and indemnification could produce the same result, as could an aggressive program of supervisory monitoring and discipline. Yet, although more robust regimes of criminal, civil, or administrative sanctions could dramatically reduce the rate at which officers use unlawful force, they could also dramatically reduce the effectiveness of policing. Assessing this possibility requires consideration of the unusually high rates of violent crime in impoverished communities, the case for aggressive policing in which officers proactively intervene at  criminogenic hot spots of crime, and, finally, the incentive effects of a regime that substantially increases the likelihood of sanctions police officers face. It is to these fraught subjects that Part II is addressed.

A. The Racial Skew in Violent Crime

As we have seen, black people are disproportionately victimized by police violence.121 Black people, however, are also disproportionately victimized by violent crime, all too often at the hands of other black people.

Consider the racial skew reflected in homicide statistics. Homicide rates over the past half-century are something of a roller coaster. The homicide victimization rate per 100,000 population in the United States rose from 4.6 in 1962 to a peak of 10.2 in 1980, fell to 7.9 by 1984, subsequently peaked once more at 9.8 in 1991, only to fall sharply thereafter, reaching 4.8 by 2010.122 These aggregate figures, however, obscure the concentration of homicide victimization among urban black people.

Throughout the period 1980 - 2008, the homicide victimization rate among black people was 27.8 per 100,000 population, about six times higher than the white rate of 4.5.123 The racial gap in homicide victimization was persistent throughout this period, though with far more variability in the black rate; the homicide victimization rate for black people peaked at 39.4 per 100,000 population in 1991, fell to around 20 in 1999, and stabilized around that rate for the next decade, while the white victimization rate remained well under 10 per 100,000 population throughout the period.124

As for offending rates, the black homicide offending rate of 34.4 per 100,000 population during this period was almost eight times the white rate of 4.5 per 100,000 population.125 The racial disparity in offending rates was persistent, though, again, with far greater variability in the black rate; the homicide offending rate for black people peaked at 51.1 per 100,000 population in 1991, declined  to 24 in 2004, and then fluctuated between 28.4 in 2006 and 24.7 in 2008, while the white offending rate remained below 10 throughout the period, declining to 3.4 per 100,000 population in 2008.126

Moreover, homicide seems to be disproportionately intra-racial; from 1980 through 2008, 84% of white victims were killed by whites and 93% of black victims were killed by black people.127 In this period, 47.7% of homicide victims and 52.5% of homicide offenders were black.128

The racial disparity in homicide is striking and persistent; in 2012, for example, the black homicide victimization rate was 18.03 per 100,000 population, while the white rate was 2.65.129 In 2015, the most recent year for which data is available, the white homicide victimization rate was 2.36 and the white homicide offending rate was 1.87 per 100,000 population, while the black homicide victimization rate was 16.51 and the black offending rate was 13.18 per 100,000 population.130

Homicide is also concentrated in urban areas; in the period 1980-2008, homicide victimization rates were highest in large cities.131 Indeed, the crime spike in the 1980s and early ‘90s almost exclusively involved urban crime; during this period, homicide rates were essentially flat in cities with populations below 250,000, while the rise and subsequent fall in homicide concentrated in cities with populations exceeding 1,000,000.132 Even within cities, crime rates reflect the kind of racial disparity we have seen in national homicide rates; a random sample of large cities in 2000 found that violent crime rates were five times higher in predominantly African American neighborhoods than in predominantly white neighborhoods.133

It is not race but poverty and a variety of other socioeconomic factors that explain the racial skew in rates of violent crime. Poverty, for example, has long been understood to be a major contributor to crime.134 That explains a good deal of the apparent relationship between race and crime; black people have experienced consistently higher rates of poverty than whites.135

Poverty, however, is not the only factor that drives crime rates. Although they differ in details, there is widespread agreement among scholars that the racial skew in crime rates reflects the concentration of crime in communities with high rates of not only poverty, but also racial isolation and indicia of community instability, and black people disproportionately reside in such areas.136 Moreover, in light of the geographic concentration of crime in disadvantaged communities with segregated minority populations, it should come as little surprise that both minorities and the poor express greater concern about violent crime than do white people and the wealthy.137

A variety of theories have been advanced to explain the relationship between these socioeconomic factors and crime. Some argue that the persistence of racial discrimination creates a perception in disadvantaged black populations that legitimate avenues of social mobility are unavailable.138 Segregation is also said to loosen the connection of black residents of impoverished communities to urban labor markets and thereby undermine norms of law-abidingness.139 Others contend that because poverty produces crime, persistently high rates of racial segregation concentrate  impoverished blacks in communities that suffer high rates of violent crime.140 Another view posits that neighborhoods with concentrated disadvantage lack sufficient collective efficacy to exert mechanisms of social control over potential lawbreakers.141 Still others argue that residents of  disadvantaged neighborhoods become isolated from social norms and come to see violence as a legitimate form of self-help and a means for dispute resolution.142

Whatever the precise mechanism that produces differential rates of both offending and victimization, given that rates of offending and victimization are not racially neutral, many complexities arise when assessing whether evidence that police disproportionately target persons of color for investigation or arrest suggests that they are engaged in discrimination on the basis of race.143 Some argue, for example, that such evidence may reflect no more than a rational police response to differential rates of offending by race.144 Some studies endeavor to account for differential offending rates by race, and still find evidence of racial discrimination.145 Others, however, still question the probative weight of the statistical evidence of racial discrimination in policing.146

Regardless of what one makes of the debate over the statistical evidence, it seems clear that police violence represents a small part of the threat of violent victimization facing black people. In 2015, for example, the Federal Bureau of Investigation reported 13,455 total homicides in the United States, including 7,039 black victims.147 For the same year, The Guardian’s database of police killings indicated that there were a total of 1,146 police homicides, with 306 black victims.148 The Washington Post’s database of police shootings (not all police homicides) reflects 990 fatal police shootings in 2015, with 258 black victims.149 It seems apparent that the vast majority of black homicide victims are not killed by police.

In any event, the data reflecting the distribution of violent crime amply demonstrate that segregated, unstable, high-crime communities pose exceptionally difficult challenges for law enforcement. This leads to a consideration of the efforts contemporary policing has made to respond to those challenges.

B. The Case for Hot Spot Policing

As we have seen, there was a dramatic spike in violent crime occurring in the late 1980s and early 1990s, concentrated among urban black people.150 There is something approaching a consensus among criminologists that the crime spike was a function of the introduction of crack cocaine into major cities.151 A number of studies have demonstrated that lagged increases in violent crime followed the introduction of crack to an urban area.152 Although there are fewer studies that reliably measure the proportion of violent crime that is drug-related, the relationship between the competition in the illicit drug market and violence is well documented in the literature on urban crime.153

One organization particularly well suited to the type of violent competition that accompanied the emergence of crack during the crime-rise period is the criminal street gang; indeed, one consistent observation in the scholarly literature about gangs is their heavy involvement in drug distribution.154 Surveys of youth gang members reflect the involvement of gang youth in drug trafficking at much higher rates than other youth.155 Ethnographic research on gang crime reports that gangs endeavor to organize drug markets to maximize the economic benefits of drug dealing while using the threat of violence to inhibit competition.156 “Research has consistently documented that violence driven by conflicts within and among gangs, drug-selling crews and other criminally active groups generate the bulk of urban homicide problems.”157

The need to control identifiable turf in order to limit competition, as well as the prevalence of open-air drug markets, necessitates the use of violence and intimidation tactics.158 Gangs could hardly run open-air drug markets or obtain effective control over a neighborhood unless they were able to intimidate law-abiding community residents, rendering them unlikely to complain to the police or testify in court.159 The literature confirms the prevalence of gang intimidation as a means of inhibiting community cooperation with the authorities.160 Unsurprisingly, “offenders in gang-related and drug-related homicides are much less likely to be arrested . . . in part due to lack of witness cooperation.”161

In this violent urban context, firearms are pervasive. There is considerable evidence that members of criminal street gangs carry firearms at elevated rates.162 The same is true of those involved in drug trafficking.163 This should not be surprising; those engaged in unlawful but intensively competitive enterprises will often turn to violence as a means of enhancing their position in illegal, if lucrative, markets. The prevalence of violent competition, in turn, is likely to increase the rate at which offenders carry firearms.

Gang researchers have found that the prevalence of violence in gang-dominated neighborhoods serves to make firearms more pervasive in those communities.164 Researchers have similarly found that a perception of danger in high-crime neighborhoods becomes a stimulus for the carrying of firearms as a means of self-protection.165 Jeffrey Fagan and Deanna Wilkinson have labeled this phenomenon an “ecology of danger” in which the need to carry firearms and be prepared to use them came to be seen as essential.166

Despite the vast array of forces edging crime rates higher in the late 1980s and early 1990s, there was a subsequent dramatic decline in crime. Given that the crime spike seems to have been driven by violent competition in drug markets spurred by the demand for crack cocaine, perhaps the explanation for the subsequent crime drop is a reduction in drug-related crime. Some have advanced this argument, claiming that violent competition declined as drug trafficking became less profitable, and pointing to surveys of arrestees and ethnographic research suggesting that drug use declined during the crime-drop period.167 Yet, even granting this supposition, it is far from clear that drug trafficking declined as a result of a diminution in the demand for illegal drugs. Although it is difficult to measure the demand for cocaine, for example, the available statistics suggest no dramatic reduction during the crime-drop era.168 Indeed, cocaine-related emergency room admissions actually rose from 1994 to 2001, as did the proportion of emergency room admissions that involved crack.169 Trends in the price of crack were also not noticeably different during the crime-rise and crime-decline periods.170

In fact, there is ample reason to believe that the crime decline reflects the effects of policing. There is, for example, considerable evidence of a statistical relationship between increases in the number of police and subsequent decreases in violent crime.171 But, this explanation is, at best, incomplete. There are a great many studies regarding the effect of increased numbers of police or frequency of patrol, and their results are, at best, mixed.172 At the city level, the relationship between numbers of police and crime breaks down; studies of large cities find little relationship between the size of the police force and the magnitude and duration of the crime spike and subsequent decline.173

Nor does it stand to reason that increased policing alone is likely to reduce crime, regardless of the tactics employed. Officers who merely drive through a neighborhood on patrol are unlikely to be very effective at disrupting drug or gang activity — a gang or drug dealer with a modicum of sophistication need only post a lookout who can warn his confederates to cease any overt criminality as the squad car drives past. As Mark Moore once observed: “Police on patrol cannot see enough to intervene very often in the life of the community.”174 This rings true to my experience — when the Chicago police focused on increased vehicular patrol, the most obvious result seemed to be an increase in number of lookouts posted by street gangs. It seems likely that the tactics police use must be at least as important as the number of officers. Effective tactics, in turn, must be attentive to the relationship between crime and place.

The concentration of the crime spike in cities suggests that urban crime poses special challenges. Even in urban areas, however, crime is not randomly distributed. Crime tends to cluster in discrete geographic areas and is relatively stable within them.175 As we have seen, areas of concentrated disadvantage that experience high rates of racial isolation and social instability experience particularly high crime rates.176 The geographic “lumpiness” of crime, however, is not solely a function of community. Even within what are considered high-crime communities, crime tends to cluster in discrete places.177 This phenomenon is explained by the “routine activities” school of criminology, which posits that crime rates will be highest where three things intersect: motivated offenders, desirable targets, and a lack of guardians, such as parents, friends, neighbors, private security guards, or police, all of whom occupy space in a manner that deters crime.178

All of this suggests the need for a proactive policing strategy to combat crime effectively. Areas at which motivated offenders, desirable targets, and a lack of guardians intersect are not randomly distributed; and it is those locations at which the police are most needed to provide “guardianship” that would otherwise not be present.

In fact, the 1990s saw alterations in the tactics employed by a  great many urban police departments that moved from more passive and reactive patrol to proactive efforts at intensive and aggressive patrol of specific high-crime areas.179 Particularly impressive crime reductions occurred in New York City which, between 1991 and 2009, experienced the broadest crime declines of any major American city.180 In 1991, the size of New York’s police force began to increase, and subsequently, after the appointment of a new police chief,  the department placed greater emphasis on aggressive stop-and-frisk tactics for misdemeanor arrests for drug and public-order offenses and adopted a system of statistical analysis labeled “Compstat” that directed enforcement efforts at statistical “hot spots” of criminal activity, and imposed greater managerial accountability.181 A number of studies have concluded that this alteration in policing tactics surrounding the adoption of Compstat deserves substantial credit for New York’s crime decline.182

Few national studies have endeavored to examine the effect of changing police tactics on crime rates. There are many difficulties in conducting such an analysis — efforts to determine what tactics are most prevalent in each city, when they became prevalent, and how effectively they were employed are fraught with peril. Even so, the leading analysis found that the introduction of Compstat tactics in major cities was the only law enforcement tactic that had a demonstrable relationship to subsequent reductions in crime.183

Beyond that, an impressive number of studies throughout the nation have found that aggressive policing at hot spots, defined in what is likely the leading meta-analysis as geographically-targeted “problem-oriented policing, focused drug enforcement, increased patrol, increased gun searches and seizures, and zero-tolerance policing to control high-activity crime places,”184 produces persistent and significant crime reductions with limited displacement of crime to other locations.185 Indeed, comprehensive reviews of policing strategies have found that hot-spot policing is the only police intervention that consistently drives down crime.186 Moreover, although the available data is limited, the studies to examine the question have found that, despite concerns that hot-spot policing will increase friction between police and community, “community members had positive experiences when subjected to hot spots policing initiatives,” with “no significant impacts on fear of crime, police legitimacy, collective efficacy, or perceptions of crime or social disorder.”187

The apparent success of hot-spot policing reflects the case for proactive policing. Aggressive patrol targeting criminogenic hot spots is a mechanism for adding guardianship at locations at which motivated offenders and vulnerable victims intersect. When police do not merely drive by hot spots, but instead intervene on the streetscape, stopping and frisking those likely to be carrying guns or drugs, they perceptibly enhance the risks facing offenders at hot spots — especially those carrying guns or drugs.188

Criminology is hardly an exact science, and most criminological theories likely should be regarded as provisional.189 Still, there is a serious case to be made for the efficacy of hot-spot policing. That increasingly well-supported hypothesis, in turn, leads to a consideration of the problem of over-deterrence, or de-policing.

C. The Threat of De-Policing

The evidence that urban crime is best addressed by aggressive policing targeting criminogenic hot spots creates something of a dilemma for urban police. These tactics require police to intervene at locations where suspects are most likely to be armed and involved in unlawful activities in order to increase the risks facing offenders. For this reason, the policing tactics most likely to be effective are also most likely to produce violent confrontations between police and civilians.

As we have seen, in high-crime urban environments, offenders carry guns and drugs on the streetscape at elevated rates.190 Accordingly, officers involved in hot-spot policing operate in the face of an elevated threat of violence. This elevated threat, in turn, could well make officers more likely to perceive a need to defend themselves. Indeed, the existing research suggests that police violence is driven by the extent to which police regard themselves as confronted with danger.

The studies that have examined the issue have found that a variety of proxies for the perceived level of threat to officers is related to the rate of police violence; the available data indicates, for example, that the rate of violent crime is an important determinant of the rate of police shootings.191 Police  shootings are, also, usually a response to a particularized threat; the available data consistently indicate that officer-perceived threats of assault are involved in the clear majority of officer-involved shootings.192 Moreover, we have seen that in most police shootings, the available data indicate that the suspect was armed.193 The data also suggest that police violence is also associated with patrol: the vast majority of police shootings occur in areas open to the public.194 The research also shows that police are more likely to use force when they encounter resistance.195 There is also research suggesting that proxies for a perception of racial threat, such black-on-white homicide rates or shootings of officers by black suspects, increase rates of police homicides.196 Indeed, it may well be that factors that create a perception of racial threat aggravate implicit bias. In any event, it seems likely that officers who believe that they work in dangerous conditions will be quicker to perceive a need to use force in self-defense.197

Accordingly, the data powerfully suggest that police perceive and respond to indicia of threat in the environment they confront — whether real or, as the literature on implicit bias suggests, a product of racially skewed psychology. This should be unsurprising; a rationally self-interested officer’s behavior in the field will surely be driven by the extent to which he believes himself to be confronted by threats to his safety.198 If anything, as we have seen, there is reason to believe that officers are likely to err on the side of protecting themselves against perceived threats.199

Although no study has undertaken the difficult task of calculating rates of police violence at hot spots relative to other locations, the ample data linking rates of police violence to rates of real or perceived community violence powerfully suggest hot-spot policing presents heightened risk of violent police-civilian encounters. Precisely because hot spots are foci of criminal activity and potential violence, officers asked to intervene at hot spots have reason to be concerned for their safety. Hot-spot policing, after all, does not work merely because data analysts identify a particular location while in the safety of police headquarters. Hot-spot policing can impact on potential offenders only if officers appear at hot spots and alter the calculus of potential offenders. An officer tasked with intervening at a hot spot, however, has reason to believe that he confronts a heightened risk of violence. For just this reason, hot-spot policing confronts police with the type of threat to their safety — whether real or perceived — that research suggests drives rates of police violence.

It follows that officers assigned to hot spots will perceive a heightened need to protect themselves against potential threats. When officers perceive that their ability to respond to perceived threats is circumscribed by a threat of post-hoc review and sanctions, however, they are likely endeavor to minimize the extent to which they encounter such threats.200

Rationally self-interested officers are likely to consider the potential for sanctions when they intervene on the streetscape. As one scholarly assessment observed: “The existence of police departmental review mechanisms is often on the minds of police officers on the street when they make discretionary judgments.”201 Even if officers are confident of their professional judgment, whenever they place themselves in situations that may produce after-the-fact review of their conduct, there is some risk of sanction and some risk of error, as well, both by the officers in the stress of the moment, and by investigators and adjudicators who enjoy the benefit of hindsight in judging officer’s actions, and who may fear political backlash if they exonerate officers. Accordingly, if officers on patrol calculate that the likelihood that they will be sanctioned for undertaking to protect themselves against potential dangers as they intervene in the streetscape is sufficiently high, they are likely to be inhibited from doing so.

