Policing’s Information Problem

We spend over $100 billion each year on policing in the United States, yet have very little idea of what keeps us safe. From the adoption of new technologies like facial recognition to militarization to stop-and-frisk tactics, and much else, police in the United States pursue public safety strategies without understanding the full range of costs or benefits. There is a particularly troubling tendency to ignore the social costs (the impact policing practices have on individuals)—and the distributional costs (how policing regularly falls most heavily on racial and marginalized communities). As a result, we don’t know what keeps us safe, and there is a real risk we are doing more harm than good.

This Article asks why we as a society know so little about how to assure safety, and what we can do to change this state of affairs. The Article’s thesis is that a variety of dysfunctions around the politics of policing leads to legislative stasis and executive inertia: governing officials simply defer to the police. Blind trust in the police, fear of a powerful police lobby, and a law enforcement culture of secrecy and insulation result in a serious information failure. Legislative and executive officials (and the police themselves) simply lack the information needed to address whether police are doing a good job of keeping the public safe, and so they do little or nothing to assure public safety.

The Article proposes a set of solutions to encourage the development of information about policing, including sentinel event review, information-forcing legislation, the establishment of regulatory intermediaries, and the creation of
a policing college. Its main proposal is to encourage broad use of cost–benefit analysis (CBA) about policing. CBA, though prominently used in other
areas of government, is almost nonexistent in the field of policing. Adoption of
some combination of some of these tools is essential to ensure just, effective
public safety.

Introduction

It is remarkable, to say the least, that we spend over $100 billion a year in the United States on achieving public safety—much more if one takes counterterrorism efforts into account—and yet we have far too little understanding of what sort of policing actually works.^[1] There’s plenty of assertion premised on anecdotal evidence or on experiences that may be idiosyncratic.^[2] But when it comes to hard facts, grounded in careful study, the body of knowledge is thin gruel indeed.^[3]

Even when we have a clue what works, we rarely have any sense of whether a particular policing tactic, strategy, or technology is worth the cost. The $100 billion figure represents only hard economic expenses. It neglects entirely the social costs of keeping the public safe, which are more difficult to quantify, but often are most significant: the lives of those sworn to protect us, the inevitable harms inflicted on the public by policing tactics, or the trust that is shattered in certain communities as a result of enforcement efforts that are seen as inappropriate, wrongheaded, or racially biased.^[4] There’s also little in the way of hard analysis of distributional costs: policing regularly falls most heavily on communities of color and on the poor, imposing a tax for keeping the rest of us safe that often fails to enter our calculus at all.^[5]

Take stop-and-frisk as an example. Over the course of eight years, New York City police conducted more than four million pedestrian stops.^[6] To what end? The extent of any benefits is deeply contested; the social costs enormous and largely unvalued in any systematic way.^[7] Given that the New York Police Department (NYPD) has reduced those stops by over 90% and that crime, including gun violence, remains low, NYPD officials themselves today recognize the enormous toll imposed by aggressive enforcement efforts and doubt the necessity.^[8] Might we have avoided this in the first place?

The government’s use of emerging surveillance technologies poses similar questions of worth. Consider face recognition. One can see
the theoretical value of face recognition to apprehend wanted people or forestall crime. But it also has the potential to vastly erode personal
privacy, and experience thus far indicates room for deep concerns about racial bias.^[9] To date, there has been no meaningful weighing of anticipated
benefits against social costs, yet some policing agencies are forging ahead
blindly nonetheless.^[10]

One need not rely on controverted examples of policing tactics and technologies to see how in the dark we really are. Those in public safety face myriad questions daily about what tactics, strategies, policies, and technologies to employ, many of which are quite quotidian or noncontentious, and still the answers are unknown. What about using drones to map traffic accidents? Is it better to meet policing needs by paying overtime to officers already on the force, or by hiring more new officers? How about having more detectives versus patrol officers—which is more valuable to the department and what is the proper balance? Do crime analytics labs pay off? How about aviation units?^[11]

The recent spate of nationwide protests following the killing of George Floyd underscores the cost of our lack of information around policing. It is now clear that, at least in the eyes of many, policing frequently fails those most in need of the government’s help; that we’ve invested in the use of force and law over other needed social services; and that many in policed communities believe the harms imposed by policing exceed the benefits it provides. Those in the streets may possess no more information than we as a society generally otherwise hold, but the fact that so many have taken to the streets to protest policing surely is an indicator that something serious is amiss. And it is. We simply lack the most fundamental information about what in policing works and what doesn’t. To put a point on it, when it comes to policing, we are shockingly ignorant.

This lack of comprehension about what is needed to keep us safe is stunning if one stacks it up against the vast learning we have acquired in other areas critical to our well-being. In environmental science, immunology, workplace safety, and many related fields, we have developed huge literatures on how things work, and what we ought to do to protect ourselves.^[12] And yet, in the core area of policing—of protection against violence and disorder that has the potential to reduce our safety and sense of security dramatically—the paucity of hard knowledge is notable.^[13]

The question this Article addresses is why we are so in the dark about policing and the practice of keeping ourselves safe, and what we can do to address the problem. Its thesis is that because of a variety of dysfunctions around the politics of policing, decisionmakers and the public-at-large generally have not even tried to elicit or promote the development of the sorts of information needed to make wise choices, let alone publicize meaningfully what we do know. This Article then offers a variety of solutions to produce, and make available, information regarding the provision of public safety.

A, if not the, central theme of this Article is the urgent need to address the social costs of policing, i.e., the impact of policing practices on individuals. Even if we have an idea of what is efficacious, it is essential to ask “at what cost,” and make decisions about how to keep ourselves safe that factor in the very real harms policing can entail.^[14] Absent this, the choices we make to pursue public safety may be exceedingly poor ones, likely to impose substantial harms.

Part I examines the question of whether we are policing optimally, making the case that there is every reason to think not. This is true of the most prominent and persistent methods, from police patrol to broken-windows policing. And it also is true of particular tactics, such as militarization or the use of surveillance technologies. Not only is there serious doubt whether these approaches are effective, but once the social costs of the techniques are considered, there is very real reason to worry that we are doing more harm than good.

Part II then asks why, in an area as vital as public safety, we nonetheless might be proceeding suboptimally. Public safety affects everyone, and we have every incentive to get it right. But we don’t. Instead, we largely live with legislative stasis and executive inertia: legislatures and executive officials simply defer to the police. Why? Because blind trust in the police, a reluctance to get crosswise with them, and policing’s tendency to be secretive and insular have left us in a puddle of ignorance. We simply lack the information we need to assess whether those charged with keeping us safe are doing a good job and acting effectively. So, legislative and executive officials tend to do nothing, leaving the police to proceed as they will. Indeed, the spate of legislative action after the killing of George Floyd and other Black individuals, and the protests that have followed, only serves to draw attention to the inaction of the preceding decades. And in most places, even these reforms are relatively minor.

Part III identifies measures we could and should take to address the information problem around policing, and thus do a better job of assuring public safety. These range from the use of information-forcing legislation, to sentinel event review, to the establishment of regulatory intermediaries charged with policing the police, to the creation of something akin to the United Kingdom’s national College of Policing. Central to our discussion, though, is the need for cost–benefit analysis (CBA). The use of CBA is prominent in government and private industry, but when it comes to policing, CBA is radically underemployed, by researchers and practitioners alike.^[15] There is some scholarship that discusses CBA in the criminal justice system generally, but very little that is directed particularly at policing.^[16] One of our central goals here is to change that, in part by offering up an accessible way for practitioners to use CBA in daily decision-making, and in part by directing scholars to the questions they should pursue.

Whatever the course of action, continuing to do nothing to truly educate ourselves about policing is feckless. Although our politics around policing are broken, and have been for some time, stasis is not an acceptable response. There are plenty of initiatives that could be pursued, even in the face of political dysfunction. Some rest entirely in the hands of the police and researchers; some require the broader action of government. But in all cases, there is every incentive to pursue them. In an important sense, this Article, by shining a light on all we do not know, is a necessary step toward addressing policing’s information problem.

I. Are We Policing Optimally?

Are we doing what we need to keep ourselves safe and secure?
There are at least two sides to that question: is the government acting effectively to reduce crime and violence, and is it doing so in a way that
minimizes the various costs attendant to policing society?^[17] There is plenty of evidence that we may be policing in deleterious and suboptimal ways—creating severe financial, emotional, and psychological costs—while insufficiently advancing public safety. Subpart I(A) looks at this question through the lens of the most common policing approaches: patrol and “proactive” policing. Subpart I(B) then takes a deeper dive into some specific policing strategies and tactics such as militarization and the use of technology.

A. The Bread and Butter of Policing

Any analysis of the effectiveness of policing properly begins with patrol, the “backbone” of policing.^[18] In the early days of policing, officers walked the beat.^[19] With the advent of the automobile, patrol became motorized.^[20] Radios brought the capability for officers to stay in contact with headquarters and vice versa.^[21] Then, beginning in the 1980s, 911 systems were developed and officers were dispatched to respond to calls for service.^[22]

Randomized patrol once was seen as the be-all and end-all of assuring public safety. O.W. Wilson, one of the visionary leaders of policing and the one-time superintendent of the Chicago Police Department, declared patrol “an indispensable service that plays a leading role in the accomplishment of the police purpose.”^[23] Patrol created an impression of police omnipresence, persuading bad types that “opportunities for successful misconduct did not exist.”^[24] “It is the only form of police service that directly attempts to eliminate opportunity for misconduct.”^[25]

It was not until the 1970s that Wilson’s intuition was tested in the Kansas City patrol experiment. The chief of the Kansas City, Missouri police department at the time was Clarence Kelly (later Director of the FBI), who explained, “[m]any of us in the department had the feeling we were training, equipping, and deploying men to do a job neither we, nor anyone else, knew much about.”^[26] One of the researchers was George Kelling, later a patriarch of broken-windows policing.^[27] The research team broke Kansas City into three districts: a “reactive” district where there was no preventative patrol, and thus “substantially reduced police visibility”; a proactive district with two-to-three times the usual level of patrol, and thus increased visibility; and a control district.^[28]

What the Kansas City experiment revealed was rather startling: randomized patrol made pretty much no difference at all.^[29] Wilson’s statements were far too strong, if not altogether incorrect. Among the
three patrol districts, there were no significant differences in levels of crime, no real change in citizen’s fear of crime, and no difference in police
response times.^[30]

Incongruously, although the methodology of the Kansas City experiment has been the subject of sharp challenge, its results generally are taken as correct; and even so, random patrol still is a lot of what police do.^[31] Policing today tends in many places to follow what Lawrence Sherman has called the three Rs: random patrol, rapid response (to 911 calls), and reactive investigation.^[32] Countless studies indicate that when police are not filling out reports or taking personal time, they mostly are on patrol or answering calls.^[33] Yet, as former Baltimore cop and John Jay College of Criminal Law professor Peter Moskos points out, having cops out patrolling and waiting for calls for service keeps them from doing a variety of valuable things, including walking around neighborhoods, building relationships, and troubleshooting problem areas.^[34]

While patrol and answering 911 calls are relatively reactive, by the late 1980s and early 1990s, some departments across the nation began to shift toward two very different “proactive” strategies: hot spot and quality of life policing. The first involves stepping up patrol and enforcement at criminogenic hot spots. Hot spot policing got its start after an experiment conducted by Lawrence Sherman and David Weisburd—two of the leaders of the evidence-based policing movement—in Minneapolis in the early 1990s.^[35] They found that there was 25% less disorder in the areas that doubled the intensity of patrols than in the control areas, and no displacement into surrounding areas.^[36] The second is aggressive use of misdemeanor arrests for so-called “quality of life crimes,” including turnstile jumping, loitering, and panhandling. This tactic has gone by a variety of names, from “zero tolerance,” to “order maintenance,” to “broken windows,” the latter a reference to a famous article in the Atlantic by George Kelling and James Wilson arguing that disorder in a community leads to serious crime.^[37]

There is plenty of evidence to support the value of hot spot policing. Many studies conclude that focusing police efforts in very small areas will drive down crime without displacing it to other areas.^[38] The puzzling thing is how many agencies remain wedded to random patrol and reactive policing nonetheless—underscoring our basic point here.^[39]

In sharp contrast to hot spot policing, the evidence on quality-of-life policing is mixed at best: there’s room for skepticism that it has any particular value.^[40] It’s hard to understand what the causal mechanism would be that would explain why making low-level misdemeanor arrests on a city-wide basis brings down more serious crime, particularly in isolated areas. It is sometimes said that people who commit serious crimes commit other offenses too.^[41] But that logic is backward: the question is whether randomly arresting people for committing minor offenses is likely to ensnare serious criminals. The National Research Council’s comprehensive review of broken-windows policing studies ultimately concluded that they did not provide evidence of effectiveness in reducing serious crime.^[42]

Still, it is an error of huge proportion with regard to both quality of life and hot spot approaches to assess them by focusing solely on crime reduction, while neglecting the social costs proactive policing can impose—as well as whether alternative strategies could achieve the same benefits at lower costs. Whether we are safe is a function not only of bringing down crime, but what harms or costs policing imposes by doing so. Half a story is no story at all.

Both hot spot policing and quality-of-life policing frequently involve a resort to high volume pedestrian and traffic stops, used not only to investigate crime, but ostensibly to deter it.^[43] Putting aside the dubious constitutionality of most of this, the evidence of the efficacy of using such stops to fight crime more broadly is mixed at best.^[44] On the other hand, the costs of these stops are potentially vast and even now something we are only beginning to fully understand. There are the obvious costs: lost time, psychological stress, racial profiling harms, and lost community trust.^[45] But recent studies suggest widespread police stops may have had an impact on individual health, educational attendance and performance, and voter turnout.^[46]

Then there are the arrests. Both quality-of-life and hot spot approaches often rely on frequent arrests for low-level misdemeanors.^[47] This has meant dragging countless people into the criminal justice system that did not need to be there. Even if the penalty is not large for any given misdemeanor, the cumulative effect is a growing criminal record, difficulty finding a job, and a spiral that can lead to serious incarceration and disruption of families’ lives.^[48] There’s been far too little effort to value these costs, but what we know indicates they can be substantial.^[49]

These costs of proactive policing fall most heavily on low-income communities and communities of color.^[50] For instance, a 2014 analysis found that out of the millions of summonses issued in New York City from 2001 to 2013, about 81% of the recipients were Black or Hispanic.^[51] The correlation between race and summons was strongest for offenses like spitting, disorderly conduct, loitering, and failure to have a dog license.^[52] The same is true of stops; study after study shows the racially disproportionate impact of police-stop tactics.^[53]

Yet, at the same time as minority and low-income communities incur the costs of proactive policing, there is strong evidence those same communities suffer from underpolicing of more serious crimes, with the perverse result that they remain plagued by crime and violence.^[54] Police
offer fewer services to victims in high-crime areas, are slower to respond
to 911 calls, and provide less diligent investigations of cases involving minority victims.^[55] Minority populations consistently express dissatisfaction with their police departments for delayed response times, lack of focus on major crimes, and an overall view that police are not committed to protecting
their communities.^[56]

There’s an ultimate irony here, because existing research suggests that what really would matter in these communities—and particularly in the hot spots in these communities—is actually fixing broken windows, not simply conducting countless stops or arresting as many people as possible for low-level offenses. Anthony Braga and Brenda Bond did a careful study of hot spot policing in Lowell, Massachusetts, in which they estimated the effects on the number of calls for service the police get of very different hot spot approaches: misdemeanor arrests, providing social services, and “situational” efforts (i.e., paying attention to existing structural problems like lighting or abandoned houses).^[57] What they learned is that although the arrests in hot spots have some payoff, what works even better are situational approaches—which is to say addressing physical disorder, such as poor lighting, unkempt property, or abandoned buildings.^[58] Similarly, Braga, Andrew Papachristos, and David Hureau did a meta-analysis of the hot spot studies that indicated “problem-oriented” approaches—trying to solve situational issues—paid off twice as much in crime reduction as “traditional” policing.^[59]

The problem, once again—as the Braga, Papachristos, and Hureau piece highlights—is that even when we do have studies and data on policing, it tends to be about crime reduction alone, and too often neglects any assessment either of social costs, or nonenforcement approaches. They concluded, “we need to know more about the effects of hot spots policing approaches on the communities that the police serve,” and that they were “surprised” that none of the studies they evaluated “conducted formal cost-benefit assessments.”^[60] They should not have been so surprised; it is hard to find formal cost–benefit assessments of anything in policing.^[61]

In short, the bulk of policing’s bread-and-butter approaches look to be flat out inefficacious, hot spot policing being the exception. But all proactive approaches can impose substantial social costs, depending on how they are conducted. Yet, we police too often without considering these social costs, casting doubt on whether we are doing more harm than good.