This is yet another consequence of a regime in which police are made to internalize the costs of their activities, without internalizing corresponding benefits.202 After all, the compensation of officers is not linked to the crime rate or some other measure of efficacy; instead, civil service provisions and collective bargaining agreements usually limit supervisory control over police compensation and discipline.203 Accordingly, a regime that simply exposes officers to an enhanced risk of sanctions when they intervene in the streetscape likely biases officers toward inaction.204 The predictable outcome of a system of asymmetrical risk is over-deterrence, as one might expect in any system in which an individual faces mostly sticks and few carrots.205

The resulting phenomenon, in which police endeavor to avoid undertaking enforcement, especially in high-crime communities where the risks are greatest, has been dubbed “de-policing.”206 Although de-policing has not often been studied, there is anecdotal evidence of its appearance in the wake of scandals that led police to believe they would be subject to greater and more skeptical public scrutiny,207 and a number of studies have found increases in crime rates following events that led police to believe their use of force would be more heavily scrutinized.208 This result is likely attributable to over-deterrence, a possibility also suggested by evidence that video surveillance causes officers to become more risk-averse and less inclined to undertake proactive enforcement.209

Conversely, some have argued that police violence is likely to produce increased crime by undermining the community’s confidence and willingness to cooperate with police, thereby compromising police efficacy.210 One study found, for example, that after media reports of apparently unjustified police violence in Milwaukee, police-related 911 calls from community members sharply dropped, with the declines concentrated in predominantly black communities.211 The authors of the Milwaukee study, moreover, observed that homicide increased in Milwaukee in the six months following a published report of an apparently unjustified police beating that proved particularly controversial.212

Whether the Milwaukee homicide spike reflects correlation or causation is uncertain. After all, if offenders become aware that community residents are less likely to call police, the mechanism by which a homicide spike would result is unclear. Even if potential offenders perceive that community calls to 911 are in decline, it is doubtful that this would cause them to believe that homicides will go uninvestigated or unsolved in light of the severity of the crime, the virtual certainty that the victim’s body and other attendant circumstances will come to the attention of authorities, and the motivation of a victim’s friends and family to assist the authorities.213

Given the multiple potential causes for a crime spike in the wake of reports of police violence that produce an anti-police backlash, it is likely impossible to determine whether the homicide spike in Milwaukee was produced by reduced community cooperation with police, de-policing in the face of skeptical public scrutiny, potential offenders’ perception that the police are likely to be inhibited by that scrutiny, or some combination of these and other factors. In light of the many uncertainties regarding the response of both police and offenders to increased public scrutiny of policing, however, the possibility that police efficacy will be compromised when officers perceive that they face an increased risk of sanctions should not be casually dismissed.

The available evidence, moreover, likely understates the problem, since it is not linked to reforms that create new or more potent sanctions facing individual officers on patrol, but instead is based on studies involving the adoption of new use-of-force policies or events likely to trigger greater public scrutiny of policing.214 Adoption and vigorous implementation of the types of reforms considered in Part I above that make it easier to impose civil or criminal liability or internal discipline on officers would likely cause officers to perceive far greater risks in undertaking aggressive enforcement than were present in any of the studies performed to date, and cause potential offenders to perceive far greater restraints on police, leading to concomitantly greater increases in crime rates.

Because governments respond to political incentives, de-policing could be expected when public opinion swings against the police in the wake of perceived misconduct.215 From this, one might argue that de-policing is not of great concern because the political benefits of effective law enforcement will cause the elected officials to undertake to combat de-policing when the public’s concerns shift in the face of increased crime.216 For a number of reasons, however, this view may indulge an unrealistic assumption about the efficacy of the political checks on de-policing.

At the outset, given that police violence as well as the consequences of de-policing are most likely to be experienced in impoverished and unstable communities, the residents of those communities may lack the political influence necessary to ensure that elected officials properly calibrate the level of  policing undertaken in high-crime communities. Even in the face of political pressure to stem the effects of de-policing, however, there are limits to managerial control of police.

As we have seen, managers have limited ability to control the conduct of officers in the field, and are likely to have difficulty if they endeavor to require officers to undertake activities that they believe will either endanger themselves or lead to potentially career-ending sanctions.217 Emerging  technology, we have also seen, is no panacea and may itself promote de-policing.218 Elected officials may want proactive policing when crime rises as a consequence of de-policing or otherwise, but unless they properly align the incentives of officers in the field, they are unlikely to get the desired response.219

Perhaps most important, designing incentives to combat overdeterrence presents great difficulty. Even if public employers had a complete menu of carrots and sticks available to incentivize police performance, it would be difficult to provide officers with a useful metric for optimal policing. If police compensation were tied to reductions in crime rates, for example, overly aggressive policing would likely be the consequence. Effective policing represents a complex balance between liberty and order. While there may, in theory, be some combination of carrots and sticks that can produce optimal policing, identifying that mix is no easy matter.220 The political pressure to fight crime is therefore more likely to cause elected officials to retreat from reforms in the face of claims that they have inhibited police effectiveness rather than producing some sort of optimal equilibrium. Accordingly, if rising crime rates in the wake of de-policing generate political pressure for more efficacious policing, the most likely outcome would be to undo the reforms that police complain tie their hands. A reform agenda that fails to take account of the threat of de-policing is, in short, unsustainable.

I should add that, in light of my own experience with urban policing, concerns about over- deterrence ring true. I formerly occupied a rather senior position in Chicago’s municipal government. In that capacity, I was involved in many trainings in which we endeavored to prepare the Chicago Police Department to engage in hot-spot policing while apprising officers of the limits of their stop-and-frisk authority under the Fourth Amendment.221 After regaling me with their experiences of the relation between gangs, drugs, and guns, officers uniformly told me that they would not approach individuals they believed to be gang members or drug traffickers unless they could perform a frisk for weapons. Realizing that my elegant disquisition of the subtle distinctions drawn in the case law meant little to officers who feared for their safety, I ultimately agreed that the officers could do what their experience had taught them was necessary to keep safe. That is the reality of policing I the midst of guns, drugs, and violent crime.

To be sure, de-policing might be warranted as a corrective to overenforcement. In March 2013, for example, the New York Police Department issued a new directive requiring officers to provide a  detailed narrative justifying stops, which officers interpreted as an indication that their stops would be scrutinized and potentially sanctioned, leading to a dramatic decline in the rate of stops, while increasing the rate at which weapons were recovered during stops.222 There is reason to believe that by 2013, New York was engaged in over-enforcement; the number of investigative stops by police officers in New York rose from 314,000 in 2004 to 686,000 by 2011.223 As we have seen, New York’s program of aggressive stop-and-frisk had been associated with dramatic reductions in violent crime;224 New York’s success may well have caused it to push stop-and-frisk tactics well beyond the point of diminishing returns.225 In such a case, de-policing may be just what the doctor ordered.

There is no guarantee, however, that reforms that increase the risk that officers will be sanctioned for undertaking enforcement will be implemented only in the face of over-enforcement. To the  contrary, as we have seen, police shootings seem to most frequently occur in areas in which crime rates are also highest, which may explain at least part of the racial skew in police violence.226 Moreover, although there is no reliable test for measuring whether police are engaged in over- or underenforcement, there is widespread agreement among scholars that high-crime minority communities often experience the latter.227

Given that there is no ready means for identifying what amounts to optimal levels of policing, the best we can likely do is to identify criminogenic hot spots, where the case for aggressive patrol is easiest to make. At hot spots, the chance that police could be over-deterred or offenders under-deterred is likely to be greatest. Outside of those hot spots, in contrast, police violence may well be under-deterred by the existing regime, which, as we have seen, often makes it unduly difficult to identify and sanction misconduct.

When de-policing occurs in areas not already experiencing overenforcement, the consequences in terms of increased crime can be dire.228 The consequences are also likely to be no less racially skewed than police violence. As we have seen, violent crime is concentrated in disadvantaged minority neighborhoods.229 Thus, when de-policing produces increased crime rates, the victims are likely to be disproportionately persons of color. The increase in crime that may well follow in the wake of de-policing is also likely to exact a toll in terms of confidence in the police in these same communities. Studies of the matter consistently find that once indicia of perceived crime rates and neighborhood stability and disadvantage are taken into account, the statistical relationship between race and satisfaction with the local police largely, if not entirely, disappears.230 In other words, it is crime and instability that largely drives perceptions of local police, not race. If the police prove ineffective at fighting crime, they can hardly expect to retain the confidence of the community — especially the confidence of those most likely to be victimized by crime.231

The striking thing about the reform menu considered in Part I is that it is virtually all stick and no carrot. There is, however, little reason to believe that a regime of punitive management will produce superior performance rather than excessive deterrence. As Albert Reiss once observed when explaining the pervasive code of silence among police: “Much of the research on human behavior demonstrates that punishment is a poor way to get people to conform to a standard of conduct. More often than not punishment systems are self-defeating, encouraging elaborate procedures to circumvent them.”232 The literature advocating police reform, however, makes little effort to come to grips with the risk of over-deterrence.

Consider, for example, one item on the reform menu: requiring the use of de-escalation or force-minimization techniques that require officers to utilize the least force necessary in a  confrontation.233 Training officers in the use of these techniques may well reduce rates of police violence by alerting them to alternatives to the use of deadly force. In practice, however, an  aggressive program of disciplining officers when officials determine, with benefit of hindsight, that an officer failed to use the least force necessary in an encounter could invite nearly endless second guessing. Indeed, the Supreme Court has rejected the use of a least restrictive alternative test in assessing searches and seizures under the Fourth Amendment for just this reason.234 A rational officer with a career hanging in the balance, faced with a least-force-necessary directive and a command staff serious about enforcing it, might be well advised to try to avoid the type of encounter that requires a use of force that might later be questioned. De-policing is the likely outcome.

Accordingly, an efficacious reform agenda must be no less attentive to the risk of de-policing than to the possibility that the current regime and its tolerance for over-enforcement, its code of silence, and its difficulty in monitoring, identifying, and sanctioning unwarranted police misconduct, may well, at least in some circumstances, significantly under-deter police violence. It is to the tricky problem of constructing such an agenda that we finally turn.

III. Calibrated Reform: Pretty-Safe Harbors and Pretty-Sure Shipwrecks

The criminal law, as we have seen in Part I.A, is too blunt an instrument to offer a promising vehicle for reform. Criminal sanctions are appropriate in the most extreme cases, but threatening to throw people in jail is unlikely to prove an effective management tool.

We have also seen in Part I.B that civil liability is problematic. It is extraordinarily difficult to identify the right mix of indemnification and personal liability likely to produce enough but not too much deterrence. To be sure, the political costs of damages liability, coupled with the adverse publicity that civil litigation can generate, likely produces some pressure on government to reduce rates of police misconduct. Injunctive relief could also provide an avenue to reform. Nevertheless, whether through the product of damages liability, injunction, or political pressure, responsible managers must identify and then implement an efficacious reform agenda.

Thus, managerial reform is the indispensable means to address police violence. No program of reform, however, offers much promise unless it is politically sustainable. A regime that produces elevated crime rates or an ineffective police department is surely doomed; at the next election, the opponents of reform will have powerful arguments at their disposal. Even the advocates of structural reform litigation as a tool to reduce police violence recognize the problem; they caution that structural injunctions that lack political support among key constituencies may prove difficult to implement,235 and politically unsustainable once the inevitable end of judicial supervision arrives.236 Beyond that, as we have seen, racial minorities, even more than non-minorities, are concerned about both police misconduct and crime.237 Any approach that asks them to accept one in order to reduce the other is likely no more politically tenable among minorities than the public at large.

Given the difficulties in identifying policies that achieve enough, but not too much deterrence, reform must proceed with caution. Indeed, many may find the reforms advanced below to be distressingly limited in scope and number. The potential for unintended consequences and the serious risk that reforms may cost more lives than they save, however, argue for a Hippocratic approach, least likely to do harm.

The discussion above suggests guidelines for successful reform: reform should concentrate on those areas in which over-enforcement is most likely, while attending to the risk of over-deterrence. It is no easy task to adopt a policing strategy that is sufficiently aggressive to reap the benefits of hot-spot policing, but with sufficient accountability mechanisms to adequately deter — but not over-deter — police violence. To achieve these complex goals, the concepts of a safe harbor and sure shipwreck recommend themselves. Safe harbors provide guidance to officers that enable them to avoid discipline, thereby minimizing over-deterrence; while a sure shipwreck is a rule that identifies conduct certain to lead to severe sanction, thus providing deterrence where it is most needed.238 By targeting enforcement and incentivizing candor, both safe harbors and sure shipwrecks can be utilized to reduce police violence.

A. Targeted Enforcement

For decades, scholars have advocated the promulgation of rules that limit the scope of the discretion exercised by police departments and officers on patrol, even though they acknowledge that some degree of discretion remains inescapable in policing.239 Yet, even the advocates of this approach acknowledge its failure; the standard explanation is that elected officials have little incentive to limit the authority of the police given that their constituents are usually more concerned about fighting crime than constraining police authority.240

This insight suggests that regulation is most likely to be sustainable when it is least likely to impair the efficacy of the crime-control function of the police. Confrontations at criminogenic hot spots of crime may be inevitable if policing is to have efficacy; confrontations less likely to serve a crime-fighting function are far more difficult to justify.241

As we have seen, there is impressive evidence of the efficacy of aggressive patrol targeting criminogenic hot spots.242 Outside of these locations, however, aggressive tactics become far more problematic. Stop-and-frisk in New York, for example, likely passed the point of diminishing returns.243 One reason may be a substantial number of investigative stops not focusing on hot spots of crime, but rather on racial minorities whom officers may have regarded as “out of place” because they were in relatively white, low-crime areas.244 Stops of motor vehicles represent another example; traffic stops are unlikely to focus on criminogenic hot spots and there is considerable evidence of racial profiling reflected in traffic stop data.245 Relatedly, some police departments appear to have adopted patrol strategies aimed at generating revenue rather than targeting criminogenic hot spots.246

To focus resources where enforcement is needed, while minimizing over-enforcement, police departments can utilize Compstat-type strategies that target aggressive patrol consistent with crime data at criminogenic hot spots. Managers could promote targeted hot-spot patrol and discouraging aggressive patrol in other areas — creating an effective safe harbor for tactics consistent with a hot-spot policing strategy. Indeed, the success of Compstat has led police strategy throughout the country to become increasingly driven by geographically-targeted crime data identifying criminogenic hot spots.247

Moreover, Compstat-type strategies should be used symmetrically, encouraging properly targeted hot-spot policing but discouraging aggressive tactics elsewhere. Officers engaged in the kind of aggressive tactics that can produce violent confrontations outside of hot spots would therefore court discipline, not only for the use of such tactics in a manner inconsistent with policing strategy, but because, when an officer must utilize force to defend himself as a consequence of an unwarranted use of aggressive tactics outside of a hot spot, the officer could be disciplined not for defending himself, but for engaging in unwarranted tactics that made the use of force necessary. In this fashion, hot-spot policing could confront officers with something approaching a sure shipwreck when they utilize the kind of aggressive tactics associated with successful hot-spot policing elsewhere.

This approach is likely to prove politically sustainable because its use of evidence-based policing strategy makes it is less likely to compromise police efficacy, thereby producing political backlash. Moreover, if enforcement priorities are made apparent through a data-driven strategy open to public view and debate, the democratic legitimacy of policing is likely to be enhanced.248 At a minimum, if the public understands when, where, and why the police undertake aggressive tactics, and can examine the evidence supporting the justifications, it is more likely to be willing to support such tactics. If, on the other hand, the public is unpersuaded, police will appropriately scale back proactive efforts to bring down the crime rate, at least until the public decides stronger medicine is required.

Courts can encourage reforms along these lines. Under current Fourth Amendment doctrine, whether an investigative stop occurs in a “high crime area” is relevant to determining whether an officer had the requisite reasonable suspicion to justify the stop.249 The judicial notion of a “high crime area,” however, has never developed much precision, thereby engendering criticism from those who believe that it puts large numbers of law-abiding residents of high-crime communities at risk of unjustified search and seizure.250 In light of the case for hotspot policing, data identifying criminogenic hot spots could supplant the more indefinite “high-crime area” concept as an important basis for assessing whether stop-and-frisk tactics are justified. Moreover, this refined version of the high-crime area concept should be used symmetrically, producing not only a measure of deference when patrolling targets hot spots, but also greater skepticism about search and seizure occurring elsewhere, at least when it is based on highly ambiguous conduct. Outside of hot spots, accordingly, courts could use Fourth Amendment doctrine to heighten the legal risks facing aggressive policing, thereby encouraging police executives to do the same.

To be sure, a hot spot must be defined with some analytical rigor to ensure that a hot-spot policing strategy does not collapse into aggressively policing poor, but not wealthy, neighborhoods. The increasingly sophisticated methods of geographic crime analysis available to police executives, however, suggest that meaningful rigor in defining hot spots is achievable.251 If courts, in turn, were willing to grant deference to search-and-seizure at appropriately identified hot spots, while exercising greater skepticism elsewhere, it would encourage police executives to develop an evidence-based strategy for hot-spots policing open to both judicial and public scrutiny. In this fashion, Fourth Amendment doctrine could encourage police executives to promote the use of hot-spot policing — based on evidence disclosed and scrutinized in judicial proceedings.

If courts systematized the concept of a high-crime area through evidence-based identification of criminogenic hot spots, Fourth Amendment law would itself produce something approaching a safe harbor, encouraging police executives to formally designate hot spots to enable officers obtain the judicial deference afforded to search-and-seizure in high-crime locations. Judicial skepticism about aggressive search-and-seizure elsewhere, however, would create something approaching a sure shipwreck, induce police executives to discourage those tactics outside of hot spots.

Accordingly, if police executives incorporated the insights of hotspot policing into systems of officer accountability — perhaps with judicial encouragement — the resulting scheme would promote aggressive policing where is most readily justified and discourage it where over-enforcement is more likely. Such a strategy would thereby reduce the rate at which aggressive tactics most likely to produce violent confrontations are employed, by targeting those tactics geographically.252 Policies along these lines would also give patrol officers considerable guidance about where aggressive policing is encouraged through a safe-harbor rule, and where it most likely to lead to internal discipline — something approaching a sure shipwreck. In this fashion, systems of accountability could be keyed to the location at which a confrontation takes place, obviating the need to make systems of accountability turn entirely on difficult determinations about whether the officer’s use of force was justified.

Targeted enforcement is not the only type of safe harbor and sure shipwreck that police departments can create. Departments that adopt clear policies regarding the use of force and undertake effective training and supervision to ensure that officers understand and can follow those policies also create a kind of safe harbor, and correlative sure shipwreck when officers deviate from policy.253 To be sure, unduly restrictive policies create a risk of de-policing. To minimize that possibility, focusing tactics that give rise to the risk of violent confrontation on those areas where they are most readily justified and discouraging them elsewhere likely strikes the best balance available to urban policing.

B. Incentivizing Candor in Policing

No disciplinary system can hope to control misconduct unless it is able to reach reliable determinations about whether a suspect has committed misconduct. As we have seen, however, there are enormous difficulties in achieving sufficient reliability in factfinding to make the disciplinary system an effective route to accountability, including the code of silence adhered to by police themselves.254

We should not be surprised that a code of silence tends to develop among police. Systems of administrative discipline that offer all stick and no carrot provide officers with little incentive to candor; indeed, they are likely to promote the opposite. Consider the incentives facing police officers in the face of an allegation of misconduct. In light of the difficulties in assembling credible evidence of police misconduct, an officer who has violated policy will often have nothing to gain by confession to wrongdoing. Perhaps, in the face of a strong case of criminal liability, an accused officer may conclude it is in his interest to testify against others in return for favorable treatment, but absent that, admitting misconduct generally gains an officer little. Nor is an officer who has witnessed misconduct likely to have much incentive to suffer the ostracism that a decision to testify against a colleague is likely to elicit.