B. Specific Tactics and Technologies

Here we turn from the broad approaches of modern-day policing
to some of the more specific tools the police use to reduce crime and assure public safety. Our focus is on militarized policing and adoption of
new technologies. What we see, again, is that it is dubious whether those tactics and technologies keep us safer, yet they often seem to involve substantial costs.

1. Militarized Policing

Over five billion dollars’ worth of military equipment has been transferred from the federal government to policing agencies via grant programs or simple donations, under the Defense Department’s “1033” or other programs.^[62] Some is mundane equipment like desks and computers. But much is what we associate with military action: armored personnel carriers, M-16 assault rifles, grenade launchers, and infrared gun sights.^[63]

Some military equipment undoubtedly has its uses for policing, but the benefits of transferring all this equipment are largely speculative. Two studies show small—almost miniscule—benefits in terms of a drop in crime.^[64] A careful re-evaluation of those studies indicated the drop was spurious.^[65] There may well be financial benefits to jurisdictions if the military equipment supplants otherwise necessary costs, but there has been no systemic review of this question. Much of the military-grade equipment plainly is not what departments otherwise would (or even could) acquire.^[66]

On the other hand, there is good reason to worry about the social costs of militarizing the police. They might include escalating violence in response to use of heavy arms, negative public reaction or anxiety in response to militarized police, and the like. The reaction to the televised images from Ferguson, Missouri—the first view many in this country had of how militarized the police had become—raised serious concerns in many quarters.^[67] There’s been no formal attempt to evaluate these costs.^[68]

There is one use of militarized policing that deserves particular attention because of its prominence: Special Weapons and Tactics (SWAT). SWAT teams were created in the 1960s to address emergencies, such as hostage situations and active shooters.^[69] But by the 1980s, when the federal government started donating military equipment to local police forces, and the War on Drugs was in full force, SWAT use became much more frequent. Police now use these “tactical” teams for a broad array of situations, from those deemed “high risk”—such as serving search warrants and confronting “heavily fortified” drug houses^[70]—to the much less so: a SWAT raid was deployed in Arizona to arrest a man keeping chickens illegally in his backyard, and in Illinois, a SWAT team took someone into custody for teasing a politician on Twitter.^[71] Some jurisdictions use them to serve many of their drug warrants.^[72]

There has been very little empirical research on the effectiveness of tactical teams in reducing crime or improving public safety.^[73] One study found that 40% of SWAT teams deployed for drug offenses recovered no contraband at all, and it is unknown whether contraband was located in another 29% of cases.^[74] In 2012, about two-thirds of the raids in Maryland led to arrests, yet only a small percentage of those arrested ever were charged with a felony.^[75] We certainly have no idea whether SWAT teams were even necessary in any of these arrests or whether a less militarized approach would have been equally, or perhaps more, effective.

On the other hand, the costs of using SWAT teams can be astronomical. The hourly cost of SWAT raids can run to thousands of dollars.^[76] The toll in human lives is significant: an investigation by the New York Times identified at least eighty-one civilian and thirteen law enforcement deaths from SWAT raids from 2010 to 2016, with many others being severely injured.^[77] A recent tragic example is the killing of Breonna Taylor due to a botched raid in Louisville, Kentucky.^[78] Settlements for SWAT raids gone wrong have cost municipalities hundreds of thousands, or even millions, of dollars.^[79]

The ultimate issue with SWAT deploymeny is whether its use in particular situations achieves necessary goals net more safely, both for officers and for the public at large, but on this vital question we have little information.^[80]

2. Monitoring Technology

Rapid adoption of technology by police mirrors that of the rest of society. Digital technologies are being used widely to monitor the populace, but once again, it is unclear whether these technologies are beneficial or problematic, particularly when one considers the social costs, which too often are ignored.

a. Automated License Plate Readers.—Automated license plate readers (ALPRs) are a case in point. These are digitized cameras, fixed or mounted on police cars, which “read” license plates and compares them to various “hot” lists.^[81] If an ALPR read matches a number on a list, an alert is sent to officers.^[82] Departments also use ALPR data to detect the direction and speed a person traveled; over time, the stored data can reveal a vehicle’s (and its user’s) travel patterns.^[83]

Law enforcement agencies report a broad range of benefits, including an increased recovery rate of stolen vehicles and arrests, and even a boost
in officer productivity.^[84] There also is some evidence that publicity surrounding the installment of an ALPR system can reduce crime. Reports
of stolen cars in New York City declined from 17,855 in 2005, the year
before ALPRs were first introduced in the city, to 10,334 in 2010, and the former chief spokesperson for the NYPD attributed this drastic decrease in part to ALPRs.^[85]

On the other hand, several studies question the benefits of ALPRs, claiming that this expensive technology has, at best, a minimal impact on crime reduction. One study in northern Virginia concluded that there was no statistically significant deterrent effect of ALPR deployment, and an analysis of ALPR use in Mesa, Arizona found no evidence that the technology reduced vehicle theft rates.^[86]

Why departments even have ALPRs is an ever-changing story, reflecting the all-too-common phenomenon of departments acquiring the latest technology, then figuring out what to do with it. Auto theft was the primary justification initially, but in light of declining car thefts in the United States. and doubts about ALPR efficacy, the device now often is used either to identify drivers with outstanding warrants—many of which are for traffic offenses, minor fines, and even insurance lapses on the car—or as evidence in ordinary crime enforcement.^[87] Some jurisdictions look to be geofencing their communities by putting ALPRs on major routes in and out of town.^[88]

ALPRs are expensive though, which obviously has to be weighed against any benefits. Each camera costs between $10,000 and $16,000, and that is before the costs of installation, maintenance, training, and data storage.^[89] An entire ALPR system in even a small jurisdiction can therefore cost up to $400,000.^[90]

Camera costs may well be coming down, which will only make
the technology more pervasive, but the costs of ALPRs move well beyond monetary outlays. They raise serious privacy concerns because they collect every license plate that passes by, though only a miniscule fraction of
plate scans are linked to criminal activity or even vehicle registration
issues.^[91] Jurisdictions store this information for periods of time. The company Vigilant Solutions (now owned by Motorola) collects millions of license plate reads, stores them, and has data-sharing agreements with many departments.^[92] The Electronic Frontier Foundation, a nonprofit organization dedicated to protecting individuals’ privacy rights, has suggested that the ALPR data in the aggregate could chill First Amendment activity, as it “can be used to target drivers who visit sensitive places such as health centers, immigration clinics, gun shops, union halls, protests, or centers of religious worship.”^[93] There is evidence that individual officers have abused the
ALPR technology, including looking up license plates of cars parked near
an LGBTQ bar to blackmail the car owners, or searching the database to track romantic interests.^[94]

One of the largest costs of ALPRs is their uneven deployment. Mobile ALPRs travel on police cars, making it inevitable that their usage will fall most heavily on the most policed communities.^[95] When the technology is used not for serious crime, like auto theft, but to enforce warrants for low‑level offenses, or for fine enforcement—which is often the case—there is also the potential cost of racial bias and loss of community trust as a result of overpolicing.^[96]

b. Bulk Data Collection at the Federal Level.—The uncertain value of the use of policing technology is not just a local problem. The federal government’s bulk collection of our telephone call metadata, with the objective of detecting terrorist threats and foiling them before they occur, provides an example of a costly national program that appears not to have paid off.^[97]

Although government officials across party lines defended the bulk data program once Edward Snowden revealed its existence, the classified nature of the NSA’s work made it difficult to determine effectiveness.^[98] There were many who disputed the value of the program. A 2014 New America Foundation investigation called claims that bulk data collection thwarted many attacks “overblown” and “misleading,” concluding that bulk data collection had no discernable impact in preventing terrorist attacks, and only played a minor role in 1.8% of the terrorism cases studied.^[99]

Bulk data collection did involve significant costs, however. A survey conducted in 2015 found that 61% of those aware of the federal surveillance programs had become less confident that the programs were in the public’s best interest.^[100] Similarly, a 2018 survey found that respondents generally had “more confidence in other institutions, such as cellphone manufacturers and credit card companies, to protect their data” than they had in the federal government.^[101] The U.S. economy may have suffered as a result of the bulk data collection program. The New America report found that the programs led to declining sales for American companies overseas and a flight from American cloud computing companies as foreign companies sought to move their data outside of the United States.^[102]

Despite vigorous defenses of the program, and the untold time and money spent on it, the NSA quietly stopped using the program in 2019, reporting that it had deleted all the data collected since 2015 because of “irregularities” in collection.^[103] Indeed, the Ninth Circuit Court of Appeals recently held that the program was illegal and possibly unconstitutional.^[104]

c. Body-Worn Cameras.—Body-worn cameras (BWCs) provide a good place to close out this discussion because they were adopted explicitly to assist with the social costs of policing. Although some departments already were looking at body cameras prior to the events in Ferguson, those events spurred adoption after President Obama allocated federal dollars to assist departments in equipping officers with BWCs.^[105] The hope was that BWCs would reduce the use of force, lower complaint rates against the police, and help resolve those that were leveled.^[106] A primary purpose of adoption was to provide transparency—particularly around police use of deadly force—and improve trust in, and legitimacy of, the police.^[107]

BWCs turned out to be a hugely expensive proposition. Acquiring the cameras themselves proved the small part. Storage costs for footage were very high, as well as personnel costs to deal with the program, including redacting footage for release.^[108] These costs are so high that some departments have shelved their plans to adopt BWCs or even abandoned programs in progress.^[109]

Already, there is doubt that BWCs are fulfilling their promise. A careful randomized control trial in the District of Columbia found that BWCs did not reduce force or complaints.^[110] A multinational study in Europe showed reductions in some departments and not others, though this ultimately turned out to depend in part on the underlying policies about when officers turned on their cameras.^[111]

What has become clear is that BWCs can have unintended consequences that already are alienating communities affected by heavy policing and have given rise to skepticism among social justice advocates that BWCs were the way to go. In some jurisdictions, such as Phoenix and Las Vegas, it turns out that arrests go up among officers equipped with BWCs.^[112] Why this is so is unclear. Perhaps officers feel compelled to arrest for things they would have let go, given that the incident is taped. Or perhaps they feel on firmer ground arresting with a record to back them up. Whatever the case, a higher number of arrests, particularly in marginalized communities and communities of color, is hardly why anyone supported BWCs. They also are being used to provide evidence in criminal trials; the camera points to the public, not the police, and is being used to help the police and prosecution.^[113] There are serious discussions of using BWCs equipped with facial recognition, which has raised very acute civil liberties concerns.^[114] Several organizations that might have been expected to support BWCs as a police accountability measure (or in fact did support them) now are dubious or oppose them.^[115]

Perhaps most important, it is not at all clear BWCs are providing the transparency that was intended by their adoption. Several jurisdictions
that have had officer-involved shootings of members of the public have
not released footage, released partial footage, or released footage under duress.^[116] It turns out many departments adopted BWC policies that have very stringent limits on release, either to complainants about police behavior
or after a police shooting.^[117] Some states even have passed laws to limit footage release.^[118]

* * *

This Part has raised serious questions about whether the tactics and technologies of policing are making us as safe as we might be, let alone at what cost. The examples discussed here could be extended to many other aspects of policing—for example, to the use of closed-circuit television (CCTV), predictive policing algorithms, or facial recognition. The short of it is, we spend a great deal of money on achieving public safety, but it is not at all clear we are doing the best we could be, and too often the choices we make might do more harm than good.

II. Explaining the Failure to Optimize Policing

Why is it that we do such a poor job of assuring policing occurs in a way that actually furthers public safety? In subpart II(A) we frame our exploration of this question against the familiar public choice story told about harshness in the criminal justice system. We argue that unlike prosecution and sentencing, the distribution of costs and benefits around policing should yield attentive and sound decision-making. Instead, what we see is stasis and inaction on the part of legislative and executive officials. Subpart II(B) asks what explains this inaction, concluding that the political dynamic around policing both suffers from, and exacerbates, a serious failure to produce or elicit the information needed to regulate policing practices. Legislative and executive officials—and society at large—habitually defer to those who police us, without insisting that they (or anyone) provide the information necessary to evaluate the choices that are made. We simply trust the police or are too afraid to question their tactics. But because the police themselves are not motivated to obtain the necessary information, the result all too often is—as we saw in Part I—decision-making in the dark.

A. The Public Choice Account of Public Safety and Policing

We begin with the tale, deeply rooted in public choice theory, as to why the criminal justice system as a whole leans toward harsh and punitive outcomes.^[119] This account, which by now has become common wisdom, is told primarily in the context of prosecution and sentencing.

1. The Oft-Told Tale of Legislative and Prosecutorial Harshness

According to the now familiar public choice account, legislators—who only pursue legislation if their efforts can be translated into voter support or campaign contributions—tilt toward seeking tough-on-crime policies and longer sentences.^[120] That’s because first and foremost, voters want to be safe. They get their information from a media that sensationalizes crime and promotes fear.^[121] This feeds into voters’ cognitive biases, leading them to believe they are at risk.^[122] In response, those voters demand harsh punishment.^[123] Powerful interest groups, such as prosecutors, support these harsh measures, largely because more crimes on the books and higher sentences make it easier to charge and convict criminals, and in this way prosecutors—many of whom also face election—can satisfy voters’ demand for toughness.^[124] Legislators are willing if not eager to meet voter demand because appearing “soft on crime” has proven to be a threat to legislators’ electoral longevity.^[125] Legislators and prosecutors also pander to the public, fear mongering about crime and feeding the medicine of toughness that the public seems all too eager to swallow.^[126]

Absolutely central to this story is the fact that harshness in the criminal justice system falls largely on those lacking political clout: minority populations and people in or headed to prison.^[127] Without sufficient political power, those affected by the machinery of criminal justice cannot turn the system around, even if it is operating in suboptimal ways. They simply bear the costs most of the rest of us avoid.

The result is a vicious spiral of harshness. Voters demand tough-on‑crime policies. And officials yield to voter demands, even if the outcome is a criminal justice system that is far tougher than necessary to keep us safe.

This story is breaking down somewhat at present: there is important movement toward incarceration and sentencing reform. Advocates successfully have made the case that overcharging and tough sentences cost voters money that could be spent elsewhere, and ruin lives unnecessarily.^[128] Several states have passed laws that reduce prison and jail sentences and increase opportunities for early release from prison.^[129] Other states are creating statewide criminal justice reform offices and restoring formerly incarcerated people’s voting rights.^[130] Prosecutors nationwide are running and winning on progressive campaign platforms, including pretrial diversion and less punishment.^[131] The First Step Act passed in 2018 with overwhelming bipartisan support: it expands early-release programs for individuals incarcerated in federal prison and reduces some mandatory minimum federal sentences.^[132] Nonetheless, the familiar narrative is strong enough that reform appears to have hard limits, and there is reason to worry about backsliding should crime rates begin to rise appreciably.^[133]

2. How Policing Is Different

The distribution of societal costs around policing differs sufficiently from criminalization and sentencing that we ought to expect a different set of legislative and executive outcomes.