An effective disciplinary system must accordingly develop incentives that encourage candor. It would offer sufficient benefits to officers who admit fault that they could plausibly see candor as in their interest even when they have violated policy or observed others do so — something approaching a safe harbor. It would, conversely, harshly penalize a lack of candor — something approaching a sure shipwreck.

Even in the face of an officer’s assertion of the Fifth Amendment right to be free from compelled self-incrimination, a police department can require that the officer answer questions relating to his duties, although such statements may not be used as evidence in a criminal proceeding.255 The ability to insist that an officer submit to questioning offers a route to incentivizing candor. Something approaching a safe harbor could be created by capping the maximum administrative penalty facing an officer guilty of misconduct but who has told the truth in his statements concerning that misconduct. A cap of this character would create an incentive to candor of a type common in criminal sentencing.256

Consider a maximum penalty of a six-month suspension (coupled with an appropriate period of probation with suitable conditions) for an officer found to have been forthcoming and truthful. For the typical police officer, losing a half-year of salary is an enormous sanction, but even so, the officer may resume his career. In serious cases in which termination would otherwise be a serious  possibility, a sanctions cap of this character creates a substantial incentive to candor — a pretty-safe harbor.

This approach would have to be coupled with a sure shipwreck for officers who make false statements. If a finding that an officer made a knowing false statement concerning the officer’s performance of duties resulted in a mandatory sanction of dismissal, officers would confront a potent incentive for candor.257 Ex ante, not only the suspect, but all other officers on the scene would run great risks if they made false statements that might be disproved by video or other reliable evidence, even if that evidence did not conclusively establish the use of unlawful force.

In this fashion, treating false statements as a sure shipwreck leverages the promise of video evidence, since that evidence can result in career-ending sanctions even when it does not document the use of unlawful or out-of-policy force. While the difficulties of proving an officer has knowingly made a false statement should not be underestimated, they are far less than the difficulties of proving the underlying misconduct. Any variation between an officer’s initial statement and subsequently acquired video or other credible evidence could well support an administrative finding that would end an officer’s career without need to prove that the officer used excessive force.258 All other officers who witnessed the event would face a similar threat. In such a regime, the code of silence would take a body blow.259

Most departments currently have policies that subject officers who make false statements to discipline,260 although these violations are not often treated seriously.261 Police departments, however, have ample reason to take these violations cause for discharge. Not only do false statements undermine managerial control, but the Due Process Clause likely obligates the prosecution to disclose that an officer has engaged in misconduct that bears on his credibility in all subsequent cases in which the officer testifies.262 Evidence that an officer has engaged in misconduct of the type that could fatally compromise his credibility as a witness might, for that reason alone, justify termination.263 Moreover, while one might sympathize with an officer who made a  momentary error in judgment when confronted with an apparent threat in the field, there is far less reason to be lenient with an officer who, upon sober reflection and after the fact, decided to lie to conceal misconduct, whether his own or that of another.

When I discussed similar proposals with stakeholders during my service in Chicago government, I found a wall of opposition among police and civil rights activists alike. Both bridled at constraining discretion to fire officers found guilty of serious misconduct, and both usually viewed false statements as less serious than the use of excessive force. Yet, nothing in the reform menu canvassed above offers much hope of dismantling the code of silence; in fact, as the likelihood and severity of sanctions increase, the code of silence may well strengthen as officers perceive a common interest in undermining the disciplinary system. And, as long as the code of silence persists, neither police management nor those who seek to hold it accountable are likely to achieve satisfactory results. A substantial measure of safe harbor for candor, and a sure shipwreck for falsity, is the surest way to undermine that code. However counter-intuitive, affording leniency for candor, even in the wake of misconduct, while uniformly discharging the untruthful, is likely essential to dismantling the code of silence that prevents effective oversight of policing.

It is also easy to overstate the cost of a pretty-safe harbor from administrative discipline. Although a pretty-safe-harbor regime for truthful officers provides considerable protection against administrative discipline, nothing would prevent officers guilty of criminal misconduct from facing criminal liability if sufficient evidence can be assembled independent of their compelled statements.264 Nor would a regime protect such an officer from civil liability, subject to applicable indemnification obligations.

Consider the systemic consequences of a pretty-safe-harbor for candor and a pretty-sure-shipwreck for lies. Police misconduct cannot survive a disciplinary process capable of getting to the bottom of an allegation of police misconduct. If the cost of obstructing an investigation rises, and the cost of admitting fault declines, the disciplinary system becomes better able to produce reliable outcomes. If officers come to perceive that they cannot cover up misconduct, moreover, they are less likely to break the rules in the first place.

Moreover, if public testimony and administrative findings disclose misconduct, even if guilty officers are suspended rather than fired, the public will be engaged and is likely to demand better training and supervision. As with the adoption of express policies governing aggressive patrol, greater transparency of this character promotes democratic accountability in policing.

IV. Conclusion

The inevitability of trade-offs should be particularly clear when it comes to policing high-crime, disadvantaged neighborhoods, which present particularly complex law-enforcement problems. One can single-mindedly pursue reductions in police violence, but some other objective is likely to be compromised in the process. Reducing the rate of police violence by accepting a sharply higher likelihood of de-policing and its attendant costs is neither an appealing nor a sustainable course.

To be sure, there is a temptation to deny the existence of trade-offs; many doubtless would prefer to believe that effective police executives can use sanctions alone to reduce the rate of police violence without compromising the efficacy of policing. Elementary behavioral economics, however, suggests that this will be extraordinarily difficult to accomplish — when one increases the costs facing officers who undertake to intervene in a streetscape rife with potential danger, the rate at which officers will be willing to intervene will necessarily decline, while the agency costs of endeavoring to maintain effective supervision will rise as well. No one can know for sure the magnitude of de-policing and its attendant increases in crime that will result as officers face an increased risks of sanction, but a policy that courts this risk gambles with inner-city lives.

If the risk of sanctions is to increase, it must be accompanied by greater certainty about when officers will court sanctions, to minimize the risk of over-deterrence. That is the appeal of a policy rooted in safe harbors and sure shipwrecks. We could do a lot worse. We already have.

The material in all ABA publications is copyrighted and may be reprinted by permission only. Request reprint permission here.
  1. See, e.g., EGON BITTNER, THE FUNCTIONS OF THE POLICE IN MODERN SOCIETY 101-06 (1970); DOUGLAS W. PEREZ, COMMON SENSE ABOUT POLICE REVIEW 19-34 (1994); JEROME H. SKOLNICK, JUSTICE WITHOUT TRIAL 4-6 (1966); WILLIAM A. WESTLEY, VIOLENCE AND THE POLICE 118-50 (1970); Kenneth Adams, Measuring the Prevalence of Police Abuse of Force, in POLICE VIOLENCE: UNDERSTANDING AND CONTROLLING POLICE ABUSE OF FORCE 52, 61-64 (William A. Geller & Hans Toch eds., 1996) [hereinafter POLICE VIOLENCE]; Barbara E. Armacost, Organizational Culture and Police Misconduct, 72 GEO. WASH. L. REV. 453, 493-507 (2004); Alexa P. Freeman, Unscheduled Departures: The Circumvention of Just Sentencing for Police Brutality, 47 HASTINGS L.J. 677, 684-710 (1996); John V. Jacobi, Prosecuting Police Misconduct, 2000 WIS. L. REV. 789, 794-802.
  2. The discussion that follows generally employs the terms “black people” or “black” rather than African American in order to reflect the underlying source material, which generally tracks race and not citizenship.
  4. See, e.g., David A. Klinger, On the Problems and Promise of Research on Lethal Police Violence: A Research Note, 16 HOMICIDE STUD. 78 (2012); Richard Rosenfeld, Ferguson and Police Use of Deadly Force, 80 MO. L. REV. 1077, 1087-88, 1092 (2015). Cf. Franklin E. Zimring, How Many Shootings by Police?, 2016 U. CHI. LEG. F. 691, 705-10 (estimating that official statistics reflect no more than half of all law-enforcement homicides).
  5. See Franklin E. Zimring & Brittany Arsiniega, Trends in Killing of and by Police: A Preliminary Analysis, 13 OHIO ST. J. CRIM. L. 247, 250-52 (2015) (finding that police homicides dropped by about a third between 1976 and 2012, while the death rate of police officers from violent assaults during the same period dropped by around three fourths, a divergence that may reflect the rising rate at which officers wore protective vests while on duty).
  6. See, e.g., ANDREA M. BURCH, U.S. DEP’T OF JUST., ARREST-RELATED DEATHS, 2003-2009, at 2 (Nov. 2011) (report on arrest-related deaths between 2003 and 2009 found than 32% of the victims were Black and 20% were Hispanic); COMM’N TO REV. POLICE POL’Y & PRAC., NAT’L RES. COUNCIL, FAIRNESS AND EFFECTIVENESS IN POLICING: THE EVIDENCE 258-59 (Wesley Skogan & Kathleen Frydl eds., 2004) [hereinafter FAIRNESS AND EFFECTIVENESS IN POLICING] (reviewing evidence that racial minorities are disproportionately likely to be victims of police shootings); WILLIAM A. GELLER & MICHAEL SCOTT, DEADLY FORCE: WHAT WE KNOW 147-52 (1992) (same); Hubert G. Locke, The Color of Law and the Issue of Color: Race and the Abuse of Police Power, in POLICE VIOLENCE, supra note 1, at 129, 135-37 (same); Nirej Sekhon, Blue on Black: An Empirical Assessment of Police Shootings, 55 AM. CRIM. L. REV. (forthcoming), abstract_id=2700724 (survey of FBI supplementary homicide reports for 2012 disclosed that black people comprised 31% of the victims of all justifiable police homicides, although blacks were only 13% of the population); Cody T. Ross, A Multi-Level Bayesian Analysis of Racial Bias in Police Shootings at the County-Level in the United States, 2011-14, at 10-14, PLOS (Nov. 5, 2015), (finding at the county level elevated rates at which both armed and unarmed black people are shot by police compared to whites); The Counted: People Killed by Police in the U.S., THE GUARDIAN, [hereinafter The Counted], (last visited Sept. 16, 2016) (displaying a database of police homicides maintained by The Guardian newspaper indicating that in 2015 there were 1,145 police killings of civilians in the United States, with a homicide rate of 7.22 per million population for black people, 3.51 per million for Hispanics, and only 2.94 per million for whites); Zimring & Arsiniega, supra note 5, at 261 (survey of police killings in 2011 and 2012 finding that black people are killed by police at about three times their representation in the general population).
  7. See, e.g., CHRISTINE EITH & MATTHEW R. DUROSE, U.S. DEP’T OF JUST., NCJ 234599, CONTACTS BETWEEN POLICE AND THE PUBLIC, 2008, at 12 (2011) (detailing analysis of 2008 data finding that while police used force in only 1.4% of contacts with civilians, it was used in 3.4% of contacts with black or African Americans, in 1.6% of contacts with Hispanics and 1.2% of contacts with whites); PHILIP ATIBA GOFF ET AL., CTR. FOR POLICING EQUITY, THE SCIENCE OF JUSTICE: RACE, ARRESTS, AND POLICE USE FORCE 15-18 ( July 2016) (outlining study of data collected between 2010 and 2015 in twelve law-enforcement agencies finding dramatically higher rates at which force was used incident to the arrest of black suspects compared to white suspects); Brad W. Smith & Malcolm D. Holmes, Police Use of Excessive Force in Minority Communities: A Test of the Minority Threat, Place, and Community Accountability Hypotheses, 61 SOC. PROBS. 83, 93-98 (2014) (finding that Black and Hispanic representation in the population of large cities is an important predictor of the number of sustained complaints of excessive force by police officers).
  8. See, e.g., HOLMES & SMITH, supra note 3, at 7-8; William Terrill & Michael D. Reisig, Neighborhood Context and Police Use of Force, 40 J. RES. CRIME & DELINQ. 291, 292-93 (2003); Robert E. Worden, The Causes of Police Brutality: Theory and Evidence on Police Use of Force, in POLICE VIOLENCE, supra note 1, at 23, 23-32. For a recent example of a model of police violence that partakes of all of these theories, see Devon W. Carbado, Blue-On-Black Violence: A Provisional Model of Some of the Causes, 104 GEO. L.J. 1479, 1485-1507 (2016).
  9. See, e.g., David A. Harris, The Dangers of Racialized Perceptions and Thinking by Law Enforcement, in DEADLY INJUSTICE: TRAYVON MARTIN, RACE, AND THE CRIMINAL JUSTICE SYSTEM 146, 152-59 (Devon Johnson, Patricia Y. Warren & Amy Farrell eds., 2015); Devon W. Carbado & Patrick Rock, What Exposes African Americans to Police Violence?, 51 HARV. C.R.-C.L. L. REV. 159, 168-73 (2016); Kenneth Lawson, Police Shootings of Black Men and Implicit Racial Bias: Can’t We All Just Get Along, 37 U. HAW. L. REV. 339, 350-61 (2015); L. Song Richardson & Phillip Atiba Goff, Self-Defense and the Suspicion Heuristic, 98 IOWA L. REV. 293, 296-314 (2012). A related explanation focuses on evidence that black males pose a particular threat to masculine self-confidence that leads to police violence. See, e.g., L. Song Richardson & Phillip Atiba Goff, Interrogating Racial Violence, 12 OHIO ST. J. CRIM. L. 115, 128-42 (2015) [hereinafter Interrogating Racial Violence]. Since both theories rest on associations between a suspect’s race and an officer’s perception of threat, they will both be treated as implicit bias in the discussion below.
  11. See, e.g., Jeffrey M. Jones, Confidence in Local Police Drops to 10-Year Low, GALLUP (Nov. 10, 2005),
  12. For a helpful survey of the leading reform proposals, see Bill Ong Hing, From Ferguson to Palestine: Disrupting Race-Based Policing, 59 HOW. L.J. 559, 584-603 (2016).
  13. See, e.g., Freeman, supra note 1, at 711-12 (“[P]olice brutality is a crime, and if we mean to take police brutality seriously, it needs to be treated as seriously as we treat other crimes.”); Craig B. Futterman, Chaclyn Hunt & Jamie Kalven, Youth/Police Encounters on Chicago’s South Side: Acknowledging the Realities, 2016 U. CHI. LEG. F. 125, 181 (“Treat police brutality and corruption as the serious crimes that they are . . . . If [t]he [officer] has engaged in criminal behavior, prosecute him.”).
  14. See Rachel A. Harmon, When Is Police Violence Justified?, 102 NW. U.L. REV. 1119, 1146-50, 1166 (2008). For a helpful discussion of the scope of the justification defense available to police officers as they undertake to enforce the law, see 2 PAUL H. ROBINSON, CRIMINAL LAW DEFENSES § 142 (1984 & West Supp. 2015). In some jurisdictions, police also face liability under statutes that criminalize forms of official misconduct. See Monu Bedi, Toward a Uniform Code of Police Justice, 2016 U. CHI. LEG. F. 13, 27.
  15. See Bedi, supra note 14, at 28-37.
  16. See 18 U.S.C. § 242 (2012).
  17. 471 U.S. 1 (1985).
  18. Id. at 7, 11. The Fourth Amendment provides:

    The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no Warrants shall issue, but upon probable cause, supported by Oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized.

    U.S. CONST. amend. IV.
  19. Garner, 471 U.S. at 3; see also id. at 11-12 (“Where the officer has probable cause to believe that the suspect poses a threat of serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force. Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he has committed a crime involving the threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, when feasible, some warning has been given.”). The Garner standard has also influenced state-law public-authority defenses, with a recent survey finding that 36 states employ the Garner standard when assessing criminal liability for the use of deadly force. See Chad Flanders & Joseph Welling, Police Use of Deadly Force: State Statutes 30 Years after Garner, 35 ST. LOUIS U. PUB. L. REV. 109, 120-24 (2016).
  20. See, e.g., Scott v. Harris, 550 U.S. 372, 381-86 (2007) (forcing fleeing vehicle off road during high-speed pursuit); Graham v. Connor, 490 U.S. 386, 394-97 (1989) (using physical force during arrest).
  21. See, e.g., United States v. DiSantis, 565 F.3d 354, 363-64 (7th Cir. 2009).
  22. Cnty. of Sacramento v. Lewis, 523 U.S. 833, 846-47 (1998) (quoting Collins v. City of Harker Heights, 503 U.S. 115, 128 (1992)). See generally Rochin v. California, 342 U.S. 165, 172 (1952) (“This is conduct that shocks the conscience. Illegally breaking into the privacy of the petitioner, the struggle to open his mouth and remove what was there, the forcible extraction of his stomach’s contents—this course of proceeding by agents of government to obtain evidence is bound to offend even hardened sensibilities. They are methods too close to the rack and the screw to permit of constitutional differentiation.”).
  23. See, e.g., United States v. Giordano, 260 F. Supp. 2d 477, 484 (D. Conn. 2002), aff ’d, 442 F.3d 50 (2d Cir. 2006).
  24. See, e.g., FAIRNESS AND EFFECTIVENESS IN POLICING; supra note 6, at 276-77; GELLER & SCOTT, supra note 6, at 292-93; SAMUEL WALKER, THE NEW WORLD OF POLICE ACCOUNTABILITY 34-35 (2005); Mary M. Cheh, Are Lawsuits an Answer to Police Accountability?, in POLICE VIOLENCE, supra note 1, at 247, 251-52; Roger Goldman, The Importance of State Law in Police Reform, 60 ST. LOUIS U.L.J. 363, 366 (2016); Kate Levine, How We Prosecute Police, 104 GEO. L.J. 745, 763-64 (2016).
  25. See, e.g., Armacost, supra note 1, at 466-67; Cheh, supra note 24, at 252; Peter L. Davis, Rodney King and the Decriminalization of Police Brutality in America: Direct and Judicial Access to the Grand Jury as a Remedy for Victims of Police Brutality When the Prosecutor Declines To Prosecute, 53MD. L. REV. 271, 289-91 (1994); Jacobi, supra note 1, at 802-06; Kate Levine, Who Shouldn’t Prosecute the Police, 101 IOWA L. REV. 1447, 1465-72 (2016); Louis B. Schwartz, Complaints Against the Police: Experience of the Community Rights Division of the Philadelphia District Attorney’s Office, 118 U. PA. L. REV. 1023, 1024-25 (1970).
  26. See, e.g., PRESIDENT’S TASK FORCE REPORT, supra note 10, at 21 (advocating independent prosecutors); Davis, supra note 25, at 296-98 (advocating direct citizen access to grand juries); Jacobi, supra note 1, at 811-12 (advocating federal prosecutions); Levine, supra note 25, at 1487-96 (considering various alternatives); Kami Chavis Simmons, Increasing Police Accountability: Restoring Trust and Legitimacy through the Appointment of Independent Prosecutors, 49 WASH. U.J.L. & POL’Y 137, 148-54 (2015) (advocating independent prosecutors).
  27. See, e.g., Levine, supra note 25, at 1491-92 (“Assigning police cases to federal prosecutors would go a long way toward resolving any conflict inherent in state prosecutions of police—these attorneys are generally not beholden to the local police, and federal prosecutors are insulated from the political pressures that might plague state actors.” (footnotes omitted)).
  28. See, e.g., Jacobi, supra note 1, at 806-16 (proposing amendments to federal criminal law).
  29. E.g., In re Winship, 397 U.S. 358, 363-64 (1970).
  30. See, e.g., ALBERT J. REISS, JR., THE POLICE AND THE PUBLIC 191-92 (1971); Armacost, supra note 1, at 466; Cheh, supra note 24, at 266; Freeman, supra note 1, at 723-24; Asit S. Panwata, The Failure of Local and Federal Prosecutors to Curb Police Brutality, 30 FORDHAM URB. L.J. 639, 644 (2003).
  31. See, e.g., HERMAN GOLDSTEIN, POLICING A FREE SOCIETY 165-66 (1977); REISS, supra note 30, at 212-14; JEROME H. SKOLNICK & JAMES J. FYFE, ABOVE THE LAW: POLICE AND THE USE OF EXCESSIVE FORCE 110-12 (1993); WALKER, supra note 24, at 64-65; WESTLEY, supra note 1, at 111-18; Gabriel J. Chin & Scott C. Wells, The “Blue Wall of Silence” as Evidence of Bias and Motive to Lie: A New Approach to Police Perjury, 59 U. PITT. L. REV. 233, 250-56 (1998); Morgan Cloud, The Dirty Little Secret, 43 EMORY L.J. 1311, 1315-21 (1994); Donald A. Dripps, Police, Plus Perjury, Equals Polygraph, 86 J. CRIM. L. & CRIMINOLOGY 639, 698-703 (1996); Futterman, Hunt & Kalven, supra note 13, at 183-85; Aziz Z. Huq & Richard H. McAdams, Litigating the Blue Wall of Silence: How to Challenge the Police Privilege to Delay Investigation, 2016 U. CHI. LEG. F. 213, 214-19; Jennifer L. Koepke, Note, The Failure to Breach the Blue Wall of Silence: The Circling of the Wagons to Protect Police Perjury, 39 WASHBURNE L.J. 211, 221-22 (2001).
  32. PAUL CHEVIGNY, EDGE OF THE KNIFE: POLICE VIOLENCE IN THE AMERICAS 92 (1995). Accord, e.g., HOLMES & SMITH, supra note 3, at 137 (“The low visibility of police work and the blue wall of silence mean that details of many incidents of police brutality are known only to the police officers and citizens involved. As a result, excessive force complaints are determined to be unfounded for lack of credible evidence.”).
  33. See, e.g., SKOLNICK & FYFE, supra note 31, at 195-98, 199; Cheh, supra note 24, at 253.
  34. WALKER, supra note 24, at 35. For discussions of juries’ reluctance to convict in these cases, see PEREZ, supra note 1, at 51; SKOLNICK & FYFE, supra note 31, at 19; Freeman, supra note 1, at 726; and Goldman supra note 24, at 366-67. With respect to judicial deference, the Supreme Court cautioned, in the course of holding in a civil case that officers did not violate the Fourth Amendment when shooting at a fleeing vehicle following a traffic stop, killing both occupants:

    We analyze this question from the perspective “of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight.” We thus “allo[w] for the fact that police officers are often forced to make split-second judgments—in circumstances that are tense, uncertain, and rapidly evolving—about the amount of force that is necessary in a particular situation.”

    Plumhoff v. Rickard, 134 S. Ct. 2012, 2020 (2014) (brackets in original and citations omitted) (quoting Graham v. Connor, 490 U.S. 386, 396-97 (1989)).
  35. See supra text accompanying note 9.
  36. See Jennifer S. Hunt, Race, Ethnicity, and Culture in Jury Decision Making, 11 ANN. REV. L. & SOC. SCI. 269, 271-77 (2015).
  37. See, e.g., POLICE EXECUTIVE RESEARCH FORUM, U.S. DEP’T OF JUST., IMPLEMENTING A BODY-WORN CAMERA PROGRAM: RECOMMENDATIONS AND LESSONS LEARNED 5-9 (2014); Futterman, Hunt & Kalven, supra note 13, at 187-90; David A. Harris, Picture This: Body-Worn Video Devices (Head Cams) as Tools for Ensuring Fourth Amendment Compliance by Police, 43 TEX. TECH L. REV. 357, 363-66 (2010); Hing, supra note 12, at 579-81; Kami Chavis Simmons, Body-Mounted Police Cameras: A Primer on Police Accountability vs. Privacy, 58 HOW. LJ. 881, 884-87 (2015); Karson Kampfe, Note,Police-Worn Body Cameras: Balancing Privacy and Accountability through State and Police Department Action, 76 OHIO ST. L.J. 1153, 1161-68 (2015); Iesha S. Nunes, Note, “Hands Up Don’t Shoot”: Police Misconduct and the Need for Body Cameras, 67 FLA. L. REV. 1811, 1823-30 (2015); Waleska Suero, Note, Lessons from Floyd v. City of New York: Designing Race-Based Remedies for Equal Protection Violations in Stop-and-Frisk Cases, 7 GEO. J.L & MOD. CRITICAL RACE PERSP. 139, 143 (2015); Note, Developments in the Law—Policing, 128 HARV. L. REV. 1707, 1717-1818 (2015) [hereinafter Developments].
  38. See, e.g., Howard M. Wasserman, Orwell’s Vision: Video and the Future of Civil Rights Enforcement, 68MD. L. REV. 600, 618-30 (2009). There are also privacy objections to the use of police-worn body cameras. For a helpful discussion, see Kelly Freund, Note, When Cameras Are Rolling: Privacy Implications of Body-Mounted Cameras on Police, 49 COLUM. J.L. & SOC. PROBS. 91 (2015).
  39. See, e.g., Dan M. Kahan, David A. Hoffman & Donald Braman, Whose Eyes Are You Going to Believe? Scott v. Harris and the Perils of Cognitive Illiberalism, 122 HARV. L. REV. 837, 865-81 (2009); Roseanna Summers, Note, Will Putting Body Cameras on Police Reduce Polarization?, 125 YALE L.J. 1304, 1345-52 (2016).
  40. SKOLNICK & FYFE, supra note 31, at 197-98.
  42. See supra text accompanying notes 17-23.
  43. See, e.g., John C. Jeffries, Jr., Damages for Constitutional Violations: The Relation of Risk to Injury in Constitutional Torts, 75 VA. L. REV. 1461, 1461-74 (1989); Larry Kramer & Alan O. Sykes, Municipal Liability under § 1983: A Legal and Economic Analysis, 1987 SUP. CT. REV. 249, 276-87; Daniel J. Meltzer, Deterring Constitutional Violations by Law Enforcement Officials: Plaintiffs and Defendants as Private Attorneys General, 88 COLUM. L. REV. 247, 253-78 (1988).
  44. See, e.g., FAIRNESS AND EFFECTIVES IN POLICING, supra note 6, at 278-80; WALKER, supra note 24, at 32-33; Barry Friedman & Mario Ponomarenko, Democratic Policing, 90 N.Y.U. L. REV. 1827, 1867-70 (2015); Joanna C. Schwartz, Police Indemnification, 89 N.Y.U. L. REV. 885, 954-59 (2014).
  45. See supra text accompanying notes 30-36.
  46. See FDIC v. Meyer, 510 U.S. 471, 477 (1994).
  47. 28 U.S.C. § 2680(a)-(h) (2012).
  48. E.g., Bivens v. Six Unknown Named Agents, 403 U.S. 388, 396 (1971).
  49. See, e.g., Saucier v. Katz, 533 U.S. 194, 200-01 (2001).
  50. See, e.g., Mullenix v. Luna, 136 S. Ct. 305, 308-12 (2015) (per curiam); Plumhoff v. Rickard, 134 S. Ct. 2012, 2022-24 (2014); Brosseau v. Haugen, 543 U.S. 194, 197-201 (2004) (per curiam). Some have argued that because the constitutional standard for excessive force is defined at a high level of generality that places few specific limitations on the use of force, that standard, coupled with the availability of qualified immunity when an officer does not violate clearly established law, unduly circumscribes damages liability. See, e.g., Harmon, supra note 14, at 1140-46.
  51. See Will v. Mich. Dep’t of St. Police, 491 U.S. 58, 65-71 (1989).
  52. See, e.g., Hafer v. Melo, 502 U.S. 21, 27-29 (1991).
  53. See, e.g., Scheuer v. Rhodes, 416 U.S. 232, 239-49 (1974).
  54. See Bd. of Cnty. Comm’rs v. Brown, 520 U.S. 397, 403-04 (1997); City of Canton v. Harris, 489 U.S. 315, 323-26 (1991).
  55. See City of Newport v. Fact News Concerts, Inc., 453 U.S. 247, 258-70 (1981).
  56. See, e.g., Pearson v. Callahan, 555 U.S. 223, 231 (2009).
  57. For a discussion of the scope and extent of these immunities, see Lawrence Rosenthal, A Theory of Governmental Damages Liability: Torts, Constitutional Torts, and Takings, 9 U. PA. J. CONST. L. 797, 804-13 (2007).
  58. See, e.g., Richard Emery & Ilann Margalit Maazel, Why Civil Rights Lawsuits Do Not Deter Police Misconduct: The Conundrum of Indemnification and a Proposed Solution, 28 FORDHAM URB. L.J. 587, 590-96 (2000); Rosenthal, supra note 57, at 812-13 n.51, 819-20; Schwartz, supra note 44, at 912-37; Martin A. Schwartz, Should Juries Be Informed that Municipality Will Indemnify Officer’s § 1983 Liability for Constitutional Wrongdoing?, 86 IOWA L. REV. 1209, 1216-23 (2000).
  59. See, e.g., Alan O. Sykes, The Economics of Vicarious Liability, 93 YALE L.J. 1231, 1239-43 (1984).
  60. For proposals along these lines, see, for example, Emery & Maazel, supra note 58, at 597-600; and Schwartz, supra note 44, at 953-54.
  61. See, e.g., Schwartz, supra note 44, at 952.
  62. See, e.g., Donald A. Dripps, The “New” Exclusionary Rule Debate: From “Still Preoccupied with 1985” to “Virtual Deterrence,” 37 FORDHAM URB. L.J. 743, 763-64 (2010) (“Individual officers do not internalize either the benefits or the costs of Fourth Amendment activity. When the police apprehend an offender, they may improve their performance evaluations and gain prestige within the force. They do not, however, pocket what the community is willing to pay to prosecute and punish the offender.”); Richard A. Posner, Excessive Sanctions for Governmental Misconduct in Criminal Cases, 57 WASH. L. REV. 635, 640 (1982) (“Police and other law-enforcement personnel are compensated on a salaried rather than piece-rate basis, so that even if they perform their duties with extraordinary zeal and effectiveness they do not receive financial rewards commensurate with their performance.”).
  63. Cf. Myron W. Orfield, Jr., The Exclusionary Rule and Deterrence: An Empirical Study of Chicago Narcotics Officers, 54 U. CHI. L. REV. 1016, 1053 (1987) (detailing survey of Chicago narcotics officers found that 95% believed that a regime of damages liability replacing the exclusionary rule for search-and-seizure in violation of the Fourth Amendment “police would be afraid to conduct the searches they should make”).
  64. For proposals along these lines, see, for example, Paul Hoffman, The Feds, Lies, and Videotapes: The Need for an Effective Federal Role in Controlling Police Abuse in Urban America, 66 S. CAL. L. REV. 1453, 1518-21 (1993) (proposing direct liability of public agencies for compensatory damages caused by constitutional torts of officers, abolishing qualified immunity, prohibiting indemnification for punitive damages, and abolishing witnesses’ immunity from damages liability); Huq & McAdams, supra note 31, at 241-45 (advocating damages liability for use of procedures that inhibit effective investigations of police misconduct); Jon O. Newman, Suing the Lawbreakers: Proposals to Strengthen the Section 1983 Damage Remedy for Law Enforcers’ Misconduct, 87 YALE L.J. 447, 453-65 (1978) (advocating expansion of standing and recoverable damages, permitting suit against governmental defendants, changing the burden of proof, and restricting defenses).
  65. See, e.g., CHEVIGNY, supra note 32, at 101-05; Schwartz, supra note 44, at 955-56; Samuel Walker & Morgan MacDonald, An Alternative Remedy for Police Misconduct: A Model “Pattern or Practice” Statute, 19 GEO. MASON U. C.R.L.J. 479, 494-95 (2009).
  66. See Daryl J. Levinson, Making Government Pay: Markets, Politics, and the Allocation of Constitutional Costs, 67 U. CHI. L. REV. 345, 348-57 (2000).
  67. See id. at 370.
  68. See, e.g., FAIRNESS AND EFFECTIVENESS IN POLICING, supra note 6, at 279-80; SKOLNICK & FYFE, supra note 31, at 205-07.
  69. For a more elaborate argument along these lines, see Rosenthal, supra note 57, at 831-41. Relatedly, Joanna Schwartz, in a study of large police departments, concluded that many of them appear to respond to litigation costs by making some efforts to reduce misconduct. See Joanna C. Schwartz, What Police Learn from Lawsuits, 33 CARDOZO L. REV. 841, 859-61 (2012). Similarly, there is considerable evidence that insurers have effectively compelled police departments to undertake reforms that reduce liabilities. See John Rappaport, An Insurance-Based Typology of Police Misconduct, 2016U. CHI. LEG. F. 369, 376-82. Relatedly, Christopher Slobogin has observed that “[t]he single area in which most police departments have both rigorous training and systematic administrative rules is in the use of force, which happens to be one of the few domains where police are successfully sued for large sums of money.” Christopher Slobogin, Why Liberals Should Chuck the Exclusionary Rule, 1999 U. ILL. L. REV. 363, 396 (1999) (footnote omitted). In his detailed examination of law and policy in this area, however, Charles Epp makes a case that judicial decisions, by articulating legal constraints on the use of force, shaped police behavior not by the economic effects of liability, but by articulating legal norms that law enforcement agencies, given relation to legal norms, come to internalize. See CHARLES A. EPP, MAKING RIGHTS REAL: ACTIVISTS, BUREAUCRATS, AND THE CREATION OF THE LEGALISTIC STATE 59-114 (2009).
  70. See, e.g., Myriam E. Gilles, In Defense of Making Government Pay: The Deterrent Effect of Constitutional Tort Remedies, 35 GA. L. REV. 845, 859-66 (2001). The perverse outcome that this possibility presents, however, is that government officials will agree to settle litigation to prevent information potentially embarrassing to incumbent officeholders from coming to light. See Rosenthal, supra note 57, at 829. Cf. Schwartz, supra note 69, at 886 (“The threat of litigation, discipline, and other negative repercussions may have caused government personnel—including, at times, personnel in the five departments in this Study—to hide or misrepresent the kinds of information crucial to performance-improvement efforts.” (footnote omitted)); see also Marc L. Miller & Ronald F. Wright, Secret Police and the Mysterious Case of the Missing Tort Claims, 52 BUFF. L. REV. 757, 775-77 (2004) (noting the prevalence of settlements of police misconduct litigation with confidentiality provisions that keep the settlements secret).
  71. See, e.g., FAIRNESS AND EFFECTIVENESS IN POLICING, supra note 6, at 279; Miller & Wright, supra note 70, at 781-82; Schwartz, supra note 44, at 956-57; Walker & MacDonald, supra note 65, at 494-95.
  72. See Joanna C. Schwartz, How Governments Pay: Lawsuits, Budgets, and Police Reform, 63 UCLA L. REV. 1144, 1165-93 (2016) (finding that approximately half of jurisdictions surveyed required some contribution to the payment of judgments from police budgets, but budgetary arrangements rarely produced meaningful budgetary pressure on police departments).
  73. George C. Thomas III, Judges Are Not Economists and Other Reasons to Be Skeptical of Contingent Suppression Orders: A Response to Professor Dripps, 38 AM. CRIM. L. REV. 47, 55 (2001).
  74. For a more elaborate argument along these lines, see Rosenthal, supra note 57, at 845-47.
  75. E.g., Pearson v. Callahan, 555 U.S. 223, 242-43 (2009).
  76. See City of Los Angeles v. Lyons, 461 U.S. 95, 101-10 (1983). Cf. Clapper v. Amnesty Int’l USA, 133 S. Ct. 1138, 1142-45 (2013) (holding plaintiffs lacked standing to seek injunction against allegedly unlawful surveillance because they could not establish that they would be subject to such surveillance in the future).
  77. See, e.g., Armacost, supra note 1, at 491-93.
  78. 42 U.S.C. § 14141 (2012).
  79. See, e.g., Armacost, supra note 1, at 525-31; Rachel A. Harmon, Promoting Civil Rights through Proactive Police Reform, 62 STAN. L. REV. 1, 23-52 (2009); Stephen Rushin, Structural Reform Litigation in American Police Departments, 99MINN. L. REV. 1343, 1396-406 (2015).
  80. See, e.g., Walker & MacDonald, supra note 65, at 536-49.
  81. See, e.g., Myriam E. Gilles, Reinventing Structural Reform Litigation: Deputizing Private Citizens in the Enforcement of Civil Rights, 100 COLUM. L. REV. 1384, 1414-24 (2000).
  82. See, e.g., Harmon, supra note 79, at 20-21; Stephen Rushin, Using Data to Reduce Police Violence, 57 B.C. L. REV. 117, 151-69 (2016); Rushin, supra note 79, at 1408-18; Kami Chavis Simmons, Cooperative Federalism and Police Reform: Using Congressional Spending Power to Promote Police Accountability, 62 ALA. L. REV. 351, 373-76 (2011); Walker & MacDonald, supra note 65, at 518-36.
  83. See, e.g., WALKER, supra note 24, at 22-28; Rachel A. Harmon, Limited Leverage: Federal Remedies and Policing Reform, 32 ST. LOUIS U. PUB. L. REV. 33, 38-39 (2012); Rachel Moran, Ending the Internal Affairs Farce, 64 BUFF. L. REV. 837, 853-68 (2016).
  84. See, e.g., Catherine L. Fisk & L. Song Richardson, Police Unions, 85 GEO. WASH. L. REV. __, __ (forthcoming), (manuscript at 30-40); Rachel Harmon, The Problem of Policing, 110 MICH. L. REV. 765, 795-801 (2015); Huq & McAdams, supra note 31, at 221-26; Kevin M. Keenan & Samuel Walker, An Impediment to Police Accountability? An Analysis of Law Enforcement Officers’ Bills of Rights, 14 B.U. PUB. INT. L.J. 185, 203-41 (2005); Kate Levine, Police Suspects, 116 COLUM. L. REV. 1196, 1220-27 (2016); Schwartz, supra note 25, at 1027-30; Seth W. Stoughton, The Incidental Regulation of Policing, 98 MINN. L. REV. 2179, 2205-17 (2014).
  85. See, e.g., REISS, supra note 30, at 189-90; SKOLNICK & FYFE, supra note 31, at 197.
  86. See, e.g., PRESIDENT’S TASK FORCE REPORT, supra note 10, at 26 (recommending “some form of civilian oversight”); PEREZ, supra note 1, at 250-74 (recommending comprehensive civilian monitoring); SKOLNICK & FYFE, supra note 31, at 220-29 (recommending a civilian body with a role in investigation or adjudication of police disciplinary matters); WALKER, supra note 24, at 135-70 (recommending a civilian auditing function); Futterman, Hunt & Kalven, supra note 13, at 176-79 (recommending some form of civilian oversight); Wayne A. Kerstetter, Who Disciplines the Police? Who Should?, in POLICE LEADERSHIP IN AMERICA: CRISIS AND OPPORTUNITY 149, 171-81 (William A. Geller ed., 1985) (recommending civilian monitoring); Moran, supra note 83, at 882-904 (recommending an independent review agency to investigate and discipline officers); Werner E. Peterson, Police Accountability and Civilian Oversight of Policing: an American Perspective, in COMPLAINTS AGAINST THE POLICE: THE TREND TO EXTERNAL REVIEW 259, 269-83 (Andrew J. Goldsmith ed., 1991) (recommending civilian involvement in investigation and adjudication of complaints).
  87. See, e.g., CHEVIGNY, supra note 32, at 88-98; WALKER, supra note 24, at 117-78; Armacost, supra note 1, at 538-40.
  88. See, e.g., PRESIDENT’S TASK FORCE REPORT, supra note 10, at 9-12; Rod K. Brunson & Jacinta M. Gau, Race, Place, and Policing the Inner-City, in THE OXFORD HANDBOOK OF POLICE AND POLICING 362, 367-76 (Michael D. Reisig & Robert J. Kane eds., 2014) [hereinafter OXFORD HANDBOOK OF POLICE]; Futterman, Hunt & Kalven, supra note 13, at 206-08; Wayne A. Kerstetter, Toward Justice for All: Procedural Justice and the Review of Citizen Complaints, in POLICE VIOLENCE, supra note 1, at 234, 240-43; Stephen J. Schulhofer, Tom R. Tyler & Aziz Z. Huq, American Policing at a Crossroads: Unsustainable Policies and the Procedural Justice Alternative, 101 J. CRIM. L. & CRIMINOLOGY 335, 345-61 (2011). Cf. Seth W. Stoughton, Principled Policing: Warrior Cops and Guardian Officers, 51 WAKE FOREST L. REV. 611, 666-74 (2016) (advocating reforms to police culture that inculcate greater respect for community members).