Crime isn’t evenly distributed in society, and therefore neither are the benefits of fighting it, which might lead to indifference about what the police do. Much of the crime that motivates politics, such as homicide, disproportionately plagues poorer and more marginalized communities.^[134] So, if policing operated rationally, more policing would benefit those communities disproportionately as well. Given that, one might assume there would then be less appetite among politicians to spend additional money or pay sufficient attention to reducing crime in neighborhoods where most legislative and executive officials do not live.

But rationality is not the order of the day here; the same fearmongering that plays out around criminalization and sentencing takes hold around policing too. When crime rates rise, or when voters are scared into worrying about them, legislators respond by getting tough on crime generally, calling for more aggressive policing.^[135] We don’t necessarily get better policing; we just get more of it.

Still, the wider dispersion of costs around policing ought to trigger a different response from elected officials, keeping policing’s excesses in check and insisting upon a better job of achieving public safety. Unlike prosecution and sentencing, policing falls on a much broader swath of people, most of whom either are entirely innocent of wrongdoing or at least have committed only the most minor of offenses, such as traffic violations. William Stuntz, one of the foremost scholars of public choice around public safety, makes the point that (for example) the police stop tens of millions of people annually, and “no politician can afford to ignore the interests of that many constituents.”^[136]

That’s not entirely fair because—as Stuntz himself recognizes—policing (like crime) unquestionably falls more heavily on marginalized communities.^[137] For decades, police have stopped, searched, surveilled, and arrested a population that is disproportionately comprised of racial minorities.^[138] Even at large numbers, this is a community that historically has had less voice.^[139]

Nonetheless, there is enough truth to Stuntz’s broader point that we’d expect policing to be at least somewhat responsive to it. Even if we aren’t all stopped each year by the police, many of us get stopped at some point.^[140] We know what it is like to feel a cop on our tail and can comprehend the uncomfortable touch of suspicion. Most of the populace experiences the dread of police enforcement to some degree, even if relatively minor.

More important, modern-day policing is changing in ways that causes (and increasingly will cause) its burdens to fall on a larger segment of the population. As Chris Slobogin aptly puts it, we are becoming a society of panvasive surveillance.^[141] Policing has been moving from a reactive catch‑the-suspect model to a more proactive, deterrent approach that snares more and more people.^[142] This net of deterrence is woven of CCTV, automated license plate readers (ALPRs), monitoring of social media, drunk driving and security checkpoints, airport security, administrative searches of small and large businesses alike, with much more on the way.^[143] These tactics undoubtedly will continue to fall more heavily on overpoliced populations. Still, deterrent-based policing lands on more of us, more often.

In addition, the “error costs” of wayward policing would seem to favor careful legislative and executive control of policing. “Error costs” refer to the mistakes of policing, either by engaging in enforcement against those who are undeserving or failing to police where policing is needed. When it comes to crime definition and sentencing, if the system is making errors incarcerating those who should not be—or incarcerating them for far too long—very few people experience this. Given the perceived benefits, the majority supports harshness.^[144] Although the false positives of policing undoubtedly fall more heavily on more-policed communities, consistent with the argument above, more of us still can relate to the problems of overpolicing, are experiencing it under the surveillance state, and should be demanding controls on policing.

But most consequentially, the false negative error costs—the failure of the police to do a good job of detecting or fighting crime—presumably would lead the populace to insist upon effective legislative or executive supervision of policing agencies to ensure we are kept safe. Safety, after all, is foundational. It is something we all care about, deeply. And it should stand to reason that those who govern in our name would want to get things right. Otherwise, they run the risk of losing their offices.

All told, then, we would expect that those who govern us would keep a close eye enough on the police to assure the polity is kept safe.

3. Stasis Around Policing

And indeed, we are seeing some activity on the part of elected officials to supervise the police, particularly around policing technology, but also in response to 2020’s protests. The reason for the attention to surveillance is that it affects a wider range of people than street policing tactics such as use of force. (We’ll get to recent street policing reform efforts in a moment.) Several states have begun to regulate the use of emerging technologies like drones.^[145] California requires police departments to put their policies for license plate readers and stingrays on the web.^[146] A small number of municipalities have adopted legislative schemes to supervise, or require authorization of, policing surveillance systems.^[147]

Still, legislative and executive supervision of the police under policing technologies really is quite meager, far less than what the public choice story based on the distribution of costs and benefits might suggest. Places like San Francisco, Seattle, and Oakland, which regulate policing surveillance systems, are in a distinct minority, and it is assuredly not because those are the only places where the police are using panvasive surveillance. In most states, invasive technologies like drones, license plate readers, and predictive policing algorithms remain unregulated altogether.^[148] Any sort of legislative regulation of policing is patchwork and episodic at best.^[149] Even getting Congress to amend the woefully out-of-date Electronic Communications Privacy Act (ECPA) has proven a Herculean challenge.^[150]

Besides—and this is a critical point that often gets lost in policing debates—even when there is some energy to ensure police do not violate rights (such as reform efforts in the wake of the police brutality protests of 2020), there is almost no regulation of policing to ensure it is efficacious. Note the distinction. We see the occasional measure to limit what the police can do, to minimize the social costs. But it’s hard to identify any legislative or executive effort of any sort designed to ensure that the police actually
are doing a good job of keeping us safe. Imagine any other area of government running this way. From the most mundane (plumbing) to the most critical (nuclear power plants), we learn, regulate, set standards, and monitor compliance.^[151]

What we traditionally have seen around policing, in contrast to many other areas of governance, is a huge pile of legislative and executive inaction. Stasis is the order of the day. Which means, in effect, that policing agencies mostly make their own decisions as to what policing looks like on the ground, unfettered by supervision or direction. They decide on what equipment to purchase, on where it is deployed, and on the tactics to be used.^[152] If we are not as safe as we should be, and are bearing unnecessary costs, it is stasis and deference to the police that explains why.

And yes, there have been very recent reform efforts, something we will address in short order. Those efforts occurred just as this Article goes to print. Still, there already are signs they are losing steam. Because stasis and deference are the norm.

B. Policing’s Information Failure

But why this stasis and deference? The oddity should be apparent. We all want to be safe. We all are bound to suffer policing’s failures in some way. And increasingly we feel the burden of policing’s costs. So why do those whose job it is to govern the police largely sit on their hands?

1. The Stuntz Explanation

Bill Stuntz has an argument, one that he advanced repeatedly before his untimely death: that legislators don’t legislate because the Supreme Court took their incentive away by constitutionalizing policing.^[153] By constitutionalizing criminal procedure, judges yanked the regulatory carpet out from under these legislative bodies.^[154] Not only have courts beaten legislatures to the regulatory punch, but once they have, additional regulation has little appeal.^[155] Any further regulatory measures would raise the costs of policing (i.e., let more criminals go free) with little additional benefit in terms of reining in misconduct. Voila: the result is largely legislative inaction around policing.^[156]

As brilliant as Stuntz’s work generally is, this argument is deeply problematic in a way that sheds light on just how perverse stasis around policing can be. Stuntz focuses, as do many who write in this space, on a tiny swath of what real policing is about. The assumption that the police are even marginally regulated by court decisions misses most of what is important about policing.^[157]

Take use of force, much in the news today. Supreme Court decisions in this area, such as Tennessee v. Garner^[158] (prohibiting the shooting of fleeing suspects absent identifiable risk) or Graham v. Connor^[159] (requiring that force be “objectively reasonable”), barely cover the field. The utter failure of Garner and Graham have led the public of late to focus on tangible measures like body cameras and de-escalation training, or banning one form of force, such as chokeholds.^[160] But as Maria Ponomarenko argues in a recent paper, even these measures—which, by the way have received some legislative attention despite Stuntz’s marginal costs argument—do not begin to scratch the surface of what could be regulated.^[161] Use of force is affected by, among other things: recruitment, training, supervision by sergeants, force reporting, force investigations, and discipline. Every bit of this could be dealt with legislatively or administratively if some officials deigned to pay attention. But they largely don’t.

Policing surveillance really underscores the error of Stuntz’s argument, raising questions of why legislative bodies remain so silent. The bulk of police surveillance is almost entirely unregulated by constitutional law.^[162] The use of most police surveillance tools doesn’t even constitute a “search” within the meaning of the Fourth Amendment.^[163] Informants and spies, pen registers, third-party searches, CCTV, and license plate readers simply are beyond the bounds of constitutional law at present.^[164] So there, at least, even under Stuntz’s thesis, we’d expect, and would have expected for decades as this surveillance mechanism built up, some sort of effective government regulation. Instead what we see is just a few states or cities doing any regulating on these issues at all.^[165]

And note well once again: all these examples of regulation are about controlling the excesses of the police. Nothing in constitutional law has one iota to say about assuring policing itself is effective, yet here too legislative bodies and executive officials do little. They don’t mandate trials or demonstration projects to assure policing tactics are effective; they don’t set goals or standards for broad community safety. They just appear to assume the police have got things perfectly under control.

2. A Stunning Lack of Information

The actual problem here is not that legislators feel displaced by courts, but that far too much of policing lives in a dark hole of ignorance. There is insufficient information on the efficacy of many policing practices, and a shameful lack of knowledge—or even curiosity—about the magnitude of the costs of those practices, especially social costs. As a result, it is flat out difficult (or even ill-advised) for officials to take action. How does one know what to do, or even recognize that action is required, without adequate information? (This is not the whole problem, to be sure: as we’ll explain momentarily, legislative stultification also results from too much trust in, and fear of, the police. But even trust and fear are to some degree the product of ignorance.)

When information about police tactics is readily available, it often becomes salient, spurring legislative and executive action. There is no better example than the wave of proposed reforms catalyzed by the tragic death of George Floyd and massive protests against police brutality. In a matter of weeks, at least sixteen state legislatures introduced, amended, or passed 159 bills and resolutions related to policing,^[166] and the federal government released three plans that would make police reform mandatory nationwide.^[167] Similarly, after the public watched the troubling militarization of the police play out in Ferguson, some states started to rein in the police.^[168] California, for instance, passed a variety of data-collection reforms that increase accountability of law enforcement, including a bill mandating that all California police departments collect data on serious incidents of use of force.^[169] At least five states—Colorado, Connecticut, Illinois, Utah, and Washington—enacted legislation that either limits some types of use of force or requires increased police training on the definition of justifiable force.^[170] And a host of states explicitly created legislation that expanded when it is legal for an individual to record the police.^[171]

Public salience also spurred action around racial profiling. North Carolina, for instance, passed a law requiring the collection of traffic stop data in response to allegations in the state general assembly that police were stopping drivers because of their race.^[172] Just a few months after Connecticut police officers were arrested for targeting Latinos in their traffic stops, the state legislature passed a bill that created reporting requirements for police conducting stops.^[173] In response to similar concerns, Missouri adopted a law requiring data collection of stops.^[174] Information and salience motivate police reform, but these too often are in short supply.

3. Causes of Our Lack of Information

Our lack of information is the result of an unhealthy dynamic between the police and political leadership. Legislative and executive officials either trust blindly in the police or fear their political clout too much to supervise policing. The police, for their part, treasure secrecy, and often seem to resist sharing and learning from information and research. The result is ignorance.

a. Blind Trust in Police.—Officials could require the necessary information to allow regulation from the police, but instead they almost always simply defer, in what is surely a telling example of blind faith.^[175] This may be most apparent in the judicial system, where police commonly are critical witnesses in criminal trials.^[176] It’s hardly a secret that police often lie on the stand, so much so that it has been given its own name: testilying.^[177] A New York Times study found that police lying on the stand is an “entrenched perjury problem several decades in the making.”^[178] And yet, as Anna Lvovsky details, over time judges have come to “recognize police work as a matter of professional expertise.”^[179] They seem to rely on what the police say, even when knowing it is unreliable.

Similar examples abound of elected officials simply trusting in what the police claim, without requiring backup information, even when all signs point in the opposite direction. The Portland Police Department incorrectly tweeted that milkshakes being handed out by a left-wing group during an alt-right protest contained cement.^[180] The city’s mayor, Ted Wheeler, defended the police’s decision to share unconfirmed information with the public, though members of the left-wing group received death threats as a result of the tweet.^[181] The Elkhart, Indiana mayor defended the police department’s pursuit policy even after two police pursuits ended in the death of both suspects, without doing much to investigate whether that policy made sense.^[182] Famously, former Chicago mayor Rahm Emanuel supported the police department’s decision not to turn over video footage in the shooting of Laquan McDonald in 2014, because of concerns over possible public riots.^[183] A survey of over seventy mayors across the country revealed that while those mayors were disturbed by recent examples of excessive police use of force, they still were more worried about officer safety than the safety of people of color in police encounters and felt that their police officers were well-trained to handle use-of-force situations.^[184]

b. Fear of the Police Lobby.—It is entirely possible that the reason why government actors defer to the police is not because they trust them but because they fear them. They are reluctant to rouse the police lobby, further stultifying any chance at getting the information we need.^[185]

Legislative bodies and executive officials who might try to pry information from the police face a formidable lobby. The lobby against regulating law enforcement includes police unions, as well as their allies in prosecutors’ offices.^[186] They are a potent lobbying force, both in terms of their membership numbers and the campaign contributions they provide.^[187] Law enforcement officials care even more about matters that affect them directly than they do about sentencing, and they are not shy about letting their voices be heard.^[188] We can see the influence of the police in laws adopted in many states providing not regulation of the police, but a Law Enforcement Bill of Rights.^[189] Congress has been urged by a vast array of interest and industry groups to update the Electronic Communications Privacy Act, but DOJ and SEC enforcement officials have stalled needed reforms.^[190]

The impact of police lobbying is frequently evident. When Prince George’s County narcotics officers and the Sheriff’s Office SWAT team shot and killed Berwyn Heights Mayor Cheye Calvo’s dogs in a misplaced SWAT raid, Calvo proposed a bill in the Maryland legislature that did nothing more than require reporting on the use of SWAT.^[191] You’d have thought he’d proposed taking their service revolvers away. It passed over huge police opposition but was allowed to expire without renewal.^[192] The only state that followed Maryland’s example was Utah, where a bill directed the state’s police agencies to report how and when they used their SWAT teams.^[193] California has required reporting on the racial demographics of police stops, but it is a fight there, as elsewhere, when police are asked to do so.^[194] The federal Freedom of Information Act (FOIA) and states’ analogous laws technically cover law enforcement agencies, but legislators and the courts often exempt them from disclosing information to the public on the grounds of public or officer safety or ongoing investigations.^[195] In 2015, for example, South Carolina Governor Nikki Haley signed into law a bill requiring state and local law enforcement officers to wear body cameras, but included within that law a blanket exemption from the state’s FOIA equivalent for body-worn camera videos.^[196]

Underlying legislative reluctance to face off with the police lobby is a deep-seated fear—precisely the same as we saw with regard to legislation about criminalization and sentencing—of being called soft on crime at election time.^[197] It’s hard to understand this extreme deference other than as a function of the fact that legislators and other elected officials are scared to death to oppose police given the consequences should something go south. It’s not like there aren’t opposing arguments to much of what the police lobby is for, and there are active interest groups on the other side.^[198] But it is one thing for your opponent to call you out when facing the voters. It is quite another for the police to be doing so. Every politician’s nightmare has to be denying the police what they say they need or making the police do something they say is a threat to public safety, only to have some serious crime or disaster occur.^[199] For this reason, when the police want something, legislators frequently jump. And when they don’t want it, those charged with governing the police head for the sidelines. It was only in the face of something as traumatic as the nationwide protests in the spring and summer of 2020 that some legislatures were motivated to act.

c. Law Enforcement Culture.—Of course, we might hope the police themselves would provide information about their work. Even if the police displayed little interest in developing information about the social costs of policing, one would think that professionalism would lead them to want basic information on what works. But the police seem no more interested in acquiring or relying on information than legislative and executive bodies.

i. Undue secrecy.—There is a prevailing culture of insularity and defensiveness that traps valuable information within policing, keeping it even (or especially) from those who would govern policing.^[200] There is no doubt that some degree of secrecy is required in policing, perhaps more than in other areas of government. No one wants to endanger officers or the public by spilling operational or tactical details, such as protocols for dealing with active shooters. Yet, the amount of information that has been withheld by police in the name of secrecy is wholly out of proportion to what safety actually requires. As Rachel Harmon has noted, those who want information from the police often have to fight tooth and nail to get it.^[201] Agencies simply are loath to open themselves up to scrutiny.^[202]

Emblematic of this norm of secrecy is the “blue wall of silence,” which goes up after a police officer is accused of misconduct.^[203] We’ve been treated to stunning examples of this phenomenon, such as the cover-up of the killing of Laquan McDonald in Chicago, or suppressing the details of the death of Daniel Prude in police hands in Rochester.^[204] This kind of secrecy can be altogether insidious to regulation.