  89. See, e.g., Richardson & Goff, Interrogating Racial Violence, supra note 9, at 143-47. For a helpful discussion of the relationship between procedural justice and community policing, see LAURA KUNARD & CHARLENE MOE, U.S. DEP’T OF JUST., PROCEDURAL JUSTICE FOR LAW ENFORCEMENT: AN OVERVIEW 11-13 (2015). For discussion of a variant in which community groups organize to provide regular scrutiny of police activity, see Jocelyn Simonson, Copwatching, 91 CALIF. L. REV. 391, 407-27, 443-44 (2015).
  90. Cf. MARK L. DANTZKER, UNDERSTANDING TODAY’S POLICE 218 (4th ed. 2010) (“It has often been observed that [community-oriented policing] succeeds in communities that need it least: affluent, white, low-crime areas.”); HOLMES & SMITH, supra note 3, at 138 (“[T]he impoverished neighborhoods most in need of community policing programs are exactly where these programs may be least effective. The ‘community’ envisioned in community policing does not exist in America’s most crime-ridden and socially disorganized neighborhoods . . . .” (citations omitted)); WESLEY G. SKOGAN ET AL., ON THE BEAT: POLICE AND COMMUNITY PROBLEM SOLVING 223-24 (1999) (“[W]e found that the native, ‘homegrown’ capacity of study neighborhoods to defend themselves largely mirrored the pattern of privilege and privation in American society in general. Poor and internally divided beats found it more difficult to translate their common values into practice, while better-off beats had an easier time.”).
  91. See, e.g., William Terrill & Stephen D. Mastrofski, Situational and Officer-Based Determinants of Police Coercion, 19 JUST. Q. 215, 236 (2002) (“Suspects who displayed disrespectful behavior toward officers were no more likely to have force used on them than were those who were respectful.” (emphasis deleted)).
  92. Cf. David H. Bayley & James Garofalo, The Management of Violence by Police Patrol Officers, 27 CRIMINOLOGY 1, 20 (1989) (outlining an observational study, concluding “the employment of clever defusing tactics once contact has been made is probably not the key to minimizing force”); William Terrill & Eugene A. Paoline III, Citizen Complaints as a Threat to Police Legitimacy: The Role of Officers’ Occupational Attitudes, 31 J. CONTEMP. CRIM. JUST. 192, 204-05 (2015) (finding that officers with more positive views of community policing generate more discourtesy complaints, perhaps because officers undertaking community-policing initiatives are perceived as overstepping their roles).
  93. See EPP, MAYNARD-MOODY & HAIDER-MARKEL, supra note 3, at 114-33. Indeed, one of the leading advocates of incorporating procedural-justice norms into policing strategy has acknowledged that many factors come to influence community perceptions of police beyond the extent to which community members are treated fairly in individual encounters with officers, including the number of investigative stops, their intrusiveness, and perceptions of police competence. See Tracey L. Meares, The Law and Social Science of Stop and Frisk, 2014 ANN. REV. L. & SOC. SCI. 335, 348.
  94. Cf. Wesley G. Skogan, Maaten Van Craen & Cari Hennessey, Training for Procedural Justice, 11 J. EXPERIMENTAL CRIMINOLOGY 319, 334 (2015) (outlining a study of training to encourage Chicago police officers to adhere to norms of procedural justice in policing and concluding: “[I]n the long run, training does not appear to have had an impact on the willingness of Chicago officers to trust the public.”).
  95. See, e.g., John R. Lott, Jr. & Carlisle E. Moody, Do White Police Officers Unfairly Target Black Suspects? 16 (Sep. 20, 2016) (unpublished manuscript), (in a national dataset of police homicides, community policing had no effect on police homicide rate); Smith & Holmes, supra note 7, at 97 (finding community policing and civilian review did not reduce sustained complaints of excessive force against police).
  96. See, e.g., PRESIDENT’S TASK FORCE, supra note 10, at 20-24; Futterman, Hunt, & Kalven, supra note 13, at 166-70; Hing, supra note 12, at 583-94.
  97. See, e.g., Futterman, Hunt, & Kalven, supra note 13, at 170-72; Rushin, supra note 82, at 132-36.
  98. Relatedly, some have advocated enhancing racial diversity on police forces to enhance the confidence of the entire community in the police. See, e.g., LAURIE FRIDELL ET AL., POL. EXEC. RES. FORUM, RACIALLY BIASED POLICING: A PRINCIPLED RESPONSE 68-70 (2001); SKLANSKY, supra note 3, at 143-51. Whatever the virtues of enhanced racial diversity, however, there is little reason to believe that it would reduce the incidence of unlawful police violence against persons of color. There is little evidence that police officers of color are less likely to use force against suspects of color. See, e.g., JODI M. BROWN & PATRICK A. LANGAN, U.S. DEP’T OF JUST., POLICING AND HOMICIDE 1976-98: JUSTIFIABLE HOMICIDE BY POLICE, POLICE OFFICERS MURDERED BY FELONS 12 (Mar. 2001); GELLER & SCOTT, supra note 6, at 157-61; HOLMES & SMITH, supra note 3, at 135-36; WILLIAM TERRILL, POLICE COERCION: APPLICATION OF FORCE CONTINUUM 19, 127, 134 (2001); Kenneth Adams, What We Know About Police Use of Force, in USE OF FORCE BY POLICE: OVERVIEW OF NATIONAL AND LOCAL DATA 1, 6-7 (1999) [hereinafter USE OF FORCE BY POLICE]; Lott & Moody, supra note 95, at 14-15, 16-18; Sekhon, supra note 6, at 15-18; Brad W. Smith, The Impact of Police Officer Diversity on Police-Caused Homicides, 31 POL’Y STUD. J. 147, 158-59 (2003); Smith & Holmes, supra note 7, at 97. But cf. James P. McElvain & Augustine J. Kpsowa, Police Officer Characteristics and the Likelihood of Using Deadly Force, 35 CRIM. JUST. & BEHAV. 505, 514-16 (2008) (finding in Riverside County, California, white officers are more likely to be involved in shootings than Black or Hispanic officers); Joscha Legewie & Jeffrey Fagan, Group Threat, Police Officer Diversity, and the Deadly Use of Police Force 27-31 (Columbia Pub. Law, Research Paper No. 14-512, May 15, 2016), papers.cfm?abstract_id=2778692 (finding greater racial diversity in urban police forces reduces rates of police shooting of black people). There is a similar paucity of evidence suggesting that female police officers are less likely to be involved in police violence. See, e.g., HOLMES & SMITH, supra note 3, at 136; Lott & Moody, supra note 95, at 19-20.
  99. See, e.g., DANTZKER, supra note 90, at 287; FRIDELL ET AL., supra note 98, at 79-98; GOLDSTEIN, supra note 31, at 172-73; PRESIDENT’S TASK FORCE REPORT, supra note 10, at 20-21; James J. Fyfe, Training to Reduce Police-Civilian Violence, in POLICE VIOLENCE, supra note 1, at 165, 169-78; Cynthia Lee, Race, Policing, and Lethal Force: Remedying Shooter Bias with Martial Arts Training, 79 L. & CONTEMP. PROBS., vol. 79, no. 3, at 145, 160-70 (2016); Michael D. White, Preventing Racially Biased Policing through Internal and External Controls, in RACE, ETHNICITY, AND POLICING: NEW AND ESSENTIAL READINGS, 468, 470-77 (Stephen K. Rice & Michael D. White eds., 2010) [hereinafter RACE, ETHNICITY, AND POLICING]; see also SKOLNICK & FYFE, supra note 31, at 246-50 (advocating peer- driven programs to develop policies to reduce the likelihood of violent confrontations).
  100. See, e.g., GOLDSTEIN, supra note 31, at 171-71; JEFFREY T. NOBLE & GEOFFREY P. ALPERT, MANAGING ACCOUNTABILITY SYSTEMS FOR POLICE MISCONDUCT: INTERNAL AFFAIRS AND EXTERNAL OVERSIGHT 2-39, 286-76 (2009); SKOLNICK & FYFE, supra note 31, at 231-35; WALKER, supra note 24, at 71-134; Futterman, Hunt & Kalven, supra note 13, at 179-81; Walker & MacDonald, supra note 65, at 508-10, White, supra note 99, at 475-76; see also Rushin, supra note 79, at 1379-85 (noting that most federal consent decrees governing policing impose requirements relating to training, monitoring, supervision, investigation and discipline with respect to police use of force).
  101. See, e.g., Goldman, supra note 24, at 381-86.
  102. See, e.g., Futterman, Hunt & Kalven, supra note 13, at 200; Robert J. Smith, Reducing Racially Disparate Police Outcomes: Is Implicit Bias Training the Answer?, 37 U. HAW. L. REV. 295, 300-02 (2015).
  103. See, e.g., PRESIDENT’S TASK FORCE REPORT, supra note 10, at 20-21; WALKER, supra note 24, at 51-56; Futterman, Hunt & Kalven, supra note 13, at 197-98; Hing, supra note 12, at 594-98; Carl B. Klockars, A Theory of Excessive Force and Its Control, in POLICE VIOLENCE, supra note 1, at 1, 8-11; see also Brandon L. Garrett & Seth W. Stoughton, A Tactical Fourth Amendment, 103 VA. L. REV. __, __ (forthcoming), abstract_id=2754759 (manuscript at 43-50) (advocating incorporation of de-escalation principles into Fourth Amendment and qualified immunity doctrine).
  104. See, e.g., Futterman, Hunt, & Kalven, supra note 13, at 192-96; Sekhon, supra note 6, at 45.
  105. See Jonathan Mummolo, Can New Procedures Improve the Quality of Policing? The Case of “Stop, Frisk, and Question” in New York City 19-20 (Feb. 18, 2016) (unpublished manuscript), cfm?abstract_id=2739222.
  106. See, e.g., HOLMES & SMITH, supra note 3, at 130-31 (reviewing evidence that de-biasing training may be ineffective or counterproductive); Jessica J. Sim, Joshua Correll & Melody S. Sadler, Understanding Police and Expert Performance: When Training Accentuates (vs. Exacerbates) Stereotypic Bias in the Decision to Shoot, 39 PERSONALITY & SOC. PSYCHOL. BULL. 291, 300-02 (2013) (finding that de-biasing training has limited efficacy for police officers and may exacerbate bias by suggesting race is diagnostic of danger); Smith, supra note 102, at 305 (“[W]e still do not know enough to conclude definitely that implicit bias training of police officers is an effective strategy.”).
  107. See, e.g., Joshua Correll et al., The Police Officer’s Dilemma: A Decade of Research on Racial Bias in the Decision to Shoot, 8 SOC. & PERSONALITY PSYCH. COMPASS 201 (2014) (finding evidence of implicit bias against black suspects but no effect on police decisions to shoot); Lois James, Stephen M. James & Bryan J. Vila, The Reverse Racism Effect: Are Cops More Hesitant to Shoot Black Than White Suspects?, 15 CRIMINOLOGY & PUB. POL’Y 457, 458-60 (2016) (finding evidence of implicit bias against black suspects but that officers take longer to shoot at armed black suspects and are less likely to shoot at unarmed black suspects); Lee, supra note 99, at 159 (“The contradictory findings of the various shooter bias studies suggest that implicit racial bias may be playing less of a role in police shootings than initially thought.”).
  108. See, e.g., N.Y.C. POLICE DEP’T, ANNUAL FIREARMS DISCHARGE REPORT, 2014, at 25 fig.17 (Oct. 2015) [hereinafter 2014 NEW YORK FIREARMS DISCHARGE REPORT] (reporting on 2014 officer-involved shootings indicating 97% of suspects were armed, 66% with firearms); GEORGE FACHNER & STEVEN CARTER, COLLABORATIVE REFORM INITIATIVE: AN ASSESSMENT OF DEADLY FORCE IN THE PHILADELPHIA POLICE DEPARTMENT 27 tbl.5 (2015) (outlining a study of officer-involved shootings in Philadelphia from 2007 through 2013 finding 84.6% of suspects were armed, 55.8% with a firearm); David Klinger et al., Race, Crime, and the Micro-Ecology of Deadly Force, 15 CRIMINOLOGY & PUB. POL’Y 193, 204-05 (2015) (containing study of officer-involved shootings in St. Louis between 2003 and 2012 finding at least one suspect was armed in 78.7% of cases); Lott & Moody, supra note 95, at 11 (In a national dataset of police homicides “[m]ost of the suspects, 60 percent, were armed with a firearm, 18% with a knife or cutting instrument, and 4% of the suspects used a vehicle as a weapon.”); Sekhon, supra note 6, at 21 (studying officer-involved shootings in Chicago between 2006 and 2014, finding a firearm was recovered from the suspect in the majority of cases).
  109. Sekhon, supra note 6, at 27.
  110. See, e.g., JOHN BREHM & SCOTT GATES, WORKING, SHIRKING, AND SABOTAGE: BUREAUCRATIC RESPONSE TO A DEMOCRATIC PUBLIC 131-71 (1993); MICHAEL K. BROWN, WORKING THE STREET: POLICE DISCRETION AND THE DILEMMAS OF REFORM 96-131 (1980); GOLDSTEIN, supra note 31, at 101-27; HOLMES & SMITH, supra note 3, at 131-32; REISS, supra note 30, at 200-02; SKOLNICK & FYFE, supra note 31, at 123-24; SAMUEL WALKER, POLICE ACCOUNTABILITY: THE ROLE OF CITIZEN OVERSIGHT 8 (2001); JAMES Q. WILSON, VARIETIES OF POLICE BEHAVIOR: THE MANAGEMENT OF LAW AND ORDER IN EIGHT COMMUNITIES 64-69 (1969); David N. Allen, Police Supervision on the Street: An Analysis of Supervisory Officer Interaction During the Shift, 10 J. CRIM. JUST. 91, 104-06 (1982); Jeffrey Manditch Prottas, The Power of the Street-Level Bureaucrat in Public Service Bureaucracies, 13 URB. AFF. REV. 285, 306-10 (1978).
  111. Cf., e.g., MICHAEL LIPSKY, TOWARD A THEORY OF STREET-LEVEL BUREAUCRACY 16 (1969) (“The consequences of failure to guard against physical threat are so severe that the tendency will develop to employ safety mechanisms as often, rather than as little as possible.”).
  112. See, e.g., GELLER & SCOTT, supra note 6, at 275 (“A restrictive shooting policy is no better than its enforcement.”).
  113. David Rudovsky, Police Abuse: Can the Violence Be Contained, 27 HARV. C.R.-C.L. L. REV. 465, 493 (1992).
  114. See supra text accompanying notes 30-32. One could simply rely on officers’ reports, which do not always document adequate justifications for the use of force; for a discussion of this approach, see Loren T. Atherly & Mathew J. Hickman, Controlling Use of Force: Identifying Police Use of Force through Analysis of Administrative Records, 8 POLICING 123 (2014). If such reports routinely became the basis for imposing discipline, however, one could be confident that officers would take care to complete reports in a fashion that would insulate them from sanctions.
  115. Similarly, Paul Chevigny observed that in 1989, the civilian review board in New York that heard complaints against police “reported that it had never had a case in which a police witness testified against another.” CHEVIGNY, supra note 32, at 92.
  116. See supra text accompanying notes 37-39.
  117. Compare, e.g., Barak Ariel, William A. Farrar & Alex Sutherland, The Effect of Body-Warn Cameras on Use of Force and Citizens’ Complaints Against the Police: A Randomized Trial. 31 J. QUANTITATIVE CRIMINOLOGY 509, 523-24 (2015) (detailing a controlled experiment in Rialto, California, finding the use of force by officers wearing cameras was substantially lower than the rate for other officers), and Wesley G. Jennings, Mathew D. Lynch & Lorie A. Fridell, Evaluating the Impact of Body-Worn Cameras (BWCs) on Response-to-Resistance and Serious External Complaints: Evidence from the Orlando Police Department (OPD) Experience Utilizing a Randomized Controlled Experiment, 43 J. CRIM. JUST. 480, 485 (2015) (detailing a controlled experiment in Orlando, Florida, finding the use of force by and external complaints against officers wearing cameras was substantially lower than the rate for other officers), with, e.g., Barak Ariel et al., Report: Increases in Police Use of Force in the Presence of Body-Worn Cameras Are Driven by Officer Discretion: A Protocol Based Subgroup Analysis of Ten Randomized Experiments, 12 J. EXPERIMENTAL CRIMINOLOGY 453, 454-63 (2016) (outlining multisite, controlled and randomized study finding body-warn cameras had no consistent effect on police use of force and produced reductions when officers had least discretion in determining when camera was operating), Lott & Moody, supra note 95, at 15 (reviewing national dataset on police homicides and finding no relationship between presence of body cameras and homicide rate); and Min-Seok Pang & Paul A. Pavlou, Armed with Technology: The Impact on Fatal Shootings by the Police 24-25, 27-29, 34-35 (Fox Sch. of Bus., Research Paper No. 16-020, July 14, 2016) (unpublished manuscript), 2808662 (reviewing national data and finding the use of body-mounted cameras produced a lagged 3.64% increase in police homicide rate, with increases strongest for African Americans and Hispanic victims, perhaps because officers come to learn that cameras can assist in justifying the use of force).
  118. The closest that anyone has come is Donald Dripps’s proposal to permit the use of polygraph evidence at hearings on motions to suppress evidence on the grounds of police misconduct in order to reduce policy perjury, arguing that despite the error rate that inheres in the use of such evidence, it would increase the overall reliability of factfinding at suppression hearings. See Dripps, supra note 31, at 703-15. One could use the same approach in disciplinary hearings involving allegations of police misconduct, although one might wonder whether adjudicators would be willing to impose potentially career-ending sanctions based on polygraph evidence. Professor Dripps eventually concluded that judges were likely to be reluctant to use polygraph evidence to suppress probative evidence of serious wrongdoing. See Donald Dripps, The Case for the Contingent Exclusionary Rule, 38 AM. CRIM. L. REV. 1, 21-22 (2001).