It’s understandable—albeit unacceptable—to not want to turn over information that points to wrongdoing; it’s harder to explain refusing to provide information to the public that enables performance evaluation and suggestions for improvement. When the ACLU conducted a national study on SWAT teams, more than half of the police departments contacted simply didn’t respond to its requests for information.^[205] And a study by the Center for Policing Equity was unable to gather enough data related to police shootings to make any determinations about racial disparities precisely because departments refused to release such data—even though the Center allowed the information to be submitted anonymously.^[206] The Chicago Police Department also failed to fulfill a simple FOIA request by the Chicago Reporter—though the information was to be used in a database that tracks public civil suits against police officers, the newspaper was asking for nothing other than a simple list of the names of CPD’s sworn officers.^[207]

Even when the police do report, the quality of the data often is unreliable. For years, the federal government has collected basic crime statistics to produce the Uniform Crime Report (UCR).^[208] The quality and quantity of data varies drastically by jurisdiction—police departments often do not submit an entire year’s worth of data or may only submit data for certain types of offenses.^[209] More important, police and policing agencies also regularly are found to have manipulated crime reports, in particular downgrading crimes to lower offenses and discouraging victims from filing complaints to make crime statistics appear better than they really are.^[210] As we all learned after the killing of Michael Brown in Ferguson, Missouri, we can’t even manage to gather from the police sound data on how many people they shoot and kill every year.^[211]

ii. Reluctance to learn.—To be clear, there is a great deal of research around policing. There are countless criminology departments at universities, and a number of criminology journals that publish scholarship on policing.^[212] Many of these regularly publish papers on what works and what does not around policing. And there are divisions of the U.S. Department of Justice—such as the National Institute of Justice, and the Bureau of Justice Assistance—that hand out grants and publish work to advance our knowledge of policing.^[213] (Even in this academic work, however, there is little attention paid to the social costs of policing. Too often the focus is on “what works,” rather than “at what cost?”)^[214]

But there exists a very serious gap between academic research and policing practice.^[215] Cynthia Lum, a former police officer and now a leading criminologist, explains that academics are taught research tools, not how “to disseminate research and translate it into meaningful forms” for police officials.^[216] The police feel that real policing issues are not of concern in universities and that when studies are done, the “real beneficiaries . . . are the researchers and not the police.”^[217] At the same time, policing agencies don’t take advantage of what is learned by researchers.^[218] Leading policing scholars such as Gary Cordner and Cynthia Lum have written extensively about the failure of research to affect practice.^[219]

Since the late 1990s, there has been a movement for “evidence-based policing,” but this movement has made shockingly little headway among police themselves. In theory, evidence-based policing would “use[] research to guide practice and evaluate practitioners.”^[220] Lawrence Sherman began arguing for the approach in 1998, drawing from the movement for the evidence-based practice of medicine.^[221] Since then, there are evidence-based policing (EBP) organizations in several countries, including the U.S., U.K., and Canada.^[222] But as everyone involved in the EBP movement recognizes, including—if not especially—Sherman, uptake has been slow.^[223] The annual meeting of the International Association of Chiefs of Police regularly attracts some 15,000 participants;^[224] the American Society of Evidence-Based Policing meetings have around 300.^[225] As Cynthia Lum, one of its foremost advocates, has acknowledged, “Despite its potential, . . . evidence-based policing has not rapidly diffused into American policing.”^[226] Other movement leaders, like David Weisburd and Peter Neyroud, echo that: “[T]he evidence-based model for developing practices and policies has not been widely adopted by police agencies.”^[227]

Instead, too much of policing continues to be driven by the “idiosyncrasies of the incident, anecdotes and stories, officers’ experiences, political and social crises, standard operating procedures, moral panics, political ideology, pressure-group interests, police organizational, strategic, and tactical culture, and other whims, hunches, feelings, and best guesses.”^[228] Sherman even has tried an evidence-based argument to support evidence-based policing.^[229] That this should be necessary, or that making decisions based on “evidence” would be an innovation in the twenty-first century, is telling. As Sherman convincingly argues, any trade that fails to take account of science and learning cannot really call itself a profession.^[230]

d. Failure of Administrative and Judicial Remedies.—Finally, we might at least hope to obtain information from policing after it has gone awry, but administrative and judicial remedies often fail to force information and sometimes fail even to function at all. Tight union contracts shield much of police discipline—situations, obviously, where something went wrong—from public disclosure.^[231] Civilian boards review complaints about police misconduct—again, instances where things arguably went wrong and need regulation—but reports of their inefficacy are legion, and they often similarly are hobbled from producing systemic information about policing.^[232] What goes wrong in policing could surface in the context of criminal prosecutions that follow arrests, but judges are notoriously loathe to rule against the police in exclusion motions, and most cases plead out anyway, ensuring relevant facts are not elicited publicly.^[233] The doctrine of qualified immunity shields law enforcement officers from personally being held liable for constitutional violations unless there is a violation of “clearly established law.”^[234] States are immune altogether from suits about policing.^[235] Municipalities face no liability unless there is a policy in place, rewarding a failure to have policy and further eliminating an incentive to produce information around policymaking.^[236] Most suits against municipalities settle, meaning there is no careful scrutiny of what the police have done.^[237]

* * *

Policing’s information deficit is the result of a sort of vicious circle. If elected officials had information, they would know if regulation was needed. By the same token, if they tried regulating, that might force information from interest groups and the police themselves. We don’t try, or try much, so we don’t get the sort of information that is needed. Similarly, elsewhere in government agencies—and the police are, after all, an agency—are required to engage in notice and comment rulemaking if they want to adopt practices or rules that affect us all.^[238] That too has an information forcing function. But policing manages to escape these requirements, for reasons that have been documented elsewhere.^[239]

Policing can’t improve without information, and we just don’t have it. So, the question becomes: What can we do to produce the information needed to see that policing is accomplishing its public safety mission?

III. Addressing the Information Problem

The question is how to generate, and make available, the information society requires in order to ensure public safety. Given policing’s cultural bent toward insularity, and legislative reticence, the fixes are going to be neither quick nor easy. Nonetheless, there are encouraging signs of change already happening. This Part identifies some of the key initiatives that ought to be pursued so that we have the information necessary to ensure all of society is safe.

A. Cost–Benefit Analysis

Other regulatory fields have addressed dysfunctional or suboptimal regulation by using cost–benefit analysis (CBA) to generate necessary information and incorporate it into the decision-making process. Policing should do the same. We focus initially, and particularly closely, on
CBA, because when policing has proven suboptimal, the reason often has been a failure to take account of the full range of costs and benefits, especially social costs.

1. The Value of CBA

CBA has been the cornerstone of the United States’ regulatory system since the early 1980s, when President Ronald Reagan assumed office with the explicit goal of dismantling the regulatory state.^[240] One of the measures he put into place was Executive Order 12291, which mandated CBA of all regulatory measures.^[241] Although the motivation was that under careful CBA scrutiny many regulations would prove inefficacious or not worth the cost, irony had its way. One of the first measures the Reagan Administration targeted was the Carter Administration’s phase-out of leaded gasoline.^[242] But after a high-quality CBA revealed the potential consequences of slowing down the process, President Reagan changed course and implemented a more accelerated phase-out program.^[243]

As the leaded gasoline example shows, CBA serves as a “natural corrective” to the lack of information that leads to ill-advised, reactive decisions.^[244] It provides a systematic way to collect information about
policy choices and compare options. It also increases transparency in the regulatory process, providing a way for experts to participate in the evaluation of policy options.^[245]

Despite the clear value of CBA, it barely is used in the realm of policing. As noted in the Introduction, there is a small but important body of literature on cost–benefit analysis in the criminal justice arena more generally, but very little on policing tactics, techniques, and technology.^[246] In her important work evaluating the impact of federal grant programs around policing, Rachel Harmon makes the essential point that although policing imposes significant costs, they often are undercounted or ignored entirely in setting policy.^[247] Researchers and practitioners alike fail to give CBA its due in policing space.

Yet, where CBA has been used in criminal justice, it can make a clear difference. Although no jurisdiction since has adopted a “full-scale” CBA for sentencing policies, Rachel Barkow explains that many of today’s reform measures were motivated by jurisdictions requiring consideration of the “direct costs of sentencing.”^[248] It would be ideal, Barkow observes, “to require decisionmakers to reflect on all of the costs and benefits of proposed sentencing legislation before its adoption,” but even in its lesser form, information obtained from looking at costs can have tremendous influence.^[249]

2. The Proper Decision-Maker

When considering the application of CBA to policing, an essential point is that decision-making regarding policing is not, and should not be, for the police alone. We have abdicated too much of the decision-making around policing to chief law enforcement officers (CLEOs), be it a police chief or a sheriff.^[250] Beyond the imperative for popular involvement in deciding how to police, there’s a more practical point: even using CBA, public safety decisions made only by CLEOs are likely to suffer from significant spillover effects and externalities.

The incentives that drive CLEOs distort their frame of reference. Take the example of license plate readers. We’ve already seen how there are tangible costs—such as equipment, data storage, and officer training—as well as tangible benefits in terms of the number of apprehended individuals.^[251] A CLEO is likely to make decisions based on a very narrow frame of reference, and favor acquiring ALPRs because they provide an efficient way of apprehending more individuals.^[252] But “apprehending more individuals” means more arrests, and that alone points to serious dysfunction in policing decision-making.

Although the police tend to think of arrests as “benefits”—getting someone who did something in violation of the law off the street—most social scientists think of an arrest not as a benefit, but as a cost.^[253] For years, officers have been rewarded for arresting people, and police departments have counted arrests in the benefit category. But in reality, arrests impose a number of burdens, whether it is court time, prison beds, or the outward ripple effects on the families of those who are justice-system involved.^[254] These costs exist outside the CLEOs frame of reference, which is the problem. The correct measure of a benefit around policing is not someone taken into custody: it is preventing crime and fostering an increased perception of public safety.^[255] An arrest may take a car thief off the street, but the market for stolen car parts may be such that another thief will step in to take his place.^[256]

Adopting a holistic approach to policing policy that takes full account of all the costs and benefits is not always easy, in part because the federalist and localist structure of law enforcement in the United States creates a mismatch of incentives and spillover effects.^[257] The jurisdiction responsible for law enforcement may not be the jurisdiction that bears the cost of punishment.^[258] Until there is a system that causes all decision makers in the criminal justice system to internalize the costs of their decisions, externalities and spillovers are likely to drive public safety decision-making.^[259] The broader the decision-making frame, the better.

3. Making CBA Workable for Practitioners

In order to accomplish its purposes, CBA has to be workable for practitioners, including political appointees and the police. CBA can be a bit rarefied, attracting a sophisticated set of researchers. But it need not be understood that way. At bottom, it is simply a decision tool.^[260] CBA needs to be seen in a practical light, perhaps as nothing more than an “invitation to balance a range of variables,” taking them all into account.^[261] The goal of conducting CBAs is to see what we can learn, and what difficulty we can avoid, by thinking through a problem comprehensively.^[262]

It would be a notable advance for policing decision-making for the command staff of a policing agency, individuals on a city council, or a mayor’s office, to do nothing more than sit around a table and think rigorously through the costs and benefits of an approach: making a list of pros and cons, and trying in some rough fashion to weigh them. Just doing this sort of forward thinking would have an immediate impact on policy in this space.

4. The Full Range of Costs and Benefits

What’s critical, though, is that whoever is making policing decisions does so in a way that takes account of the full range of costs and benefits. Too often in policing there is an “act now, develop a rationale later,” mentality. Neighbors complain about prostitution, so the vice squad kicks into gear.^[263] The country is attacked by foreign terrorists, and the Federal Bureau of Investigation pivots, fundamentally changing its mission, from law enforcement to intelligence gathering.^[264] A serial murderer is on the loose, and all available resources are devoted to apprehending the individual.^[265] In the face of gun violence, or persistent violent crime, departments adopt programs like stop-and-frisk.^[266]

Although moving expeditiously to address public safety challenges is explicable, especially in a crisis, it also has its downsides.^[267] Examples here are manifold. As we’ve seen, body cameras were adopted to provide transparency and accountability to policing, at a truly staggering cost, but were put into place with policies that did not provide for disclosure of footage. Stingray cell tracking devices were acquired by policing agencies under nondisclosure agreements with the FBI and the manufacturer requiring that their use not be revealed, even in court, failing to foresee the outrage this would provoke among some segments of the public, including members of the U.S. Senate and judges hearing criminal cases.^[268] Those who advocated programmatic use of stop-and-frisk seem not to have thought through the enormous social costs, including loss of community trust and a backlash in affected communities that has hindered cooperation necessary to achieve public safety.

Police officials hardly are the only ones who have forged policy this way. In the 1980s, social scientists focused on the cost of crime, and developed techniques for monetizing victimization.^[269] Taking these costs into account, increased enforcement and stringent interventions seemed warranted. These interventions led in turn to a sharp spike in incarceration.^[270] But no one seems to have considered the social costs of increased enforcement, and whether, with those costs considered, the enforcement measures—including mass incarceration—still were cost beneficial. Today’s concerns about mass incarceration were the all-too-predictable response to a set of policies that ought to have been considered better on the front end.^[271]

The trick is thinking ahead to the full range of costs. Direct costs, particularly monetary ones, like the cost of acquiring ALPRs or body cameras, tend to be acknowledged. Even yet, agencies have failed to think beyond acquisition costs to storage costs, and thus were caught off guard as these costs mounted.^[272] Some agencies have abandoned body camera programs in the face of these costs.^[273] A simple CBA up front may have saved much misdirection.

On the other hand, all too often as we have seen, social costs—the costs that a form of policing imposes on those who are policed—have been ignored. In part this may be because the costs are intangible, and difficult to value, something we will address momentarily. Nonetheless, failing to account for them can be devastating, as the example of programmatic stop-and-frisk in New York made clear.^[274] There also might be social benefits: for example, a community policing model that increases community trust, which itself has a value.^[275]

Decision makers need to account for a number of additional factors. There are opportunity costs.^[276] Could the money spent on ALPRs or bulk collection of data be put to better use? Resources are not unlimited.^[277] There are downstream costs, which can get lost with too narrow a frame of reference, as the arrest example makes clear.^[278] And there often are unintended consequences, which might have come to mind with more reflection on the front end.^[279] These can be both unintended costs—e.g., frequent misdemeanor arrests utilized as part of broken-windows policing may have made obtaining a job difficult, turning some folks to illicit activity to feed and care for families—and unintended benefits—such as ALPR data that once stored made it easier to solve some crimes.