  119. Cf. Armacost, supra note 1, at 494-507 (arguing that police violence should often be understood as reflecting organizational culture that tolerates violence against vulnerable populations).
  120. Cf. Carbado, supra note 8, at 1524 (“If police officers know that their violent conduct will be considered justifiable force, or that they will be immune from civil liability or indemnified if they are found civilly liable, they are less likely to exercise care with respect to when and how they employ violent force.”).
  121. See supra text accompanying notes 6-7.
  124. COOPER & SMITH, supra note 122, at 11 fig.17.
  125. BRAGA & BRUNSON, supra note 123, at 5-6.
  126. COOPER & SMITH, supra note 122, at 11 fig.18.
  127. Id. at 13 fig.19.
  128. Id. at 12 tbl.7.
  130. 130. For these calculations, the mid-year 2015 population estimates of the census bureau were used showing 247,784,609 white and 42,632,530 black or African American residents of the United States. See U.S. CENSUS BUR., DEP’T OF COMM., ANNUAL ESTIMATES OF THE RESIDENT POPULATION BY AGE, SEX, AND HISPANIC ORIGIN FOR THE UNITED STATES, STATES, AND COUNTIES (2015), The number of homicide victims by race came from FBI, U.S. DEP’T OF JUST., CRIME IN THE UNITED STATES, 2015, EXPANDED HOMICIDE DATA tbl.1 (2015), https://ucr.fbi/gov/crime-in-he-u.s./2015/tables/expanded_homicide_ data_table_1_ murder_victims_by_race_ethnicity_and_sex_2015.xls. The number of homicide offenders by race came from id. tbl.3,
  131. COOPER & SMITH, supra note 122, at 29-31, figs.46-47.
  133. See RUTH D. PETERSON & LAUREN J. KRIVO, DIVERGENT SOCIAL WORLDS: NEIGHBORHOOD CRIME AND THE RACIAL-SPATIAL DIVIDE 17-18 fig.2.1 (2010). Because a racial skew is reflected in rates of both offending and violent victimization and in statistics generated by both police and victims, it is unlikely to reflect some form of racial selection bias by police. For a helpful discussion of this issue, see BARRY LATZER, THE RISE AND FALL OF VIOLENT CRIME IN AMERICA 131-34 (2016).
  134. For useful surveys of the research on the relationship of poverty and crime, see, for example, ELLIOT CURRIE, CRIME AND PUNISHMENT IN AMERICA 120-47 (1998); and GARY LAFREE, LOSING LEGITIMACY: STREET CRIME AND THE DECLINE OF SOCIAL INSTITUTIONS IN AMERICA 117-34 (1998).
  136. See, e.g., PETERSON & KRIVO, supra note 133, at 71-109; ROBERT J. SAMPSON, GREAT AMERICAN CITY: CHICAGO AND THE ENDURING NEIGHBORHOOD EFFECT 155-78 (2012); Ronald C. Kramer, Poverty, Inequality, and Youth Violence, 567 ANNALS AM. ACAD. POL. & SOC. SCI. 123 (2000); Michael T. Light & Casey T. Harris, Race, Space, and Violence: Exploring Spatial Dependence in Structural Covariates of White and Black Violent Crime, 28 J. QUANTITATIVE CRIMINOLOGY 559, 572-75 (2012); Darrell Steffensmeier et al., Scope and Conceptual Issues in Testing the Race-Crime Invariance Theory: Black, White, and Hispanic Comparisons, 48 CRIMINOLOGY 1133, 1155-63 (2000).
  137. A 2016 Gallup poll, for example, disclosed the following percentages of subgroups defined by race and income who expressed “a great deal” of concern about crime and violence:
    National Adults  -  53
    Whites  -  46
    Nonwhites  -  68
    Less than $30,000  -  66
    $30,000 to $74,999  -  57
    $75,000+  -  36
    Alyssa Davis, In U.S., Concern about Crime Climbs to 15-Year High, GALLUP (Apr. 5, 2016),
  138. See, e.g., Ruth D. Peterson & Lauren Krivo, Racial Segregation and Black Urban Homicide, 71 SOC. FORCES 1001, 1018-21 (1993).
  139. See, e.g., KAREN F. PARKER, UNEQUAL CRIME DECLINE: THEORIZING RACE, URBAN INEQUALITY, AND CRIMINAL VIOLENCE 83-110 (2008); WILLIAM JULIUS WILSON, WHEN WORK DISAPPEARS: THE WORLD OF THE NEW URBAN POOR 22-25 (1996); Edward Shihadeh & Nicole Flynn, Segregation and Crime: The Effect of Black Social Isolation on Rates of Black Urban Violence, 74 SOC. FORCES 1325, 1341-48 (1996).
  140. See, e.g., Douglas S. Massey, Getting Away with Murder: Segregation and Violent Crime in Urban America, 143 U. PA. L. REV. 1203, 1210-29 (1995).
  141. See, e.g., SAMPSON, supra note 136, at 149-54.
  142. See, e.g., Charis E. Kubrin & Ronald Weitzer, Retaliatory Homicide: Concentrated Disadvantage and Neighborhood Culture, 50 SOC. PROBS. 157, 170-78 (2003).
  143. For helpful discussions of the difficulties in determining whether the high proportion of victims of police shootings is fairly characterized as racial discrimination, see, for example, GELLER & SCOTT, supra note 6, at 152-55, 200-19; and Sekhon, supra note 6, at 26-39. For more general discussions of the difficulties in assessing statistical evidence of racial discrimination, see, for example, Gregg Ridgeway & John McDonald, Methods for Assessing Racially Biased Policing, in RACE, ETHNICITY, AND POLICING, supra note 99, at 180, 181-98; Lawrence Rosenthal, Pragmatism, Originalism, and the Case Against Terry V. Ohio, 43 TEX. TECH L. REV. 299, 347-53 (2010); and Sonja Starr, Testing Racial Profiling: Empirical Assessment of Disparate Treatment By Police, 2016 U. CHI. LEG. F. 485, 504-18.
  144. See, e.g., Vani K. Borooah, Racial Bias in Police Stops and Searches: An Economic Analysis, 17 EUR. J. POL. ECON. 17, 32-33 (2001); Jeff Dominitz & John Knowles, Crime Minimization and Racial Bias: What Can We Learn from Police Search Data?, 116 ECON. J. F368, F368-71 (2006); John Knowles, Nicola Persico & Petra Todd, Racial Bias in Motor Vehicle Searches: Theory and Evidence, 109 J. POL. ECON. 203, 205 (2001).
  145. See, e.g., Floyd v. City of New York, 959 F. Supp. 2d 540, 583-89 (S.D.N.Y. 2013) (street stops in New York); EPP, MAYNARD-MOODY & HAIDER-MARKEL, supra note 3, at 100-06 (traffic stops in Kansas City area); Kate Antonovics & Brian G. Knight, A New Look at Racial Profiling: Evidence from the Boston Police Department, 91 REV. ECON. & STATS. 163, 175-77 (2009) (traffic stops in Boston); Roland G. Fryer, Jr., An Empirical Analysis of Racial Differences in Police Use of Force 16-21 ( July 2016) (unpublished working paper, Nat’l Bureau of Econ. Res.), (street stops in New York); Andrew Gelman, Jeffrey Fagan & Alex Kiss, An Analysis of the New York City Police Department’s “Stop and Frisk” Policy in the Context of Claims of Racial Bias, 102 J. AM. STAT. ASS’N 813, 818-20 (2007) (same); David A. Harris, The Stories, the Statistics, and the Law: Why “Driving While Black” Matters, 84 MINN. L. REV. 265, 277-81 (1999) (highway stops in New Jersey and Maryland); Sean Childers, Note, Discrimination During Traffic Stops: How an Economic Account Justifying Racial Profiling Falls Short, 87 N.Y.U. L. REV. 1025, 1030-46 (2012) (Illinois traffic stops). For more general criticisms of the economic defense of policing that disproportionately targets racial minorities, see, for example, Steven N. Durlauf, Assessing Racial Profiling, 116 ECON. J. F402, F406-07 (2006); and Bernard Harcourt, Rethinking Racial Profiling: A Critique of the Economics, Civil Liberties, and Constitutional Literature, and of Criminal Profiling More Generally, 71 U. CHI. L. REV. 1275, 1295-1314 (2004).
  146. See, e.g., NICK SELBY, BEN SINGLETON & ED FLOSI, IN CONTEXT: UNDERSTANDING POLICE KILLINGS OF UNARMED CIVILIANS 70-73 (2016); (study of police homicides of unarmed civilians in 2015 finding no evidence of targeting victims by race); Decio Coviello & Nicola Persico, An Economic Analysis of Black-White Disparities in the New York Police Department’s Stop-and-Frisk Program, 44 J. LEG. STUD. 315 (2015) (questioning evidence of bias in New York City street stops); Fryer, supra note 145, at 21-26 (finding no statistically significant racial differences for rates of officer-involved shootings).
  147. FBI, supra note 130, at tbl.1, https://ucr.fbi/gov/crime-in-the-u.s./2015/tables/expanded_homicide_data_table_1_murder_victims_by_race_ethnicity_and_sex_2015.xls.
  148. The Counted, supra note 6.
  149. Investigation: People Shot and Killed by Police This Year, WASH. POST, (last visited Sept. 15, 2016).
  150. See supra text accompanying notes 122-33.
  151. See, e.g., ALFRED BLUMSTEIN & JACQUELINE COHEN, DIFFUSION PROCESSES IN HOMICIDE 6-9 (Nat’l Crim. Just. Ref. Serv., July 17, 1999); FOX, LEVIN & QUINET, supra note 132, at 87-88; BENJAMIN PEARSON-NELSON, UNDERSTANDING HOMICIDE TRENDS: THE SOCIAL CONTEXT OF A HOMICIDE EPIDEMIC 37-41 (2008); Alfred Blumstein & Joel Wallman, The Crime Drop and Beyond, 2006 ANN. REV. SOC. SCI. 125, 131 (2006); Philip J. Cook & John H. Laub, After the Epidemic: Recent Trends in Youth Violence in the United States, in CRIME & JUSTICE: A REVIEW OF RESEARCH 1, 21-31 (Michael Tonry ed., 2002).
  152. See, e.g., Eric Baumer et al., The Influence of Crack Cocaine on Robbery, Burglary, and Homicide Rates: A Cross-City, Longitudinal Analysis, 35 J. RES. CRIME & DELINQ. 316, 328-30 (1998); Daniel Cork, Examining Space-Time Interaction in City-Level Homicide Data: Crack Markets and the Diffusion of Guns Among Youth, 15 J. QUANTITATIVE CRIMINOLOGY 379, 403-04 (1999); Jeff Grogger & Michael Willis, The Emergence of Crack Cocaine and the Rise of Urban Crime Rates, 82 REV. ECON. & STATS. 519, 525-28 (2000); Steven F. Messner et al., Locating the Vanguard in Rising and Falling Homicide Rates Across U.S. Cities, 43 CRIMINOLOGY 661, 677-87 (2005); Graham Ousey & Matthew R. Lee, Examining the Conditional Nature of the Illicit Drug Market-Homicide Relationship: A Partial Test of Contingent Causation, 40 CRIMINOLOGY 73, 94-98 (2002).
  153. See, e.g., Lallen J. Johnson, Policing Urban Drug Markets, in OXFORD HANDBOOK OF POLICE, supra note 88, at 34, 36-39. What is probably the best study of this issue—a sampling of homicides in New York City during an eight-month period in 1988—found that 52.7% of homicides were drug-related; of those, 60% involved crack, and 74% of drug-related homicides were classified as “systemic” or involving “the normally aggressive patterns of interaction within the systems of drug use and distribution” as opposed to homicides that were a function of the pharmacological effects of drugs or the economic compulsion to commit crimes to finance drug use. See Paul J. Goldstein et al., Crack and Homicide in New York City, 1988: A Conceptually Based Event Analysis, 16 CONTEMP. DRUG PROBS. 651, 655-56, 681-82 (1989). Indeed, there is relatively little evidence that drug use is itself criminogenic, though few doubt that illegality produces violent competition. See, e.g., Shima Baradaran, Drug and Violence, 88 S. CAL. L. REV. 227, 271-90 (2015) (doubting a relation between drug use and crime but acknowledging that the illegality of drugs creates illegal markets involving violent crime). Given that most controlled substances are likely to remain illegal in the foreseeable future, the discussion that follows takes it as given that violent competition surrounding drug dealing is likely to persist.
  154. See, e.g., JAMES C. HOWELL & SCOTT H. DECKER, U.S. DEP’T OF JUST., THE YOUTH GANGS, DRUGS, AND VIOLENCE CONNECTION 2-5, 7 ( Jan. 1999); C. RONALD HUFF, U.S. DEP’T OF JUST., COMPARING THE CRIMINAL BEHAVIOR OF YOUTH GANGS AND AT-RISK YOUTHS 4, 7 (Oct. 1998); HERBERT C. COVEY, SCOTT MENARD & ROBERT J. FRANZESE, JUVENILE GANGS 51-54 (2d ed. 1997); SCOTT H. DECKER & BARRICK VAN WINKLE, LIFE IN THE GANG: FAMILY, FRIENDS, AND VIOLENCE 153-71 (1996); MARTIN SANCHEZ JANKOWSKI, ISLANDS IN THE STREET: GANGS AND AMERICAN URBAN SOCIETY 120-21 (1991); JOAN W. MOORE, HOMEBOYS: GANGS, DRUGS, AND PRISON IN THE BARRIOS OF LOS ANGELES 75-93 (1978); FELIX M. PADILLA, THE GANG AS AN AMERICAN ENTERPRISE 97-117, 129-51 (1993); CARL S. TAYLOR, DANGEROUS SOCIETY 92, 97-99 (1990); Brenda C. Coughlin & Sudhir Alladi Venkatesh, The Urban Street Gang After 1970, 2003 ANN. REV. SOC. 41, 43-45 (2003); Finn-Aage Esbensen & David Huizinga, Gangs, Drugs, and Delinquency in a Survey of Urban Youth, 31 CRIMINOLOGY 565, 573-75 (1993); Jeffrey Fagan, The Social Organization of Drug Use and Drug Dealing Among Urban Gangs, 27 CRIMINOLOGY 633, 635, 648-51 (1989); Ronald Glick, Survival, Income, and Status: Drug Dealing in the Chicago Puerto Rican Community, in DRUGS IN HISPANIC COMMUNITIES 77, 77-101 (Ronald Glick & Joan Moore eds., 1990); Tom Mieczkowski, Geeking Up and Throwing Down: Heroin Street Life in Detroit, 24 CRIMINOLOGY 645, 648-49 (1986). To be sure, there is ample evidence that less structured and organized gangs are less effective drug sellers. See, e.g., MALCOLM W. KLEIN, THE AMERICAN STREET GANG: ITS NATURE, PREVALENCE, AND CONTROL 40-42, 126-29 (1995). Still, there is widespread agreement that better organized gangs have become adept at drug trafficking. See, e.g., ARLEN EGLEY, JR. ET AL., U.S. DEP’T OF JUST., NATIONAL YOUTH GANG SURVEY 1999-2001, at 33-34 ( July 2006); Scott H. Decker, Youth Gangs and Violent Behavior, in THE CAMBRIDGE HANDBOOK OF VIOLENT BEHAVIOR AND AGGRESSION 388, 392-93 (Daniel J. Flannery et al. eds., 2007); Finn-Aage Esbensen, et al., Initiation of Drug Use, Drug Sales, and Violent Offending Among a Sample of Gang and Nongang Youth, in GANGS IN AMERICA III 37, 39-40 (C. Ronald Huff ed., 2002); Malcolm W. Klein, Street Gang Cycles, in RIME 217, 219-21, 227 (James Q. Wilson & Joan Petersilia eds., 1998) [hereinafter CRIME]; Malcolm W. Klein & Cheryl Lee Maxson, Gangs and Crack Cocaine Trafficking, in DRUGS AND CRIME: EVALUATING PUBLIC POLICY OPTIONS 42, 47-52 (Doris Layton McKenzie & Craig D. Uchida eds., 1994).
  155. See, e.g. , DECKER & VAN WINKLE, supra note 154, at 159-60; TERENCE P. THORNBERRY ET AL., GANGS AND DELINQUENCY IN DEVELOPMENTAL PERSPECTIVE 110-15 (2003); C. Ronald Huff, Comparing the Criminal Behavior of Youth Gangs and At-Risk Youth, in AMERICAN YOUTH GANGS AT THE MILLENNIUM 78, 81-82 n.3 (Finn-Aage Esbensen, Stephen F. Tibbets & Larry Gaines eds., 2004).
  156. See, e.g., DECKER & VAN WINKLE, supra note 154, at 163-64; JANKOWSKI, supra note 154, at 126-29; PADILLA, supra note 154, at 129-66; IRVING A. SPERGEL, THE YOUTH GANG PROBLEM: A COMMUNITY APPROACH 47-49 (1995); Ansley Hamid, The Political Economy of Crack-Related Violence, 17 CONTEMP. DRUG PROBS. 31, 61-63 (1990); Jerome H. Skolnick et al., The Social Structure of Street Drug Dealing, 9 AM. J. POLICE 1, 16-17 (1990).
  157. BRAGA & BRUNSON, supra note 123, at 8 (citations omitted).
  158. See, e.g., DECKER & VAN WINKLE, supra note 154, at 163-64.
  160. See, e.g., id.; Bruce D. Johnson et al., Drug Abuse in the Inner City: Impact on Hard-Drug Users and the Community, in CRIME & JUSTICE: A REVIEW OF RESEARCH 13, 35-37 (Michael Tonry & James Q. Wilson eds., 1990).
  161. BRAGA & BRUNSON, supra note 123, at 6. Although most research finds no differences between the investigative efforts made by police for homicide cases involving black and white victims, some studies have reached a contrary conclusion. See id. at 6-7.
  162. See, e.g., DECKER & VAN WINKLE, supra note 154, at 175-76; JAMES C. HOWELL, GANGS IN AMERICA’S COMMUNITIES 218 (2012); JOSEPH F. SHELEY & JAMES D. WRIGHT, IN THE LINE OF FIRE: YOUTHS, GUNS, AND VIOLENCE IN URBAN AMERICA 95-103 (1995); THORNBERRY ET AL., supra note 155, at 123-25, 131; Beth Bjerregaard & Alan J. Lizotte, Gun Ownership and Gang Membership, 86 J. CRIM. L. & CRIMINOLOGY 37, 46-53 (1995); Decker, supra note 154, at 391-92; Huff, supra note 155, at 77, 83; Alan J. Lizotte et al., Factors Influencing Gun Carrying Among Young Urban Males over the Adolescent-Young Adult Life Course, 38 CRIMINOLOGY 811, 812-13 (2000).
  163. See, e.g., SHELEY & WRIGHT, supra note 162, at 75-76, 83-93; Alfred Blumstein, Youth Violence, Guns, and the Illicit-Drug Industry, 86 J. CRIM. L. & CRIMINOLOGY 10, 29-31 (1995); Lizotte et al., supra note 162, at 814-16, 826-28.
  164. See, e.g., James C. Howell, Youth Gang Homicides: A Literature Review, 45 CRIME & DELINQ. 208, 216 (1999).