Among all these costs and benefits, there is one in particular—distributional costs—that looms large in policing and presents somewhat of a difficulty for economic approaches. A chosen course of action may prove to be the net best for society as a whole from a cost–benefit perspective, and yet the costs may fall disproportionately on a particular group.^[280] These often are referred to as “transfers” between the winners and the losers.^[281]

Distributional effects are of particular concern in the policing and public safety context, given that it is well established that the costs of policing—from stop-and-frisk to policing algorithms—often fall disproportionately on minority populations or historically marginalized communities.^[282] Black communities in particular have incurred severe costs at the hand of policing. Even if many policing tactics were a net benefit to society (and as we have seen there is room to be skeptical, a point we return to in a moment), there is room for concern about the distribution of costs.

Classical economic theory tends to take distributive effects out of the cost–benefit formula altogether, leaving them to the political process for remediation.^[283] Economists tend to say that the first task is to make the pie as big as possible, then we can start dividing it.^[284] In theory, division occurs through some political process that is sensitive to transfers, and remediates them in some way, for example by taxing the entire population then making a transfer to the group bearing the costs.^[285]

The problem is that this two-stage process just doesn’t work. As many note, that second step—remediation—all too rarely comes to pass.^[286] To the extent this is true, distributional inequities linger.

Equity alone compels dealing with distributional costs up front, but a failure to deal with them also may have negative long-term effects on public safety as a whole.^[287] A good example of this is the long-term loss of legitimacy of the police in certain communities, particularly communities of color, as a result of heavy policing that also resulted in high levels of incarceration.^[288] That resulting loss of trust and legitimacy has translated into a refusal to cooperate with the authorities, even in reporting and addressing gun violence.^[289] If this sort of effect can be anticipated as a result of a single policy, then it should be factored directly into the CBA of that policy. But if something like the erosion of trust and legitimacy is the cumulative result of a variety of policies over time, it becomes more difficult to incorporate and must be considered on the front end in some fashion.

5. Ongoing Analysis

Cost–benefit analysis is not, and should not be, reserved for decision-making on the front end; programs should be subject to constant re-evaluation. Many of the tactics of policing that now are considered ineffective or overly costly—such as random patrol, or aggressive stop‑and‑frisk—nonetheless have been employed for years. If something is not working, or indeed if there is no evidence it is working, why continue doing it? Inertia is a bad reason.

One advantage of committing to an ongoing cost–benefit approach is that evaluation can be built into the front end, helping to tackle one of
the great challenges in this space: how to assess effectiveness. The gold standard in such assessment today is a randomized controlled trial, or RCT.^[290] In an RCT, a geographic area is randomized, and then the treatment or intervention is tried in some areas and not in the others, the “control” areas.^[291] If, after the experiment is run, there is an effect in the treatment areas but not the control areas, the conclusion is that the treatment or intervention explains what happened.^[292]

But among the difficulties with RCTs, one large one is ethical.^[293] Suppose that the police firmly believe that Intervention X will make a community notably safer, perhaps bringing down homicides. Randomization now presents us with a sort of Sophie’s Choice: in order to learn something, we are going to deny some neighborhoods the treatment, which in theory means some murders will occur that could have been prevented.^[294] It is easy to see why police chiefs are squeamish about this approach, and university Institutional Review Boards are concerned about experimental choices like these.^[295]

If one thinks ahead, however, and assumes ongoing evaluation is part of policing, it may be possible to find a solution to these challenges. One possibility is a “rollout” knowing where the intervention would happen next in a planned way.^[296] So, for example, if the gunshot detection program ShotSpotter is being rolled out slowly starting in District A, but everyone knows District B has a gunfire problem and is going to be the next place, District B can act as a sort of control. Planning ahead can facilitate evaluation—even if it is less strict than these sorts of controlled experiments.

6. Solving the Expertise Problem: Valuing Intangibles as an Example

Of course, employing an RCT or any technical approach to evaluating a project requires a certain amount of expertise, leading to a word on the challenges of using CBA. As we’ve hoped to have made clear, there’s a great deal of progress to be made simply by lay utilization of an approach to decision-making that thinks ahead about the full range of costs and benefits. Still, really wringing information from a CBA requires the use of technically trained individuals and clever approaches. The trick is to figure out how to make progress while valuing both local experience and expert capacity.

Valuing intangibles—one of the greatest challenges in this space—provides a model for how lay individuals and experts can work together to make CBA meaningful. Many of the costs of policing will be hard to value. These might include community trust, psychological impacts of encounters with the police, or the impact of perceptions of racial profiling. There are understandable concerns that some of these sorts of harms should not be monetized or commodified; on the other hand, if we don’t try valuing these things they simply will be ignored.^[297]

CBA provides ways to deal with these intangibles short of monetization, and some of them are easy enough for agencies to pursue on their own, or with local community assistance. First, it is a prominent lesson that if there are no identifiable benefits, there is no need to value costs.^[298] That is why it is incumbent on decision makers to the extent possible to have some assurance of the value of an approach before proceeding.^[299] Still, in the face of a serious threat, trying something may be better than doing nothing. Second, another way to deal with difficulties around valuation is by providing a qualitative description of costs or benefits, and developing an estimate of how often these are incurred.^[300] If you cannot put a figure on the benefit of having a SWAT team, is it at least possible to specify when the team will be deployed and for what sorts of incidents? On the costs side can you anticipate what may go wrong and build in a sense of those harms? If programmatic stop-and-frisk is to be used, can one say how many people will be stopped, how many are likely to be found not culpable of or in possession of anything? A third option is to value alternative measures to ameliorate the costs. Alternatives are an important part of cost–benefit analysis; indeed, the methodology has its greatest value in choosing among alternatives.^[301] There are a variety of possible measures, such as contact cards—so citizens can file complaints easily if they feel they have been stopped unfairly—or training in procedural justice, or tracking whether certain officers seem to be responsible for more of the racially disproportionate outcomes.^[302] Finally, there is “break even” analysis.^[303] The idea here is that if it is possible to put a firm value on some things but not others, at least one gets a sense of what is needed to break even, to justify a policy. Suppose that a full system of LPRs can be priced out, but the utility of having LPRs is more contingent. One can at least look at the cost and get a sense then of what the benefit has to look like to justify the expense. In this way, one can determine whether going down a certain road is unlikely to make sense, before the decision to proceed ever is made.^[304]

There also are sophisticated means of valuing intangibles, such as hedonic pricing and contingent valuation, for which policing agencies may benefit from expert help. In the former, economists try to use revealed preferences to value nonmarket events.^[305] Economists may look to housing prices in various neighborhoods to value crime there. Contingent valuation (CV) is a survey technique that asks respondents to say what they are willing to pay (WTP) or willing to accept (WTA) to avoid or encourage certain conduct.^[306] CV can be costly if done right, yet it may provide the best estimate of how to value important things that will be lost absent this alternative.^[307]

Though all this may seem daunting, there is no reason to think through the costs and benefits of policing approaches from the ground up in every instance. Some types of costs and benefits reappear in response to a variety of policing tactics. And each agency hardly has to do it for itself: large agencies that have the resources to do the work first often lead the way, and others follow.

In addition, the value of collaboration should not be understated. Many agencies have the capacity to engage academic partners to assist with their work. CrimeLab, RAND, the Policing Project, and many other entities are working with agencies to analyze data and answer questions about the efficacy of policing practices.^[308] So too are individual academics. These collaborations have the best chance of ensuring that practitioner and researcher thinking is married together into useful end product.

7. Funding

Using techniques like these could help us put reasonable estimates on some of the most common costs of policing, such as psychological costs or racial harms, but getting the work done can be costly, raising finally the issue of funding. It’s difficult to develop sophisticated research on the sorts of issues discussed here around public safety without adequate funding; although there are some encouraging signs, the overall picture has been disappointing.

The federal government expends roughly $100 million annually in research funding on public safety and policing, through the Office of Justice Programs, and the National Institute of Justice, among other agencies.^[309] Although this may seem like no small amount, it pales in comparison with funding for other major public safety issues. The National Cancer Institute, for example, received roughly $5.75 billion for 2019.^[310] In addition, agencies in the criminal justice space too often are buffeted by political pressures. Witness the complete turnaround during the Trump Administration in funding for both the Community Oriented Policing Services office at DOJ and NIJ, in terms of the projects they were considering. It was a whipsaw from attention to community collaboration to hard-core enforcement.^[311] There needs to be a long-term research strategy to answer the questions that need answering, without political meddling. In addition, way too much funding has been awarded without any attention to cost–benefit analysis, despite its adoption elsewhere in the federal government.

The funding on the private philanthropic side is even sadder. Funding around policing is itself remarkably limited compared to other social concerns such as education or health.^[312] And too little of that funding goes to research. Not only is public safety foundational, but there is increasing recognition that public safety is health.^[313] There has been a great deal of attention given to mass incarceration—as there should be—but it is somewhat startling how little of the funding in this area has gone to policing, the front door of any jail or prison. There are important players around issues like gun violence, such as the Joyce Foundation.^[314] Arnold Ventures announced a collective funding effort of $50 million on that issue.^[315] But all told, the funders of sound research around policing are simply too few.

B. Sentinel Event Review

CBA is excellent for program evaluation, on the front end or on
an ongoing basis, but there also has been little sustained effort to learn systemically from what goes wrong in policing. Exemplary are Part II’s barriers to post hoc remedies for failures. But there are other approaches that are nonblaming, and no matter the state of policing remedies, these should
be used.

Sentinel event review is widely employed in medicine, aviation, and other high-risk fields.^[316] The idea is to learn from error: to go back over a significant negative outcome, such as a plane crash or preventable death during surgery; assess any individual or structural mistakes that led to this outcome; and use what is learned during this process to improve the system and reduce the likelihood of that mistake happening again.^[317]

The National Institute of Justice recently launched a “Sentinel Events Initiative” that can help. Its goal is to provide resources for communities to review the criminal systems’ failings collaboratively.^[318] Most criminal justice agencies have error-detecting procedures in place that tend to be “single-cause, one-component” reviews—i.e., that focus on sniffing out the “bad apple” that caused the negative outcome.^[319] Yet, often it is unlikely that one person alone was responsible for the outcome. Perhaps more important, punishment-based inquiry incentivizes covering up errors, rather than uncovering them to foster learning.^[320] The NIJ’s Sentinel Events Initiative, by contrast, is: (1) nonblaming; (2) encouraging of all stakeholders to participate; and (3) an ongoing, embedded practice within communities.^[321]

Some progressive agencies have adopted sentinel review techniques to great profit. Notable among them is Tucson, where Chief Chris Magnus launched a Critical Incident Review Board (CIRB). That board already has conducted a number of reviews, including of a vehicle accident fatality in which a police officer was involved, and an immigration protest in which officers used force.^[322] Like the NIJ’s review process, the CIRB involves all stakeholders, attempts to learn from the past, and focuses on “broader issues of policy, training, supervision, and needed resources.”^[323]

C. Legislative Information Forcing

The approaches to obtaining information discussed so far are open
to any agencies and officials to pursue. The problem, as we have seen, is that too few take this course. Next, we turn to ways that legislative bodies, at
the state or municipal level, can force the production and dissemination
of information.

These approaches necessarily assume a willing legislative body, which—as we saw in Part II—is a problem of its own. However, as the examples we discuss in this section suggest, there are legislative bodies that want to motivate the production of information, and simply need the tools to overcome the reluctance of policing agencies in their jurisdictions. In addition, these approaches can help overcome legislative reluctance as well. That is because there are instances in which simply requiring information is more palatable than regulating the public safety agencies directly. As a halfway measure, it may be easier to adopt information-forcing statutes.

1. Forcing CBA

Just as Ronald Reagan’s executive order sparked a cost–benefit revolution, state and local legislative bodies could mandate the use of cost–benefit analysis as a tool to evaluate public-safety decision‑making. (Mayors or city managers could do the same, of course.) Washington State does something of this sort, and by all accounts it has been notably effective. Its Institute for Public Policy (WIPPS) houses an interdisciplinary team that conducts policy reviews at the behest of the state legislature, including the use of cost–benefit analysis.^[324] WIPPS has done work around criminal justice, but not a great deal around public safety and policing.^[325] If Washington State can do this so nicely, why not other states?

One of the challenges to mandating cost–benefit is figuring out when precisely to require it. Under federal law the standard is that any new regulation that is “economically significant” is required to undergo a centralized review that includes the development of an onerous Regulatory Impact Analysis, which may or may not translate well into the municipal context.^[326] A potential alternative—used in WIPPS—is for the state legislature or city council to dictate when a proposed policy must undergo cost–benefit analysis.^[327]

We are seeing some legislative movement requiring the production of information in areas most ripe for CBA, such as policing technology. The model for these laws is the ACLU’s Community Control over Police Surveillance (CCOPS) approach, which mandates impact reports on specific technologies, surveillance-use policies, and annual surveillance reports. Agencies cannot acquire or deploy surveillance technology without disclosing the mandated information and obtaining city council approval.^[328] Although the model is not framed in terms of CBA, it requires disclosure of much of the same information that one would expect from a CBA. And in any event, it is decidedly information forcing in a useful way.

Laws based loosely or more closely on the CCOPS model have been adopted in a number of cities. The first was Seattle in 2013.^[329] Other cities have followed, including Oakland, San Francisco, Cambridge, and Nashville.^[330] Two states—California and Maine—have enacted statewide CCOPS legislation, which requires CCOPS procedures at the local level.^[331]

The CCOPS model has its problems—it may be too onerous to achieve compliance, and policing agencies will find the “surveillance” label
for their work off-putting—but something like it is entirely appropriate if
not essential.

2. Forcing Specific Information

A less intensive alternative is to require public information in limited circumstances or on particular topics. One area Bill Stuntz pointed to as a success in this regard was statutes requiring the collection of demographic information around police stops. These laws can provide an antidote to racial profiling in a couple of ways. First, agencies that have to produce the information may as a result of doing so become attentive to the problem, and take action. Second, the public or the legislative body may use the information to insist on further measures. Today there are some twenty-five states that require the collection of racial profiling information for either or both traffic and pedestrian stops, and hundreds of municipalities also have adopted plans to collect racial profiling information in their jurisdictions.^[332] Such information allows activists and policymakers to quantify what had before been purely anecdotal, and to uncover any pattern or practice of racial profiling in policing.^[333]

Legislatively mandated demographic information around racial profiling has led to revised practices in some states. In Rhode Island, for instance, after two studies found racially disparate treatment for traffic stops, the state revised its police training around implicit bias and traffic stops, leading to a reduction in consent and probable-cause searches of cars, but an increase in the rate of contraband uncovered.^[334] In Connecticut, following a study that found that police treated motorists differently based on racial differences, the state changed its practices by requiring that officers give a stopped driver a card explaining how to file a complaint.^[335] After it was shown that North Carolina’s mandated traffic stop data resulted in massive racial disparities in traffic stops, several jurisdictions revised their practices on how they approach traffic stops, including requiring officers to obtain written consent before searching a car during a traffic stop.^[336]

These laws aren’t perfect. Many don’t cover both pedestrian and traffic stops. Louisiana’s law exempts agencies that have policies against racial profiling from making their data public, and there are no standards for these policies.^[337] Vermont’s law contains no enforcement mechanism, so predictably, many police departments aren’t complying.^[338]

Still, the lesson seems to be that when there is an identifiable social problem, the salience of the issue can motivate or allow legislators to force information, even if that is all they are able to do. In California, for instance, the state legislature enacted a law that makes police-misconduct records public, even though the state’s police union vehemently opposed the measure.^[339] The same happened in New York, following the surge in protests against police brutality after the killing of George Floyd.^[340] In Maryland, a new law was passed in the wake of the death of Freddie Gray, a man who died while in police custody, that requires law enforcement to report all officer-involved deaths to the state Governor’s Office of Crime Control and Prevention.^[341] And after the New York Times, using public arrest data required by law, uncovered staggeringly disproportionate enforcement of marijuana laws against people of color, a municipal law was passed that requires the NYPD to issue tickets for people who smoke weed in public, instead of arresting them.^[342]

The federal government can and should play an important role in this space. Congress has passed laws that require the Executive Branch to collect certain information, such as the Violent Crime Control and Law Enforcement Act, which mandates the Attorney General collect data on lethal use-of-force by police.^[343] This law is not as strong as it might be, though—it doesn’t provide the Attorney General with the means to fulfill this requirement, so it is no wonder that the Department of Justice has not successfully collected this data in any year in the past two decades.^[344]

Even when Congress passes laws requiring data collection, the Department of Justice simply has not used the congressional sticks it does possess to force the information that Congress mandated as a condition
on federal grants. Take the Death in Custody Reporting Act (DIRCA),
which requires states that receive funding under the Omnibus Crime
Control and Safe Streets Act track and submit information regarding the death of an individual who is detained or under arrest.^[345] Although the Attorney General has the discretion to impose as much as a 10% penalty of funds for nonreporting, the Department of Justice has never done so to encourage reporting.^[346]

D. Regulatory Intermediaries

Another way to produce information around public safety is to create administrative bodies that ensure this happens. Maria Ponomarenko refers to these bodies as “regulatory intermediaries,” and one of their prime functions is in producing the information needed to achieve sound public safety objectives.^[347] Some jurisdictions have adopted this approach with regard to public safety, creating public entities whose job it is to regulate the police. More jurisdictions should do so.

1. Police Commissions

One sort of regulatory intermediary that has the potential not only to produce information, but to act upon it, is police commissions. These are public agencies that have the power to set policy for policing agencies. Depending on the jurisdiction, they effectively adopt an administrative law approach to policing.

The Los Angeles Police Commission may be the most effective and powerful of these. It regularly proposes rules and offers opportunity for public views before adopting them.^[348] Recently it engaged a not-for-profit to help it gather public and police input into the proper policy on when to release officer body-worn camera video.^[349] That exercise resulted in a Commission rule that eventually became the law in the State of California.^[350] It also recently revised the LAPD’s use of force policy.

Some jurisdictions are experimenting with bodies that play the role of commissions, but only around policing technology and privacy. Oakland’s new privacy board is a good example. These bodies force information in a manner akin to CCOPS, and then have the power to act on it.^[351] It’s hard to say why these limited-jurisdiction bodies are being set up, rather than generalist police commissions, but one suspects it may be because there is wider public concern for privacy than for, say, racially discriminatory stops. If so, this unfortunately continues policing’s problem of paying insufficient attention to distributional costs.

2. Inspectors General

An easier political sell may be the establishment of Inspector General offices, whose very function is providing information to improve public safety. Inspectors General (IGs) typically perform audits of policing agency practices and make recommendations for policy change. IGs have been established in a number of major cities including Chicago, Los Angeles, and New York.^[352]

IGs are doing notable work to inform the public about police practices. The Chicago Inspector General’s Office released a report in 2018 highlighting the police department’s lack of clear directives on how it
picks, trains, and evaluates school police officers, resulting in many students getting pulled into the criminal justice system wrongfully.^[353] More recently, the Chicago IG’s Office disseminated a report that highlighted major issues with the data and transparency of the police department’s gang database.^[354] The New York Inspector General has produced nine reports on various NYPD practices, including the use of chokeholds, a failure to train officers to effectively de-escalate potentially violent situations, and the handling
of U‑visa requests for undocumented immigrants who were victims of
a crime.^[355]

Ideally, the information produced by IG investigations results in policy change.^[356] That was the case in Los Angeles around a set of controversial predictive policing tools. In March 2019, the Los Angeles Inspector
General issued a report critical of the Los Angeles Strategic Extraction
and Restoration (LASER) zones, a program started in 2011 to pinpoint
the exact locations connected to gun and gang violence, and to identify violence-involved individuals.^[357] The forty-eight-page report noted that the program lacked oversight, there was insufficient data to calculate the program’s success, and officers often used inconsistent criteria to label people who might commit violent crimes.^[358] The new information the report brought to light ultimately led to the LAPD’s decision to shut down the LASER program.^[359]

E. Notice and Comment Rulemaking

Another way to address information shortages is through notice-and-comment rulemaking, one of the key tools used in the administrative state to produce information and set policy. Agencies themselves can either be required to, or decide on their own to, engage in notice-and-comment rulemaking.^[360] Police departments are agencies under many state administrative codes that mandate notice-and-comment rulemaking in certain circumstances. Still, there are fewer than one might expect because these requirements typically only kick in when an agency is altering an individual’s rights and obligations under law, which policing agencies typically are not empowered to do.^[361]

However, agencies can and have engaged in forms of notice-and-comment rulemaking on their own. In formulating their body-camera policies, both the Camden County Police Department and the NYPD engaged a third party to perform the notice-and-comment function for them.^[362] The result was notable changes in the agency policies, though perhaps less than advocates might have hoped.^[363] A particularly notable example of notice and comment rulemaking is the Chicago Police Department’s initiative around its gang database. In response to the Inspector General report discussed above, the department decided to scrap its existing gang database and create a new one—the Criminal Enterprise Database.^[364] It opened up its policy for public comment for thirty days and agreed to take public comments into consideration when finalizing the policy.^[365] Again, although some have argued that the new database, which notifies listed individuals and provides an opportunity to appeal the designation, as well as the public comment process itself, did not go far enough in correcting the problem, it is an important step in bolstering police–community relations in Chicago.^[366]

To make notice-and-comment proceedings effective, Rachel Barkow suggests that criminal justice agencies be required to have an empirical basis for their rules, enforcing that requirement with judicial review.^[367] Similarly, municipalities may consider requiring that agencies produce racial impact statements to account for potential distributional costs of a policy.^[368] These information-producing mechanisms could ensure that policing decisions are on a sound footing and do not impose racial harms.

F. Aggregating Information and Identifying Best Practices: A National College of Policing

Information production is not an end unto itself. The suggestion of Part II is that the political process is locked down for lack of information. Many examples discussed in Part III have shown that, once information is available, stasis can be overcome.

Although our focus here is on information production in order to foster deliberation and change around public safety, there is one development that sweeps more broadly and deserves its own attention: the idea of a national college of policing. This is not a “college” in the sense of being an institution where people go for training—though there are proposals to establish a policing university in the United States, largely as a reaction to what some perceive as a failure in criminal justice programs to focus sufficiently on policing.^[369] Rather, this is an institution that would serve as a storehouse for information on policing research and a place in which to develop best practices based on the research we have.^[370] It could also help bridge the gap between academics and police by supporting partnerships similar to those funded by the Bureau of Justice Assistance’s Smart Policing Initiative and National Institute of Justice’s Building and Enhancing Criminal Justice Researchers-Practitioners Partnerships.^[371]

Development of a national college of policing in the United States would not need to start from scratch: In 2012, the United Kingdom created just such a body. The College of Policing serves three functions: (1) it develops the research for improving the evidence of what works and what does not work in policing, with the goal of ensuring “that policing practice and standards are based on knowledge, rather than custom and convention;” (2) it creates educational standards that police officers across the country have to meet; and (3) it uses the best evidence of what works to set standards of policing.^[372]

One of the College of Policing’s main developments is the What Works Centre for Crime Reduction.^[373] The effort is to review policing research and evaluate in a standardized way what works.^[374] Since its creation, it has set up an evaluation framework and reviewed more than 300 crime reduction studies. It also has created guidelines for police and researchers on cost analysis of interventions (what data to collect, how to calculate costs associated with implementing interventions, etc.).^[375]

Despite the impressive ambitions and lofty goals of the UK College of Policing, there is substantial room for critique. For example, one part of the website stresses the use of cost–benefit analysis, and provides tools to aid agencies in performing this analysis.^[376] Yet, nowhere in there is there any discussion of the social costs of policing, other than the costs of crime. For a body with as impressive a beginning as the College of Policing, this omission is unfathomable.

Still, the general approach of the College of Policing would fill notable gaps in the policing infrastructure in the United States. Currently, there simply is no central repository of information about sound policing.^[377] The National Institute of Justice runs a web-based clearinghouse for research on criminal justice, but it has only several hundred rated programs, most of which are not about policing.^[378] The American Society of Evidence-Based Policing has just started to put out policy briefs summarizing what they can find on policing tactics, in an effort to make information widely available to police officers and leadership.^[379] RAND has created a Better Policing Toolkit to help departments find effective public safety strategies and put them into action, but it is in the pilot stage.^[380] The Center for Evidence-Based Crime Policy (CEBCP) at George Mason University has a very intriguing matrix that aggregates and evaluates the results of studies, but this effort is still new and involves only a couple hundred studies.^[381] It is a tool with promise if only it were used. As we have seen, there is way too little diffusion from what researchers have learned to what police officials actually take on board
and utilize.^[382]

There has been some tentative movement toward an American college of policing, but it does not seem to be going anywhere, which is unfortunate. It’s not entirely clear why, though one reason may be our federalist and localist structure, and concern for any national body doing standard setting for the rest. To the extent this is true, it is myopic. Although some aspects of policing are and will remain local, the truth is that there is a pool of knowledge that either is not localized, or that could be tailored once we had some baseline of information.

Conclusion

We’ve argued here that we don’t know enough about what keeps us safe and what does not. We’ve looked at the public choice dysfunction in our political process that seems to accept that level of ignorance as the status quo. And we’ve developed detailed suggestions for directions for change, some that depend on the political process and some that do not.

Ultimately, the lack of progress or even serious interest about a national college of policing is just a metaphor for everything else wrong with knowledge around policing. For fear of what might happen, we seem unwilling to learn. To the extent we won’t learn, we can’t improve. To the extent we can’t improve, our public safety will not be as secure as it could be. It’s high time to change all of that.