  165. See BLUMSTEIN & COHEN, supra note 151, at 4-5; MARK R. POGREBIN, PAUL B. STRETESKY & N. PRABHA UNNITHAN, GUNS, VIOLENCE, AND CRIMINAL BEHAVIOR: THE OFFENDER’S PERSPECTIVE 69-71 (2009); SHELEY & WRIGHT, supra note 162, at 102-03, 110-13; Jeffrey Fagan & Deanna L. Wilkinson, Guns, Youth Violence, and Social Identity in Inner Cities, in YOUTH VIOLENCE 105, 174 (Michael Tonry & Mark H. Moore eds., 1998); David Hemenway et al., Gun Carrying Among Adolescents, 59 L. & CONTEMP. PROBS. 39, 44-47 (1996); Lizotte et al., supra note 162, at 813-14; Paul B. Stretesky & Mark R. Pogrebin, Gang-Related Gun Violence: Socialization, Identity, and Self, 36 J. CONTEMP. ETHNOGRAPHY 85, 105-08 (2007).
  166. Fagan & Wilkinson, supra note 165, at 174.
  167. See, e.g., LATZER, supra note 133, at 149-52.
  168. See Craig Reinarman & Harry G. Levine, The Crack Attack: Politics and Media in the Crack Scare, in CRACK IN AMERICA: DEMON DRUGS AND SOCIAL JUSTICE 20, 29-31 figs.2.1-2.3 (Craig Reinarman & Harry G. Levine eds., 1997).
  169. See SUBSTANCE ABUSE & MENTAL HEALTH SERV. ADMIN., DEP’T OF HEALTH & HUM. SER., EMERGENCY DEPARTMENT TRENDS FROM THE DRUG ABUSE WARNING NETWORK: FINAL ESTIMATES 1994-2001, at 50-53 fig.3 (Aug. 2002). Similarly, federal seizures of cocaine remained roughly constant from 1989 through 2002 See BUR. OF JUST. STATS., U.S. DEP’T OF JUST., SOURCEBOOK OF CRIMINAL JUSTICE STATISTICS: 2003, at 390 tbl.4.36 (Aug. 2004).
  171. See Steven D. Levitt, Using Electoral Cycles in Police Hiring to Estimate the Effects of Police on Crime: Reply, 92 AM. ECON. REV. 1244, 1244-46 (2002); Steven D. Levitt, Using Electoral Cycles in Police Hiring to Estimate the Effect of Police on Crime, 87 AM. ECON. REV. 270, 286 (1997); Thomas B. Marvell & Carlisle E. Moody, Specification Problems, Police Levels, and Crime Rates, 34 CRIMINOLOGY 609, 630-40 (1996); Steven Mello, Police and Crime: Evidence from COPS 2.0 ( July 6, 2016) (unpublished manuscript),; Emily Weisburst, Safety in Police Numbers: Evidence of Police Effectiveness and Foresight from Federal COPS Grant Applications (Oct. 2016) (unpublished manuscript),
  172. See FAIRNESS AND EFFECTIVENESS IN POLICING, supra note 6, at 224-25; JOHN E. CONKLIN, WHY CRIME RATES FELL 60-63 (2003); LATZER, supra note 133, at 222-23; SKOLNICK & FYFE, supra note 31, at 251-52; John E. Eck & Edward R. Maguire, Have Changes in Policing Reduced Violent Crime? An Assessment of the Evidence, in THE CRIME DROP IN AMERICA 207, 209-17 (Alfred Blumstein & Joel Wallman eds., 2d ed. 2006); Lawrence W. Sherman & John E. Eck, Policing for Crime Prevention, in EVIDENCE-BASED CRIME PREVENTION 295, 306-07 (Lawrence W. Sherman et al. eds., rev. 2006).
  173. See, e.g., PEARSON-NELSON, supra note 151, at 152-54; Graham C. Ousey & Matthew R. Lee, Homicide Trends and Illicit Drug Markets: Exploring Differences Across Time, 24 JUST. Q. 48, 73 (2007); Travis C. Pratt & Francis T. Cullen, Assessing Macro-Level Predictors and Theories of Crime: A Meta-Analysis, in CRIME AND JUSTICE: A REVIEW OF RESEARCH 373, 425-26 (Michael Tonry ed., 2005); see also Gary L. Shoesmith, Four Factors that Explain Both the Rise and Fall of US Crime, 1970-2003, 42 APPLIED ECON. 2957, 2971-72 (2010) (finding crime declines relate to arrest rates rather than size of police force).
  174. Mark Harrison Moore, Problem-Solving and Community Policing, in MODERN POLICING 99, 112 (Michael Tonry & Norval Morris eds., 1992).
  175. See, e.g., DAVID WEISBURD, ELIZABETH R. GROFF & SUE-MING YANG, THE CRIMINOLOGY OF PLACE: STREET SEGMENTS AND OUR UNDERSTANDING OF THE CRIME PROBLEM 50-53 (2012); John E. Eck, Preventing Crime at Places, in PREVENTING CRIME: WHAT WORKS, WHAT DOESN’T, WHAT’S PROMISING: A REPORT TO THE UNITED STATES CONGRESS 7-1 (Lawrence W. Sherman et al. eds., 1987); Lawrence W. Sherman, Hot Spots of Crime and Criminal Careers of Places, in CRIME AND PLACE 35, 36-39 ( John E. Eck & David Weisburd eds., 1995) [hereinafter CRIME AND PLACE].
  176. See supra text accompanying notes 134-42.
  177. See, e.g., WEISBURD, GROFF & YANG, supra note 175, at 9-14; Ralph B. Taylor, Crime and Small-Scale Places: What We Know, What We Can Prevent, and What Else We Need to Know, in CRIME AND PLACE: PLENARY PAPERS OF THE 1997 CONFERENCE ON CRIMINAL JUSTICE RESEARCH AND EVALUATION 3-9 (U.S. Dep’t of Just. 1998); Lawrence W. Sherman et al., Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place, 27 CRIMINOLOGY 27 (1989).
  178. See, e.g., ROBERT J. BURSIK & HAROLD J. GRAMSICK, NEIGHBORHOODS AND CRIME: THE DIMENSIONS OF EFFECTIVE COMMUNITY CONTROL 62-72 (1993); MICHAEL R. GOTTFREDSON & TRAVIS HIRSCHI, A GENERAL THEORY OF CRIME 22-44 (1990); WEISBURD, GROFF & YANG, supra note 175, at 6-9, 90-118; Joel M. Caplan, Mapping the Spatial Influence of Crime Correllates: A Comparison of Operationalization Scheme and Implications for Crime Control and Analysis, 13 CITYSCAPE 57, 60-63 (2011); Lawrence E. Cohen & Marcus Felson, Social Change and Crime Rate Trends: A Routine Activities Approach, 44 AM. SOC. REV. 588 (1979); John E. Eck & David Weisburd, Crime Places in Crime Theory, in CRIME AND PLACE, supra note 175, at 1, 7-18.
  179. See, e.g., Eck & Maguire, supra note 172, at 228-45; Debra Livingston, Police Discretion and the Quality of Life in Public Places: Courts, Communities, and the New Policing, 97 COLUM. L. REV. 551, 572-84 (1997).
  181. See id. at 108-21.
  182. See, e.g., id. at 131-44 (concluding that New York’s policing strategies associated with Compstat provide the only explanation for its unusual crime decline consistent with the data); David Weisburd et al., Do Stop, Question, and Frisk Practices Deter Crime?, 15 CRIMINOLOGY & PUB. POL’Y 31, 42-47 (2016) (finding stop-frisk-andquestion tactics produce significant crime reductions in the immediate area and aftermath of an intervention); David Weisburd, Cody Telep and Brian Lawton, Could Innovations in Policing Have Contributed to the New York City Crime Drop even in a Period of Declining Police Strength? The Case of Stop, Question and Frisk as a Hot Spots Policing Strategy, 31 JUST. Q. 129, 136-46 (2014) (discussing evidence that stop and frisk tactics focusing on hot spots enabled New York to reduce crime even as its police force shrunk). But cf. Richard Rosenfeld & Robert Fornago, The Impact of Police Stops on Precinct Robbery and Burglary Rates, 2003-2010, 31 JUST. Q. 96, 116-18 (2014) (finding no effect of rates of stop, frisk, and question on precinct robbery and burglary rates, but cautioning that use of annual data and the large size of precincts may limit the utility of the analysis).
  184. Anthony A. Braga, Andrew V. Papachristos & David M. Hureau, The Effects of Hot Spots Policing on Crime: An Updated Systematic Review and Meta-Analysis, 41 JUST. Q. 633, 655 (2014).
  185. See Blumstein & Wallman, supra note 151, at 136-37; Braga, Papachristos & Hureau, supra note 184, at 643-60; Daniel S. Nagin, Robert M. Solow & Cynthia Lum, Deterrence, Criminal Opportunities, and Police, 53 CRIMINOLOGY 74, 78-79 (2015); Richard Rosenfeld, Michael J. Deckard & Emily Blackburn, The Effects of Directed Patrol and Self-Initiated Enforcement on Firearm Violence: A Randomized Controlled Study of Hot Spot Policing, 52 CRIMINOLOGY 428, 428-30, 445-47 (2014); Lawrence W. Sherman, Reducing Gun Violence: What Works, What Doesn’t, What’s Promising, in PERSPECTIVES ON CRIME AND JUSTICE: 1999-2000 LECTURE SERIES 69, 78-79 (2001); Lawrence W. Sherman, The Police, in CRIME, supra note 154, at 327, 328-34; Sherman & Eck, supra note 172, at 295, 308-10; David Weisburd, Does Hot Spots Policing Inevitably Lead to Unfair and Abusive Police Practices, Or Can We Maximize Both Fairness and Effectiveness in the New Proactive Policing?, 2016 U. CHI. LEG. F. 661, 666-71.
  186. See FAIRNESS AND EFFECTIVENESS IN POLICING, supra note 6, at 235-40; FIREARMS AND VIOLENCE, supra note 132, at 230-35; Cody Telep & David Weisburd, What Is Known about the Effectiveness of Police Practices in Reducing Crime and Disorder?, 15 POLICE Q. 331, 333-36 (2012).
  187. Braga, Papachristos & Hureau, supra note 184, at 658, 659. For a similar assessment of the empirical evidence, see Weisburd, supra note 185, at 673-76.
  188. See, e.g., Anthony A. Braga & David L. Weisburd, The Effects of Focused Deterrence Strategies on Crime: A Systematic Review and Meta-Analysis of the Empirical Evidence, 49 J. RES. CRIME & DELINQ. 323, 349-50 (2011) (“In the focused deterrence approach, the emphasis is not only on increasing the risk of offending but also on decreasing opportunity structures for violence, deflecting offenders away from crime, increasing the collective efficacy of communities, and increasing the legitimacy of police actions. Indeed, we suspect that the large effects we observe come precisely from the multifaceted ways in which this program influences criminals.”). For a formal model defending this view of the deterrent effect of hot-spot policing, see Nagin, Solow & Lum, supra note 185, at 79-94.
  189. For a useful discussion of the difficulties in conducting assessments of the efficacy of policing strategies, see MALCOLM SPARROW, HANDCUFFED: WHAT HOLDS POLICING BACK, AND THE KEYS TO REFORM 129-70 (2016).
  190. See supra text accompanying notes 162-66.
  191. See, e.g., David Jacobs & Robert M. O’Brien, The Determinants of Deadly Force: A Structural Analysis of Police Violence, 103 AM. J. SOC. 837, 851-54 (1998) (murder rate); Legewie & Fagan, supra note 98, at 24-26 (black-on-white homicide rate); John M. MacDonald et al., The Temporal Relationship Between Police Killings of Civilians and Criminal Homicide: A Refined Version of the Danger-Perception Theory, 47 CRIME & DELINQ. 155, 165-69 (2001) (civilian homicide rate); Smith, supra note 98, at 154-58 (minority population and violent crime rates); Brad W. Smith, Structural and Organizational Predictors of Homicide by Police, 27 POLICING 539, 547-51 (2004) (violent crime rate); Terrill & Reisig, supra note 8, at 303-07 (neighborhood homicide rate and measures of concentrated neighborhood disadvantage); Jonathan R. Sorenson, James W. Marquart & Deon E. Brock, Factors Related to Killings of Felons by Police Officers: A Test of the Community Violence and Conflict Hypotheses, 10 JUST. Q. 417, 438 (1993) (violent crime rate).
  192. See, e.g., GELLER & SCOTT, supra note 6, at 191; Sekhon, supra note 6, at 20-21; Zimring & Arsiniega, supra note 5, at 257-61.
  193. See supra text accompanying notes 108-09.
  194. See GELLER & SCOTT, supra note 6, at 184-85 (studies consistently show more than two-thirds of police shootings occur in public locations); 2014 NEW YORK FIREARMS DISCHARGE REPORT, supra note 108, at 23 fig.14 (69% of 2014 police shootings occurred outdoors, primarily on streets or sidewalks).
  195. See, e.g., Adams, supra note 98, at 6-7; Joel H. Garner, Christopher D. Maxwell & Cedrick G. Heraux, Characteristics Associated with the Prevalence and Severity of Force Used by the Police, 19 JUST. Q. 705, 719-20, 738 (2002); William Terrill, Police Use of Force and Suspect Resistance: The Micro-Process of the Police-Suspect Encounter, 6 POLICE Q. 51, 74-75 (2003); Terrill & Mastrofski, supra note 91, at 240; cf. GEOFFREY P. ALPERT & ROGER G. DUNHAM, UNDERSTANDING POLICE USE OF FORCE: OFFICERS, SUSPECTS, AND RECIPROCITY 167-69 (2004) (outlining a study of police in Miami-Dade and Prince George’s Counties finding police use of force escalated in rough proportion to suspect’s resistance); Geoffrey P. Alpert & Roger G. Dunham, The Force Factor: Measuring and Assessing Police Use of Force and Suspect Resistance, in USE OF FORCE BY POLICE, supra note 98, at 45, 55-56 (similar findings for Miami-Dade County and Springfield/Eugene); TERRILL, supra note 98, at 219-20 (similar finding for Indianapolis and St. Petersburg).
  196. See, e.g., HOLMES & SMITH, supra note 3, at 10-11 (reviewing evidence that indicia of racial threat produces police brutality); Joscha Legewie, Racial Profiling and the Use of Force in Stops: How Local Events Trigger Periods of Increased Discrimination, 122 AM. J. SOC. 379, 383-393 (2016) (fatal shootings of police officers by black suspects increase police use of force against blacks); Legewie & Fagan, supra note 98, at 25-27 (measures of racial polarization and black-on-white homicide rates related to higher rates of police killings of blacks but not whites). See also Jacobs & O’Brien, supra note 191, at 850-60 (finding indicia of racial inequality predict rates at which police kill blacks in cities); Allen E. Liska & Jiang Yu, Specifying and Testing the Threat Hypothesis: Police Use of Deadly Force, in SOCIAL THREAT AND SOCIAL CONTROL 53, 60-67 (Allen E. Liska ed., 1992) (finding a positive relationship between nonwhite percentage of city population and police homicide rate).
  197. Cf. Terrill & Paoline III, supra note 92, at 205 (finding that officers who viewed their job as more dangerous received more excessive force complaints). See generally Daria Roithmayr, The Dynamics of Excessive Force, 2016 U. CHI. LEG. F. 407, 416-33 (arguing that in high-crime communities conflict between police and civilians tends to escalate and produces elevated rates of excessive force).
  198. See, e.g., HOLMES & SMITH, supra note 3, at 20 (“The salience of threats perceived directly by the police should figure prominently in their street-level behavior”); Steve Herbert, Police Subculture Revisited, 36 CRIMINOLOGY 343, 357 (1998) (“Considerations of safety strongly condition how officers define and approach situations.”).
  199. See supra text accompanying note 111.
  200. Cf. Klinger et al., supra note 108, at 209-14 (detailing block- group analysis of police shootings in St. Louis finding a curvilinear relation between local firearms crime and police shootings and suggesting that in areas with greatest firearms violence, residents and/or police may undertake greater precautions to avoid confrontations).
  201. Douglas W. Perez & William Ker Muir, Administrative Review of Alleged Police Brutality, in POLICE VIOLENCE, supra note 1, at 213, 227.
  202. See supra text accompanying notes 62-63.
  203. See, e.g., Harmon, supra note 84, at 796-98.
  204. Cf. John C. Jeffries, Jr., Disaggregating Constitutional Torts, 110 YALE L.J. 259, 267 (2000) (“While negative outcomes can readily be translated into adverse legal claims, the benefits of good performance are hard to capture. These skewed incentives may bias discretionary choices. . . . The result is a bias toward inaction, defensiveness, and bureaucratic self-protection.”).
  205. Consider, for example, the potential use of polygraph evidence in police disciplinary proceedings derived from Professor Dripps’s proposal to use such evidence in hearings on motions to exclude evidence as unlawfully obtained. See supra note 118. Even if one assumes that the use of such evidence would improve the reliability of factfinding and that adjudicators would be willing to impose serious and even career-ending sanctions on officers on the basis of such evidence, the question remains whether officers would respond to the use of such evidence in ways that might deter them in the exercise of their duties—a problem that Professor Dripps did not need to address, given that the suppression of evidence involves no direct sanction against the police. Whatever else can be said of the exclusion of unlawfully-obtained evidence, because it imposes no direct cost on officers, it involves little risk of over-deterrence. See Lawrence Rosenthal, Seven Theses in Grudging Defense of the Exclusionary Rule, 10 OHIO ST. J. CRIM. L. 525, 557-62 (2013).
  206. See, e.g., Frank Rudy Cooper, Understanding “Depolicing”: Symbiosis Theory and Critical Culture Theory, 71 UMKC L. REV. 355, 357-63 (2002).
  207. See, e.g., Frank Rudy Cooper, “The “Seesaw Effect” from Racial Profiling to Depolicing: Toward a Critical Cultural Theory, in THE NEW CIVIL RIGHTS RESEARCH: A CONSTITUTIVE APPROACH 139, 145-48 (Benjamin Fleury-Steiner & Laura Beth Neilson eds., 2006) (New York); Cooper, supra note 206, at 362-63 (Seattle, Los Angeles and Cincinnati); Lewis R. Katz, “Lonesome Road”: Driving without the Fourth Amendment, 36 SEATTLE U.L. REV. 1413, 1421-22 (2013) (New Jersey).