1. .Barry Friedman, We Spend $100 Billion on Policing. We Have No Idea What Works., Wash. Post (Mar. 10, 2017, 1:49 PM), https://‌www.washingtonpost.com‌/posteverything‌/wp‌/
   2017‌/03‌/10‌/we-spend-100-billion-on-policing-we-have-no-idea-what-works‌/?noredirect=on&utm‌_term=.16a6b6587439 [https://‌perma.cc‌/32FU-CYAQ]; Jeanne Sahadi, The Cost of Fighting Terrorism, CNN (Nov. 16, 2015, 6:29 PM), https://‌money.cnn.com‌/2015‌/11‌/16‌/news‌/economy‌/cost-of-fighting-terrorism‌/index.html [https://‌perma.cc‌/G5ER-HHXH] (noting how much we spend on counterterrorism efforts); David Weisburd & Peter Neyroud, Police Science: Toward a New Paradigm, New Persps. Policing (Harvard Kennedy Sch., Cambridge, MA and Nat’l Inst. of Justice, Washington, D.C.), Jan. 2011, at 1, 3 (“[M]ost police practices are not systematically evaluated, and we still know too little about what works and under what conditions in policing.”). ↑
2. .See, e.g., Cynthia Lum, Translating Police Research into Practice, 11 Ideas Am. Policing (Police Found., Washington, D.C.), Aug. 2009, at 1, 3 (explaining that many police responses are based on guesses and idiosyncrasies of the moment). ↑
3. .See infra subpart I(B) (detailing how little we know about what policing strategies and technologies are actually effective); see also Lawrence W. Sherman, The Rise of Evidence-Based Policing: Targeting, Testing, and Tracking, in 42 Crime and Justice in America, 1975–2025, at 377, 384 (Michael Tonry ed., 2013) (“Most police practices, despite their enormous cost, are
   still untested.”). ↑
4. .See Daniel S. Nagin, Cost-Benefit Analysis of Crime Prevention Policies, 14 Criminology & Pub. Pol’y 583, 585 (2015) (“Criminal justice policies commonly have impacts that go far beyond prevention of crime and impacts on victims and perpetrators.”); see also Dan A. Black, Robert M. Solow & Lowell J. Taylor, Comments on Domínguez and Raphael, 14 Criminology & Pub. Pol’y 639, 641 (2015) (detailing the social costs imposed on those who are suspected of having committed a crime). ↑
5. .Randall Kennedy, Race, Crime, and the Law 151, 159 (1998) (arguing that racial profiling in policing imposes a “racial tax” on innocent racial minorities); Rachel A. Harmon, Federal Programs and the Real Costs of Policing, 90 N.Y.U. L. Rev. 870, 941 (2015) (explaining how only a small percentage of the public—disproportionately members of racial minorities—is subject to policing). ↑
6. .Brentin Mock, How Police Are Using Stop-and-Frisk Four Years After a Seminal Court Ruling, CityLab (Aug. 18, 2017, 6:30 AM), https://‌www.citylab.com‌/equity‌/2017‌/08‌/stop-and-frisk-four-years-after-ruled-unconstitutional‌/537264‌/ [https://‌perma.cc‌/6VM5-ALRG] (explaining that between January 2004 and June 2012, the NYPD made 4.4 million stops). ↑
7. .See Nat’l Acads. Scis., Eng’g & Med., Proactive Policing: Effects on Crime and Communities 150–51 (David Weisburd & Malay K. Majmundar eds., 2018) (describing two studies that found mixed outcomes on the benefits of pedestrian stops). ↑
8. .Ashley Southall, Decline in Stop-and-Frisk Tactic Drives Drop in Police Actions in New York, Study Says, N.Y. Times (Dec. 11, 2015), https://‌www.nytimes.com‌/2015‌/12‌/12‌/nyregion‌/end-to-stop-and-frisk-drives-drop-in-police-actions-in-new-york-study-says.html [https://‌perma.cc‌/32BU-GXVD] (noting that NYPD officials questioned the efficacy of stop-and-frisk tactics after a reduction in stops did not lead to increased crime). ↑
9. .Brian Brackeen, Facial Recognition Software Is Not Ready for Use By Law Enforcement, TechCrunch (June 25, 2018, 6:30 AM), https://‌techcrunch.com‌/2018‌/06‌/25‌/facial-recognition-software-is-not-ready-for-use-by-law-enforcement‌/ [https://‌perma.cc‌/75CG-S95Z] (explaining that the technology is not neutrally applied and does not perform equally among races). ↑
10. .See, e.g., Amazon’s Facial-Recognition Technology Is Supercharging Washington County Police, Oregonian (Apr. 30, 2019), https://‌www.oregonlive.com‌/washingtoncounty‌/2019‌/05‌/amazons-facial-recognition-technology-is-supercharging-washington-county-police.html [https://‌perma.cc‌/A86U-3MTY] (discussing how the Washington County Sherriff’s Office in Oregon implemented Rekognition, Amazon’s facial-recognition technology). ↑
11. .These are a sampling of questions suggested to one of the authors in a series of telephone conferences with Police Foundation Fellows (on file with authors). ↑
12. .Whether we act on the information, such as in the case of a warming planet, is another question altogether. ↑
13. .See infra section II(B)(2). ↑
14. .See Rachel A. Harmon, The Problem of Policing, 110 Mich. L. Rev. 761, 792–94 (2012) (discussing the harms of policing). ↑
15. .See infra note 246 and accompanying text. ↑
16. .A seminal piece by Darryl Brown focuses on prosecution, or the wages of mass incarceration. Darryl K. Brown, Cost-Benefit Analysis in Criminal Law, 92 Calif. L. Rev. 323 (2004). Rachel Barkow also recently wrote a book on using cost–benefit analysis in sentencing. Rachel Elise Barkow, Prisoner of Politics: Breaking the Cycle of Mass Incarceration (2019). For CBA studies discussing diversion programs, see NPC Res. & Ctr. for Ct. Innovation, Testing the Cost Savings of Judicial Diversion: Final Report (2013), https://‌www.courtinnovation.org‌/sites‌/default‌/files‌/documents‌/NY‌_Judicial%20Diversion‌_Cost%20Study.pdf [https://‌perma.cc‌/U9A2-7LVH]; Steve Aos, Polly Phips, Robert Barnowski & Roxanne Lieb, The Comparative Costs and Benefits of Programs to Reduce Crime, Wash. St. Inst. Pub. Pol’y (2001), https://‌www.wsipp.wa.gov‌/ReportFile‌/756‌/Wsipp‌_The-Comparative-Costs-and-Benefits-of-Programs-to-Reduce-Crime-v-4-0‌_Full-Report.pdf [https://‌perma.cc‌/6N7G-9XYM] and Alexander J. Cowell, Nahama Broner & Randolph Dupont, The Cost-Effectiveness of Criminal Justice Diversion Programs for People with Serious Mental Illness Co-Occurring with Substance Abuse: Four Case Studies, 20 J. Contemp. Crim. Just. 292 (2004). There is a literature that evaluates the “cost of crime,” which was used initially to justify the sort of harsher criminal justice measures that led to mass incarceration. E.g., David A. Anderson, The Aggregate Burden of Crime, 42 J. L. & Econ. 611, 614 (1999); Mark A. Cohen, The Monetary Value of Saving a High Risk Youth, 14 J. Quantitative Criminology 5, 11 (1998); Mark A. Cohen, Pain, Suffering, and Jury Awards: A Study of the Cost of Crime to Victims, 22 L. & Soc’y Rev. 537, 537–38 (1988). A symposium devoted to the technical aspects of the cost-of-crime literature drew some attention to these “equity” costs. Symposium, Role of the Cost-of-Crime Literature, 14 Criminology & Pub. Pol’y 583 (2015). The most significant entrant in the policing space itself is Rachel Harmon’s excellent work on the costs imposed by federal “justice” programs, particularly federal grant programs such as the provision of military equipment to local police forces, which distort local preferences around policing. See Harmon, supra note 5, at 894 (explaining how racial minorities are disproportionately policed). Finally, there is a short symposium in the Journal of Benefit-Cost Analysis, organized by one of the authors here (and participated in by both) that begins to bring attention to the critical methodological issues around the evaluation of policing practices. Symposium on Benefit-Cost Analysis of Policing, 8 J. Benefit-Cost Analysis 305 (2017). ↑
17. .In pursuing this question, we adopt a narrow view of public safety, one that focuses on crime and violence, rather than a more capacious conception that ensures people have the basic necessities to live a secure life. See, e.g., Invest-Divest, Movement for Black Lives, https://‌m4bl.org‌/policy-platforms‌/invest-divest‌/ [https://‌perma.cc‌/2PQF-MMDD]. There has been too little attention in scholarship to what public safety means. See Jeremy Waldron, Safety and Security, 85 Neb. L. Rev. 454, 456 (2006) (“Although we know that ‘security’ is a vague and ambiguous concept . . . still there has been little or no attempt in the literature of legal and political theory to bring any sort of clarity to the concept.”). In other work, one of us is exploring a broader notion of public safety. Barry Friedman, Disaggregating the Policing Function, 169 U. Pa. L. Rev. (forthcoming 2020–2021); Barry Friedman, What Is Public Safety? (June 12, 2019) (unpublished manuscript) (on file with authors). ↑
18. .Paul Michna, Patrol Officers Are Backbone of Any Police Department, Hous. Chron. (Apr. 22, 2009, 7:00 PM), https://‌www.chron.com‌/neighborhood‌/magnolia‌/opinion‌/article‌/Patrol-officers-are-backbone-of-any-police-9337519.php [https://‌perma.cc‌/8ER9-VWMJ] (noting that patrol is the “backbone of any police department” and where officers “begin [their] careers”). ↑
19. .Gary Fields & John R. Emshwiller, Putting Police Officers Back on the Beat, Wall St. J. (Mar. 12, 2015, 7:54 PM), https://‌www.wsj.com‌/articles‌/putting-police-officers-back-on-the-beat-1426201176 [https://‌perma.cc‌/Y4RA-SA9W] (explaining that in the 1800s, U.S. cities began implementing a “beat system” in their police departments). ↑
20. .Id. ↑
21. .Id. ↑
22. .See Penelope McMillan, 911: At 1 Year, Emergency System Still Maturing, L.A. Times (Sept. 30, 1985, 12:00 AM), https://‌www.latimes.com‌/archives‌/la-xpm-1985-09-30-me-17784-story.html [https://‌perma.cc‌/E2XP-WXM8] (describing the implementation of 911 programs in multiple cities in California). ↑
23. .George L. Kelling, Tony Pate, Duane Dieckman & Charles E. Brown, The Kansas City Preventive Patrol Experiment: A Summary Report 1 (1974). ↑
24. .Id. ↑
25. .Id. ↑
26. .Patrick V. Murphy, Foreword to George L. Kelling, Tony Pate, Duane Dieckman, & Charles E. Brown, The Kansas City Preventive Patrol Experiment: A Summary Report v (1974). ↑
27. .See, e.g., George L. Kelling & James Q. Wilson, Broken Windows: The Police and Neighborhood Safety, Atlantic (Mar. 1982), https://‌www.theatlantic.com‌/magazine‌/archive‌/1982‌/03‌/broken-windows‌/304465‌/ [https://‌perma.cc‌/F8HD-7HF6] (introducing broken-windows policing). ↑
28. .Joseph D. McNamara, Preface to George L. Kelling, Tony Pate, Duane Dieckman, & Charles E. Brown, The Kansas City Preventive Patrol Experiment: A Summary Report vii (1974). ↑
29. .See id. (“The experiment did show that routine preventive patrol in marked police cars has little value in preventing crime or making citizens feel safe.”). ↑
30. .Id. ↑
31. .Sherman, supra note 3, at 387 (“[T]he conclusion that random patrol did not work was widely accepted . . . None of that stopped the widespread use of random patrol.”). To be fair, some of the patrol may be “directed,” which is to say that it is not random but aimed at areas where officers believe there are issues with crime. But short of data-driven hot spot policing, which we come to momentarily, even directed patrol is at best impressionistic. ↑
32. .Id. at 378. ↑
33. .See, e.g., Theresa Ervin Conover & John Liederbach, Policing on Demand: An Observational Study of Mobilization and Citizen Encounters Across Communities, 17 Policing: Int’l J. Police Sci. & Mgmt. 170, 171 (2015) (using data from seventeen agencies to provide a breakdown of police activities, which primarily includes patrol); James Frank, Steven G. Brandl & R. Cory Watkins, The Content of Community Policing: A Comparison of the Daily Activities of Community and “Beat” Officers, 20 Policing: Int’l J. Police Strat. & Mgmt. 716, 720–22 (1997) (discussing the sizable role of community policing officers, whose primary task is to patrol); John Liederbach, Addressing the “Elephant in the Living Room”: An Observational Study of the Work of Suburban Police, 28 Policing: An Int’l J. Police Strat. & Mgmt. 415, 417 (2005) (presenting data concerning the work routines and citizen interactions of suburban police officers, which primarily included patrol); Gordon P. Whitaker, What Is Patrol Work?, Police Studs., Winter 1982, at 13 (1982) (finding that police primarily spend time on patrol, not fighting crime). ↑
34. .Peter Moskos, Cop in the Hood: My Year Policing Baltimore’s Eastern District 108, 109 (2008). ↑
35. .See generally Lawrence W. Sherman & David Weisburd, General Deterrent Effects of Police Patrol in Crime “Hot Spots”: A Randomized, Controlled Trial, 12 Just. Q. 625 (1995) (describing this experiment and its conclusions). ↑
36. .Id. at 643. ↑
37. .See Kelling & Wilson, supra note 27 (proposing that broken-windows policing in a community leads to serious crime). ↑
38. .See Anthony A. Braga, Andrew V. Papachristos & David M. Hureau, The Effects of Hot Spots Policing on Crime: An Updated Systematic Review and Meta-Analysis, 31 Just. Q. 633,
    634–35 (2014) (conducting a meta-analysis of nineteen hot spot studies generally finding crime reduction and no displacement). The drops in crime can be low in some instances, however, and it appears continued dosage is required to make those drops persist. Id. at 644, 646–47, 652–53. ↑
39. .Lum, supra note 2, at 3 (noting that while hot spot policing is an evidence-based practice, there is “little real indication that hot-spot policing is institutionalized in daily police work”); Sherman, supra note 3, at 403 (“Random patrol predominates in practice if not in rhetoric; hot spot patrol is rising rapidly, at least in rhetoric if not yet in practice.”). ↑
40. .See, e.g., Robert J. Sampson & Stephen W. Raudenbush, Disorder in Urban Neighborhoods—Does It Lead to Crime?, Nat’l Inst. of Justice 1, 4 (2001) (questioning the relationship between disorder and crime); Bernard E. Harcourt & Jens Ludwig, Broken Windows: New Evidence from New York City and a Five-City Social Experiment, 73 U. Chi. L. Rev. 271, 277 (2006) (same). ↑
41. .See William J. Bratton, N.Y. Police Dep’t, Broken Windows and Quality-of-Life Policing in New York City 2 (2015) (“[S]erious criminals committed petty crimes, too.”). ↑
42. .Nat’l Res. Council Nat’l Acad., Fairness and Effectiveness in Policing: The Evidence 229 (Wesley Skogan & Kathleen Frydl eds., 2004) (arguing that there is not strong research to support broken-windows policing). ↑
43. .Cory P. Haberman, A View Inside the “Black Box” of Hot Spots Policing from a Sample of Police Commanders, 19 Police Q. 488, 501–02 (2016) (explaining that one reason police chiefs might employ hot spot policing is due to a deterrent effect); George L. Kelling & William J. Bratton, Why We Need Broken Windows Policing, City J. (2015), https://‌www.city-journal.org‌/html‌/why-we-need-broken-windows-policing-13696.html [https://‌perma.cc‌/Q54K-R3SB] (discussing the deterrent effect of quality-of-life policing). ↑
44. .Nat’l Acads. Scis., Eng’g & Med., supra note 7, 150–51 (finding that evidence about the effects of stop-and-frisk are inconclusive). Some studies showed that programmatic stop-and-frisk reduced crime, while others contested that conclusion vigorously. Compare Alex Chohlas-Wood, Sharad Goel, Amy Shoemaker & Ravi Shroff, An Analysis of the Metropolitan Nashville Police Department’s Traffic Stop Practices, Stan. Computational Pol’y Lab 8 (2018) (finding that traffic stops had no impact on crime reduction), and Ames Grawert & James Cullen, Brennan Ctr. for Justice, Fact Sheet: Stop and Frisk’s Effect on Crime in New York City (2016), https://‌www.brennancenter.org‌/analysis‌/fact-sheet-stop-and-frisks-effect-crime-new-york-city [https://‌perma.cc‌/HQ2L-HBV7] (concluding that there is no relationship between stop and frisk and crime rates), with David Weisburd, Alese Wooditch, Sarit Weisburd & Sue-Ming Yang, Do Stop, Question, and Frisk Practices Deter Crime?: Evidence at Microunits of Space and Time, 15 Criminology & Pub. Pol’y 31, 46 (2016) (finding support that stop-and-frisk deters criminals). See generally Tracey L. Meares, The Law and Social Science of Stop and Frisk, 10 Ann. Rev. L. & Soc. Sci. 335 (2014) (discussing the evidence on the effectiveness of stop-and-frisk tactics). ↑
45. .See Abigail A. Sewell, Kevin A. Jefferson & Hedwig Lee, Living Under Surveillance: Gender, Psychological Distress, and Stop-Question-and-Frisk Policing in New York City, 159 Soc. Sci. & Med. 1, 6–7, 9 (2016) (finding increased stress among men who live in neighborhoods with a higher density of police frisking); Wesley G. Skogan, Stop-and-Frisk and Trust in Police in Chicago, 13 (Nw. Inst. for Pol’y Res., Working Paper No. 16-08, 2016), https://‌www.ipr.northwestern.edu‌/documents‌/working-papers‌/2016‌/WP-16-08.pdf [https://‌perma.cc‌/PA3D-BN2F] (finding that only about 30% of African-Americans who had been stopped were trusting of police, whereas about 55% of African-Americans who hadn’t been stopped were trusting of police). ↑