  208. See, e.g., RICHARD ROSENFELD, NAT’L INST. OF JUST., DOCUMENTING AND EXPLAINING THE 2015 HOMICIDE RISE: RESEARCH DIRECTIONS 5-21 ( June 2016) (finding increases in homicide in fifty-six large cities in 2015 less likely explained by drugrelated crime or release of prisoners than controversial police use-of-force incidents); Paul Heaton, Understanding the Effects of Anti-Profiling Policies, 57 J.L. & ECON. 29, 34-57 (2010) (finding decreased arrest rates for motor vehicle theft and increased rates of motor vehicle theft in minority communities following racial profiling scandals in New Jersey and Maryland); Stephen L. Morgan & Joel A. Pally, Ferguson, Gray, and Davis: An Analysis of Recorded Crime Incidents and Arrests in Baltimore City, March 2010, at 46-67 (Mar. 15, 2016) (finding sharp declines in arrests in Baltimore following protests against police violence in Ferguson, Missouri, and additional declines in arrests and spikes in shootings and homicides following the death of Freddy Gray while in the custody of Baltimore Police) (unpublished manuscript),; Stephen Rushin & Griffin Sims Edwards, De-Policing, 102 CORNELL L. REV. __, __ (forthcoming), (manuscript at 40-50) (finding increased crime rates following entry of federal consent decrees involving adoption of new policies relating to police use of force); Lan Shi, The Limit of Oversight in Policing: Evidence from the 2001 Cincinnati Riot, 93 J. PUB. ECON. 99, 101-11 (2009) (finding reduced arrest rates following Cincinnati riots over police misconduct). But cf. David C. Pyrooz et al., Was There a Ferguson Effect on Crime Rates in Large U.S. Cities?, 46 J. CRIM. JUST. 1, 5-8 (2016) (finding limited evidence of increased crime in eighteen large cities following riots over police misconduct in Ferguson, Missouri in a study of a narrower time period than examined in ROSENFELD, supra).
  209. See Justin T. Ready & Jacob T.N. Young, The Impact of on-Officer Video Cameras on Police-Citizen Contacts: Findings from a Controlled Experiment in Mesa, AZ, 11 J. EXPERIMENTAL CRIMINOLOGY 445, 454 (2015) (outlining a controlled experiment finding officers wearing body cameras were more risk-averse and undertook fewer stop-and-frisks and arrests); see also Benjamin J. Goold, Public Area Surveillance and Police Work: The Impact of CCTV on Police Behavior and Autonomy, 2 SURVEILLANCE & SOC’Y 191, 194 (2003) (stating two-thirds of British police officers interviewed reported that they became more cautious when under video surveillance); Howard M. Wasserman, The Uncertain Hope of Body Cameras, in FERGUSON’S FAULT LINES: THE RACE QUAKE THAT ROCKED A NATION 217, 219-22 (Kimberly Jade Norwood ed., 2016) (observing that body-mounted cameras could over-deter police).
  210. For a theory along these lines, see Tracey L. Meares, The Good Cop: Knowing the Difference Between Lawful or Effective Policing and Why It Matters, 54 WM. & MARY L. REV. 1865, 1875-79 (2013).
  211. See Mathew Desmond, Andrew Papachristos & David S. Kirk, Police Violence and Citizen Crime Reporting in the Black Community, 81 AM. SOC. REV. 857, 864-70 (2016).
  212. See id. at 870-71.
  213. Cf. id. at 870 (“[M]urder records should not be affected by fluctuations in citizen crime reporting, owing to the gravity and nature of the crime.”).
  214. See supra text accompanying notes 207-09. Cf. Rushin & Edwards, supra note 208, at 27-29, 34-36, 40-41 (treating a Department of Justice investigation of local police as a proxy for public scrutiny and the entry of a consent decree as a proxy for regulation and finding no consistent crime increases following investigations but significant and consistent crime increases following entry of decrees).
  215. For an account of de-policing in New York that presents a narrative along these lines, see Cooper, supra note 206, at 359-62.
  216. Cf. Slobogin, supra note 69, at 407-09, 412-14 (advocating a damages remedy to replace the rule barring the use of evidence obtained in violation of the Fourth Amendment and arguing that the political benefits of effective law enforcement will minimize the risk of over-deterrence).
  217. See supra text accompanying notes 110-11.
  218. See supra text accompanying notes 37-39, 117, 209.
  219. Cf. SKLANSKY, supra note 3, at 159-70 (arguing that top-down management of police departments impairs effective policing).
  220. Cf. SKOGAN ET AL., supra note 90, at 238 (“But if generating performance measures sounds difficult, linking the results to individual rewards is an even thornier problem.”).
  221. Compare, e.g., Adams v. Williams, 407 U.S. 143, 145-48 (1972) (upholding stop-and-frisk based on informant’s tip that suspect was armed and carrying narcotics in high-crime area), with Sibron v. New York, 392 U.S. 40, 63-64 (1968) (finding an officer’s observation of suspect conversing with known addicts over an eight-hour period insufficient to justify stop-and-frisk).
  222. See Mummolo, supra note 105, at 3-4, 16-20.
  223. Floyd v. City of New York, 959 F. Supp. 2d 540, 573 (S.D.N.Y. 2013).
  224. See supra text accompanying notes 180-82.
  225. For a discussion of the evidence reaching a conclusion along these lines, see Jeffrey A. Fagan et al., Street Stops and Broken Windows Revisited: The Demography and Logic of Proactive Policing in a Safe and Changing City, in RACE, ETHNICITY, AND POLICING, supra note 99, at 309, 333-39.
  226. See supra text accompanying notes 191-99.
  227. See, e.g., KENNEDY, supra note 3, at 69-75; SKLANSKY, supra note 3, at 139-41; Kubrin & Weitzer, supra note 142, at 175-76; Alexandra Natapoff, Underenforcement, 75 FORDHAM L. REV. 1715, 1722-31 (2006).
  228. Cf. Jeremy Gorner, Few Answers as Chicago Hit with Worst Violence in Nearly 20 Years, CHI. TRIB., Dec. 30, 2016, at A1 (following the release of a video of a controversial police shooting and new reporting requirements for street stops, in 2016, arrests declined 24% and street stops declined 82%, while homicides increased 57% and the number of shooting victims increased 46% in Chicago).
  229. See supra text accompanying notes 122-37.
  230. See, e.g., CHERYL MAXSON, KAREN HENNIGAN & DAVID SLOAN, U.S. DEP’T OF JUST., FACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE 7-9 ( June 2003); RONALD WEITZER & STEVEN A. TUCH, RACE AND POLICING IN AMERICA: CONFLICT AND REFORM 41-45 tbl.2.2 (2006); Liquin Cao, James Frank & Francis T. Cullen, Race, Community Context and Confidence in the Police, 15 AM. J. POLICE 3, 13-14 (1996); Paul Jesilow, J’ona Meyer & Nazi Namazzi, Public Attitudes Toward the Police, 14 AM. J. POLICE 67, 83-86 (1995); Michael D. Reisig & Roger B. Parks, Experience, Quality of Life, and Neighborhood Context: A Hierarchical Analysis of Satisfaction with Police, 17 JUST. Q. 607, 623-25 (2000); Robert J. Sampson & Dawn Jeglum Bartusch, Legal Cynicism and (Subcultural?) Tolerance of Deviance: The Neighborhood Context of Racial Differences, 32 L. & SOC’Y REV. 777, 798-800 (2000); Joseph A. Schafer, Beth M. Huebner & Timothy S. Bynum, Citizen Perceptions of Police Services: Race, Neighborhood Context, and Community Policing, 6 POLICE Q. 440, 455-63 (2003); Amie M. Schuck, Dennis P. Rosenbaum & Darnell F. Hawkins, The Influence of Race/Ethnicity, Social Class, and Neighborhood Context on Residents’ Attitudes Toward the Police, 11 POLICE Q. 496, 509, 513-14 (2008); Ronald Weitzer, Race and Policing in Different Ecological Contexts, in RACE, ETHNICITY, AND POLICING, supra note 99, at 118, 120-21. But cf. John MacDonald et al., Race, Neighborhood Context and Perceptions of Injustice by the Police in Cincinnati, 44 URB. STUD. 2567, 2571, 2578-81 (2007) (finding a statistically significant relation between race and perception of unjust police practices in Cincinnati while acknowledging that it had an unusual recent history of racial tension involving the police).
  231. Cf. David S. Kirk & Andrew V. Papachristos, Cultural Mechanisms and the Persistence of Neighborhood Violence, 116 AM. J. SOC. 1190 (2011) (arguing that when community residents come to see police as ineffective they are more likely to employ violence as a means of self-help).
  232. REISS, supra note 30, at 212.
  233. See supra text accompanying note 103.
  234. See, e.g., United States v. Martinez-Fuerte, 428 U.S. 543, 557-58 (1976) (“The logic of such elaborate least-restrictive-alternative arguments could raise nearly insuperable barriers to the exercise of virtually all search-and-seizure powers.”).
  235. See, e.g., Kami Chavis Simmons, The Politics of Policing: Ensuring Stakeholder Collaboration in the Federal Reform of Local Law Enforcement Agencies, 98 J. CRIM. L. & CRIMINOLOGY 489, 519-28 (2008).
  236. See, e.g., Rushin, supra note 79, at 1410-11.
  237. See supra text accompanying notes 10-11, 137.
  238. I have borrowed this terminology: “A safe harbor combines a rule . . . that particular facts comply with the law and will result in no penalty,” while “[a] sure shipwreck describes conduct that will definitely violate the law, while other facts remain subject to a standard as applied by the ex post judgment of future decision makers.” Susan C. Morse, Safe Harbors, Sure Shipwrecks, 49 U.C. DAVIS L. REV. 1385, 1391-92 (2016).
  239. See, e.g., KENNETH CULP DAVIS, POLICE DISCRETION 139-63 (1975); Anthony G. Amsterdam, Perspectives on the Fourth Amendment, 58 MINN. L. REV. 349, 416-28 (1974); Friedman & Ponomarenko, supra note 44, at 1877-89; Erik Luna, Principled Enforcement of Penal Codes, 4 BUFF. CRIM. L. REV. 515, 590-623 (2000); Christopher Slobogin, Policing as Administration, 165 U. PA. L. REV. 91, 123-25, 150-51 (2016).
  240. See, e.g., Donald A. Dripps, Criminal Procedure, Footnote Four, and the Theory of Public Choice; or, Why Don’t Legislatures Give a Damn about the Rights of the Accused?, 44 SYRACUSE L. REV. 1075, 1089-94 (1993); Friedman & Ponomarenko, supra note 44, at 1862-65; Slobogin, supra note 238, at 119-20.
  241. Cf. Harmon, supra note 84, at 792-94 (advocating the concept of “harmefficient” policing).
  242. See supra text accompanying notes 179-88.
  243. See supra text accompanying notes 222-25.
  244. See, e.g. Gelman, Fagan & Kiss, supra note 145, at 822.
  245. See, e.g., EPP, MAYNARD-MOODY & HAIDER-MARKEL, supra note 3, at 52-73; Harris, supra note 145, at 275-88.
  246. See, e.g., Developments, supra note 37, at 1723-33 (describing the emergence of policing strategies directed at generating revenue).
  247. See, e.g., Andrew Guthrie Ferguson, Predictive Policing and Reasonable Suspicion, 62 EMORY L.J. 259, 265-84 (2012); Kelly K. Koss, Note, Leveraging Predictive Policing Algorithms to Restore Fourth Amendment Protections in High-Crime Areas in a Post-Wardlow World, 90 CHI.-KENT L. REV. 301, 306-10 (2015).
  248. See, e.g., Erik Luna, Transparent Policing, 85 IOWA L. REV. 1107, 1170-93 (2000) (arguing that the use of crime mapping to set and publicly justify police strategy enhances the democratic accountability and legitimacy of policing).
  249. See, e.g., Illinois v. Wardlow, 528 U.S. 119, 123-24 (2000).
  250. See, e.g., Andrew Guthrie Ferguson, Crime Mapping and the Fourth Amendment: Redrawing “High-Crime Areas,” 63 HASTINGS L.J. 179, 214-19 (2011); David A. Harris, Factors for Reasonable Suspicion: When Black and Poor Means Stopped and Frisked, 69 IND. L.J. 659, 677-81 (1994); Margaret Raymond, Down on the Corner, Out in the Street: Considering the Character of the Neighborhood in Evaluating Reasonable Suspicion, 60 OHIO ST. L.J. 99, 116-24 (1999).
  251. For helpful discussions of how the “high-crime area” concept can be refined and given greater utility in light of emerging technology, see, for example, Ferguson, supra note 247, at 304-12; Michael L. Rich, Machine Learning, Automated Suspicion Algorithms, and the Fourth Amendment, 164 U. PA. L. REV. 871, 911-23 (2016); Koss, supra note 247, at 327-33. For a helpful review of the methods for identifying criminogenic hot spots, see Spencer Chainey, Lisa Thompson & Sebastian Uhlig, The Utility of Hotspot Mapping for Predicting Spatial Patterns of Crime, 21 SEC. J. 4 (2008).
  252. Cf. Pang & Pavlou, supra note 117, at 5, 24, 27-29, 34 (finding that departments that conduct statistical analysis of digitized crime data experience 2.5% fewer fatal shootings, departments in which officers use smartphones to access intelligence data have 2.72% fewer fatal shootings, with stronger reductions for African American and Hispanic victims, and theorizing that “technology use for intelligence analyses and access helps police officers make more informed decisions in using lethal force”); see also Bennett Capers, Race, Policing, and Technology, 95 N.C.L. REV. __, __ (forthcoming), (proposing that police make greater use of electronic surveillance to determine whether there is predication to approach a suspect and assess the suspect’s dangerousness).
  253. Cf. GELLER & SCOTT, supra note 6, at 257-67 (reviewing evidence that adoption of restrictive use-of-deadly-force policies reduces rates of police shootings).
  254. See supra text accompanying notes 30-32, 114-19.
  255. See, e.g., Gardner v. Broderick, 392 U.S. 273, 278 (1968) (“If appellant, a policeman, had refused to answer questions specifically, directly, and narrowly relating to the performance of his official duties, without being required to waive his immunity with respect to the use of his answers or the fruits thereof in a criminal prosecution of himself, the privilege against self-incrimination would not have been a bar to his dismissal.” (citation and footnote omitted)). For discussions of the scope of a public employer’s prerogative to compel officers to make statements, see, for example, Kate E. Bloch, Police Officers Accused of Crime: Prosecutorial and Fifth Amendment Risks Posed by Police-Elicited “Use Immunized” Statements, 1992 U. ILL. L. REV. 625, 664-79; Steven D. Clymer, Compelled Statements from Police Officers and Garrity Immunity, 76 N.Y.U. L. REV. 1309, 1314-41 (2001); and Robert L. Myers, Code of Silence: Police Shootings and the Right to Remain Silent, 26 GOLDEN GATE L. REV. 497, 523-32 (1997).
  256. See, e.g., U.S. SENTENCING GUIDELINES MANUAL § 3E1.1 (2016) (directing reduction of the defendant’s offense level in calculating guidelines sentence in light of the defendant’s “acceptance of responsibility”); § 5K1.1 (authorizing reduced sentence “[u]pon motion of the government stating that the defendant has provided substantial assistance in the investigation or prosecution of another person who has committed an offense”).
  257. An even more aggressive alternative would be to create a mandatory duty to report misconduct. See Futterman, Hunt & Kalven, supra note 13, at 186. The ex ante calculation of an officer in this regime, however, is less clear, since the officer may calculate that misconduct that he has witnessed may never come to the attention of the authorities. Indeed, ex ante, such a regime might even enhance the incentive to undertake efforts at concealment.
  258. Cf. Huq & McAdams, supra note 31, at 236 (“Those who lie about violent encounters must also worry about consistency between those accounts and the physical evidence.”).
  259. Although officers frequently are granted special procedural protections during disciplinary investigations that often enable them to delay investigative interviews and obtain representation, they typically do not include a right to review all evidence known to investigators prior to the time that formal charges are filed. See, e.g., Keenan & Walker, supra note 84, at 203-41; Huq & McAdams, supra note 31, at 221-26; Levine, supra note 84, at 1220-27. The preceding discussion, however, suggests that in order to have the greatest impact on officers’ ex ante calculations, they not be permitted to review available video and other independent evidence prior to an initial investigative interview. See Futterman, Hunt & Kalven, supra note 13, at 191 (“Forbid accused or witness officers from reviewing video before they submit their official statements or reports about an incident in order to ensure that officer/witness statements are based on their own observations and memory, and are not tainted by the viewing of the video.” (footnote omitted)).
  260. See, e.g., NOBLE & ALPERT, supra note 100, at 157-62; WALKER, supra note 24, at 65.
  261. See, e.g., CHEVIGNY, supra note 32, at 51-52, 80 (noting that the New York and Los Angeles police departments rarely imposed discipline for false statements by officers prior to police scandals in those cities).
  262. See, e.g., Jonathan Abel, Brady’s Blind Spot: Impeachment Evidence in Police Personnel Files and the Battle Splitting the Prosecution Team, 67 STAN. L. REV. 743, 789-804 (2015).
  263. For a useful discussion of the relevant considerations, see NOBLE & ALPERT, supra note 100, at 162-64.
  264. For a helpful discussion of the precautions that can be taken to ensure that compelled statements do not compromise the ability to bring a criminal case against the officer based on independent evidence, see Clymer, supra note 255, at 1374-81.