46. .See, e.g., Amanda Geller, Jeffrey Fagan, Tom Tyler & Bruce G. Link, Aggressive Policing and the Mental Health of Young Urban Men, 104 Am. J. Pub. Health 2321, 2321 (2014) (finding that participants who had been stopped and frisked exhibited increased trauma and anxiety); Joscha Legewie & Jeffrey Fagan, Aggressive Policing and the Educational Performance of Minority Youth, 84 Am. Soc. Rev. 220 (2019) (calculating that African-American boys who had been more exposed to police frisks had reduced test scores); Woo Chang Kang & Christopher T. Dawes, The Electoral Effect of Stop-and-Frisk 1 (July 11, 2017) (unpublished manuscript) (on file with authors) (finding that stop-and-frisk activity reduces voter turnout). ↑
47. .See Nat’l Acads. Scis. Eng’g & Med, supra note 7, at 126–27 (noting that hot spot policing often includes more frequent arrests for misdemeanors). ↑
48. .See, e.g., Rebecca Vallas & Sharon Dietrich, Ctr. for Am. Prog., One Strike and You’re Out: How We Can Eliminate Barriers to Economic Security and Mobility for People with Criminal Records 9 (2014) (finding that even one arrest “decreases employment opportunities more than any other employment-related stigma”); K. Babe Howell, Broken Lives from Broken Windows: The Hidden Costs of Aggressive Order-Maintenance Policing, 33 N.Y.U. Rev. L. & Soc. Change 271, 296 (2009) (discussing the lost work time and fees associated with even a minor violation in New York City); Eisha Jain, Arrests as Regulation, 67 Stan. L. Rev. 809, 833–34, 839–40, 844 (2015) (citing specific costs of arrests related to public housing, employment, and education). ↑
49. .See, e.g., Howell, supra note 48, at 274 (noting that quality-of-life policing creates “significant hidden” costs); Alexandra Natapoff, The Cost of ‘Quality of Life’ Policing: Thousands of Young Black Men Coerced to Plead Guilty to Crimes They Didn’t Commit, Wash. Post (Nov. 11, 2015, 11:00 AM), https://‌www.washingtonpost.com‌/news‌/the-watch‌/wp‌/2015‌/11‌/11‌/the-cost-of-quality-of-life-policing-thousands-of-young-black-men-coerced-to-plead-guilty-to-crimes-they-didnt-commit‌/ [https://‌perma.cc‌/ER3Z-WPGH] (noting how quality-of-life policing has led to baseless convictions of millions of defendants nationally). ↑
50. .Sarah Pitcher McDonough, Colorblindness, Discretion, and Systemic Inequities in Criminal Justice, 28 Geo. J. Legal Ethics 733, 742 (2015). ↑
51. .Sarah Ryley, Laura Bult & Dareh Gregorian, EXCLUSIVE: Daily News Analysis Finds Racial Disparities in Summonses for Minor Violations in ‘Broken Windows’ Policing, Daily News (Aug. 4, 2014, 2:00 AM), http://‌www.nydailynews.com‌/new-york‌/summons-broken-windows-racial-disparity-garner-article-1.1890567#4Mo4O4gmHoPITrMH.97 [https://‌perma.cc‌/CZ4H-QSVK]. ↑
52. .Id. ↑
53. .See, e.g., Office of the New York City Comptroller Scott M. Stringer, Addressing the Harms of Prohibition: What NYC Can Do to Support an Equitable Cannabis Industry (2018), https://‌comptroller.nyc.gov‌/reports‌/addressing-the-harms-of-prohibition-what-nyc-can-do-to-support-an-equitable-cannabis-industry‌/ [https://‌perma.cc‌/PM7F-MVNQ] (finding that marijuana stops and arrests by the NYPD in East Harlem, a predominately Latino community, were thirteen times higher than on the Upper East Side, an adjoining, predominately white, neighborhood). ↑
54. .Kennedy, supra note 5, at 19 (“The principal injury suffered by African-Americans in relation to criminal matters is not overenforcement but underenforcement of the laws.”). ↑
55. .Alexandra Natapoff, Underenforcement, 75 Fordham L. Rev. 1715, 1723 (2006). ↑
56. .Anthony A. Braga, Better Policing Can Improve Legitimacy and Reduce Mass Incarceration, 129 Harv. L. Rev. F. 233, 238 (2015). ↑
57. .See generally Anthony A. Braga & Brenda J. Bond, Policing Crime and Disorder Hot Spots: A Randomized Controlled Trial, 46 Criminology 577 (2008) (evaluating the effects of police disorder, within a problem-oriented policing framework, at crime and disorder hot spots in Lowell, Massachusetts). ↑
58. .Id. at 598–99. ↑
59. .Braga et al., supra note 38, at 656. ↑
60. .Id. at 659. ↑
61. .See infra note 246 and accompanying text. ↑
62. .Tom Jackman, Trump to Restore Program Sending Surplus Military Weapons, Equipment to Police, Wash. Post. (Aug. 27, 2017, 8:47 PM), https://‌www.washingtonpost.com‌/news‌/true-crime‌/wp‌/2017‌/08‌/27‌/trump-restores-program-sending-surplus-military-weapons-equipment-to-police‌/ [https://‌perma.cc‌/4ZPS-WQF8] (noting that $5.4 billion had been sent to local and state law enforcement). ↑
63. .Dexter Filkins, “Do Not Resist” and the Crisis of Police Militarization, New Yorker (May 13, 2016), https://‌www.newyorker.com‌/news‌/news-desk‌/do-not-resist-and-the-crisis-of-police-militarization [https://‌perma.cc‌/D7L2-KD9P] (attributing these acquisitions to the 1033 and Homeland Security programs). ↑
64. .Vincenzo Bove & Evelina Gavrilova, Police Officer on the Frontline or a Solider? The Effect of Police Militarization on Crime, Am. Econ. J.: Econ. Pol’y, Aug. 2017, at 2–3, 14 (finding that police militarization, measured by value or volume of military equipment, reduces crime rates and citizen complaints); Matthew C. Harris, Jinseong Park, Donald J. Bruce & Matthew N. Murray, Peacekeeping Force: Effects of Providing Tactical Equipment to Local Law Enforcement, Am. Econ. J.: Econ. Pol’y, Aug. 2017, at 311 (same). ↑
65. .Anna Gunderson, Elisha Cohen, Kaylyn Jackson, Tom S. Clark, Adam Glynn & Michael Leo Owens, Does Military Aid to Police Decrease Crime? Counterevidence from the Federal 1033 Program and Local Police Jurisdictions in the United States 26–27 (Feb. 5, 2019) (unpublished manuscript) (on file with authors). ↑
66. .See, e.g., War Comes Home: The Excessive Militarization of American Policing, ACLU (June 2014), https://‌www.aclu.org‌/report‌/war-comes-home-excessive-militarization-american-police [https://‌perma.cc‌/87PX-UZLV] (detailing Arizona law enforcement’s acquisitions, including 32 bomb suits, 64 armored vehicles, and 702 units of night-vision equipment). ↑
67. .See As We Remember the Militarized Response to the Ferguson Uprising, Trump Says Civilian Police Are Making ‘Good Use’ of Military Weapons, ACLU (Aug. 8, 2017, 5:15 PM), https://‌www.aclu.org‌/blog‌/racial-justice‌/race-and-criminal-justice‌/we-remember-militarized-response-ferguson-uprising [https://‌perma.cc‌/F67W-Z444] (“The militarized response to the uprising in Ferguson shocked America.”). ↑
68. .Harmon, supra note 5, at 873 (“Policymakers and economists do make more formal assessments of federal public safety programs. . . . But they consider only the budgetary outlay to the federal agency as costs. They do not contemplate any harm—and, therefore, any cost—from policing itself.”). ↑
69. .See ACLU, supra note 66, at 18. ↑
70. .Kevin Sack, Door-Busting Drug Raids Leave a Trail of Blood, N.Y. Times (Mar. 18, 2017), https://‌www.nytimes.com‌/interactive‌/2017‌/03‌/18‌/us‌/forced-entry-warrant-drug-raid.html [https://‌perma.cc‌/J4GB-QSGE]. ↑
71. .Bonnie Kristian, The Troubling Rise of SWAT Teams, Week (Jan. 19, 2015), https://‌theweek.com‌/articles‌/531458‌/troubling-rise-swat-teams [https://‌perma.cc‌/F96K-6H9D]. ↑
72. .Sack, supra note 70 (explaining that about 40% of warrants served in Utah were no-knock warrants and usually for drugs). ↑
73. .See id. (explaining that there is no federal mandate that police agencies report SWAT operations, and that only two states have required reporting). ↑
74. .ACLU, supra note 66, at 34. ↑
75. .Radley Balko, Shedding Light on the Use of SWAT Teams, Wash. Post (Feb. 17, 2014, 10:30 AM), https://‌www.washingtonpost.com‌/news‌/the-watch‌/wp‌/2014‌/02‌/17‌/shedding-light-on-the-use-of-swat-teams‌/?utm‌_term=.b94e898828fe [https://‌perma.cc‌/9TUX-U5F5]. ↑
76. .See, e.g., Joy Diaz, How Much It Costs Austin Every Time SWAT Teams Roll Out, KUT 90.5 (Aug. 4, 2014), https://‌www.kut.org‌/post‌/how-much-it-costs-austin-every-time-swat-teams-roll-out [https://‌perma.cc‌/955J-PHQM] (showing that a “small [SWAT] operation” in Austin, Texas costs $1,040 per hour, at minimum). ↑
77. .Sack, supra note 70. ↑
78. .See Richard A. Oppel Jr., Derrick Bryson Taylor & Nicholas Bogel-Borroughs, What We Know About Breonna Taylor’s Case and Death, N.Y. Times (Oct. 2, 2020), https://‌www.nytimes.com‌/article‌/breonna-taylor-police.html [https://‌perma.cc‌/3ARK-6TV7] (describing the execution of the no-knock warrant at Taylor’s home). ↑
79. .Sack, supra note 70. ↑
80. .A recent study used a nationwide panel measuring the presence of active SWAT teams in roughly 9,000 U.S. law enforcement agencies to test whether militarized policing lowers crime rates and enhances officer safety. Using within-agency comparisons across time, the study concluded that there is no evidence that deploying a SWAT team does either. Jonathan Mummolo, Militarization Fails to Enhance Police Safety or Reduce Crime but May Harm Police Reputation, 115 Proc. Nat’l Acad. Sci. 9181, 9181 (2018). ↑
81. .Street-Level Surveillance: Automated License Plate Readers, Elect. Frontier Found., https://‌www.eff.org‌/pages‌/automated-license-plate-readers-alpr [https://‌perma.cc‌/3ZJH-WD2K] [hereinafter “Street-Level Surveillance”]. ↑
82. .Id. ↑
83. .Id. ↑
84. .David J. Roberts & Meghann Casanova, U.S. Dep’t of Justice, Automated License Plate Recognition (ALPR) Use by Law Enforcement: Policy and Operational Guide 6 (2012). ↑
85. .Al Baker, Camera Scans of Car Plates Are Reshaping Police Inquiries, N.Y. Times (Apr. 11, 2011), https://‌www.nytimes.com‌/2011‌/04‌/12‌/nyregion‌/12plates.html [https://‌perma.cc‌/9A8Z-S46Z] (reporting NYPD’s claim that ALPRs were “directly responsible for the recovery of 3,659 stolen cars”). ↑
86. .Cynthia Lum, Linda Merola, Julie Willis & Breanne Cave, License Plate Recognition Technology (LPR): Impact Evaluation and Community Assessment 56–57 (2010) (finding no deterrent effect when comparing crime rates before, during, and after ALPR deployment); Bruce Taylor, Christopher Koper & Daniel Woods, Combating Auto Theft in Arizona: A Randomized Experiment with License Plate Recognition Technology 49 (2011) (finding increases in plates scanned, hits, arrests, and recoveries for stolen vehicles, but no effect on the overall theft rate). ↑
87. .Dave Maass & Jeremy Gillula, What You Can Learn from Oakland’s Raw ALPR Data, Elec. Frontier Found. (Jan. 21, 2015), https://‌www.eff.org‌/deeplinks‌/2015‌/01‌/what-we-learned-oakland-raw-alpr-data [https://‌perma.cc‌/69FC-3Y5R]. ↑
88. .ACLU, You Are Being Tracked: How License Plate Readers Are Being Used to Record Americans’ Movements 6 (2013), https://‌www.aclu.org‌/files‌/assets‌/071613-aclu-alprreport-opt-v05.pdf [https://‌perma.cc‌/B456-L5RG] [hereinafter “You Are Being Tracked”]. ↑
89. .See, e.g., Austen Erblat, Wellington’s License Plate Reader Contract Over $100,000 More than First Estimated, S. Fla. Sun Sentinel (Aug. 17, 2018), https://‌www.sun-sentinel.com‌/
    local‌/palm-beach‌/wellington‌/fl-cn-wellington-license-plate-reader-update-20180822-story.html [https://‌perma.cc‌/H7KL-P4XM] (detailing the high cost of purchasing and installing ALPRs); Rebecca Waters, Automated License Plate Recognition: An Insider’s Look, Police & Sec. News (Sept. 15, 2017), https://‌policeandsecuritynews.com‌/2017‌/09‌/15‌/automated-license-plate-recognition-insiders-look‌/ [https://‌perma.cc‌/AN69-JKER] (quoting Nate Maloney, ELSAG VP of Marketing and Communications). ↑
90. .See, e.g., Denis Cuff, Dublin Eyes More License Plate Readers to Fight Crime, East Bay Times (Nov. 27, 2017, 6:28 AM), at A8, https://‌www.eastbaytimes.com‌/2017‌/11‌/27‌/dublin-eyes-more-license-plate-readers-to-fight-crime‌/ [https://‌perma.cc‌/7RSN-WAFH] (detailing Dublin’s 2016 purchase, which included six ALPR systems on police cars, six ALPR systems on fixed poles, and two ALPR systems on portable trailers). ↑
91. .You Are Being Tracked, supra note 88, at 13 (finding that less than 0.2% of plate scans in Maryland were linked to criminal activity). ↑
92. .Tanvi Misra, Who’s Tracking Your License Plate?, Bloomberg CityLab (Dec. 6, 2018, 8:31 AM), https://‌www.citylab.com‌/equity‌/2018‌/12‌/automated-license-plate-readers-privacy-data-security-police‌/576904‌/ [https://‌perma.cc‌/72XF-QA6W]. ↑
93. .Street-Level Surveillance, supra note 81. ↑
94. .Id. ↑
95. .Maass & Gillula, supra note 87 (“The data indicates lower-income neighborhoods are disproportionately captured by ALPR patrols, with police vehicles creating a grid of license plates in the city’s poorest neighborhoods.”). ↑
96. .The Electronic Frontier Foundation has also brought attention to “warrant redemption” programs established between police departments and Vigilant Solutions, which incentivizes police officers to use ALPR systems to track down and pull over drivers with outstanding court fees. Dave Maass, “No Cost” License Plate Readers Are Turning Texas Police into Mobile Debt Collectors and Data Miners, Elec. Frontier Found. (Jan. 26, 2016), https://‌www.eff.org‌/deeplinks‌/2016‌/01‌/no-cost-license-plate-readers-are-turning-texas-police-mobile-debt-collectors-and#footnote1‌_5pdk92e [https://‌perma.cc‌/NSJ4-CSV6]. ↑
97. .Dianne Feinstein, Sen. Dianne Feinstein: Continue NSA Call-Records Program, USA Today (Oct. 20, 2013, 6:22 PM), https://‌www.usatoday.com‌/story‌/opinio‌/2013‌/10‌/20‌/nsa-call-records-program-sen-dianne-feinstein-editorials-debates‌/3112715‌/ [https://‌perma.cc‌/6G5A-RBNY]. ↑
98. .Keith Alexander, the former NSA Director, claimed that the data collection helped prevent over 50 potential terrorist attacks in more than twenty countries. Disclosure of National Security Agency Surveillance Programs: Hearing before the H. Perm. Select Comm. on Intelligence, 113th Cong. 10 (2013) (statement of Gen. Keith Alexander, Dir., Nat’l. Sec. Agency). Senator Dianne Feinstein also endorsed bulk data collection in an op-ed, stating, “[s]ince its inception, this program has played a role in stopping roughly a dozen terror plots and identifying terrorism supporters in the U.S.” Feinstein, supra note 97. ↑
99. .Peter Bergen, David Sterman, Emily Schneider & Bailey Cahill, New America Found., Do NSA’s Bulk Surveillance Programs Stop Terrorists? 1–2 (2014) https://‌na-production.s3.amazonaws.com‌/documents‌/Do-NSAs-Bulk-Surveillance-Programs-Stop-Terrorists.pdf [https://‌perma.cc‌/AB3X-C4LV]. ↑
100. .George Gao, What Americans Think About NSA Surveillance, National Security and Privacy, Pew Res. Ctr. (May 29, 2015), https://‌www.pewresearch.org‌/fact-tank‌/2015‌/05‌/29‌/what-americans-think-about-nsa-surveillance-national-security-and-privacy‌/ [https://‌perma.cc‌/JRY6-4S32]. ↑
101. .A.W. Geiger, How Americans Have Viewed Government Surveillance and Privacy Since Snowden Leaks, Pew Res. Ctr. (June 4, 2018), https://‌www.pewresearch.org‌/fact-tank‌/2018‌/06‌/04‌/how-americans-have-viewed-government-surveillance-and-privacy-since-snowden-leaks [https://‌perma.cc‌/K6KW-G5G4]. ↑
102. .Danielle Kehl with Kevin Bankston, Robyn Greene & Robert Morgus, Surveillance Costs: The NSA’s Impact on the Economy, Internet Freedom & Cybersecurity, New America’s Open Tech. Inst. (July 2014). ↑
103. .Charlie Savage, Disputed N.S.A. Phone Program Is Shut Down, Aide Says, N.Y. Times (Mar. 4, 2019), https://‌www.nytimes.com‌/2019‌/03‌/04‌/us‌/politics‌/nsa-phone-records-program-shut-down.html [https://‌perma.cc‌/H5PC-LP7E]. ↑
104. .See United States v. Moalin, 2020 WL 5225704, *9, 12 (9th Cir. Sept. 2, 2020). ↑
105. .See Press Release, U.S. Dep’t of Justice, Justice Department Awards over $23 Million in Funding for Body Worn Camera Pilot Program to Support Law Enforcement Agencies in 32 States (Sept. 21, 2015), https://‌www.justice.gov‌/opa‌/pr‌/justice-department-awards-over-23-million-funding-body-worn-camera-pilot-program-support-law [https://‌perma.cc‌/AED7-8M2A]. ↑
106. .See generally Seth W. Stoughton, Police Body-Worn Cameras, 96 N.C. L. Rev. 1363, 1378–99 (2018) (discussing the potential benefits of BWCs). ↑
107. .Press Release, U.S. Dep’t of Justice, supra note 105. ↑
108. .Police Exec. Research Forum, Costs and Benefits of Body-Worn Camera Deployments 16, 26 (2018), https://‌www.policeforum.org‌/assets‌/BWCCostBenefit.pdf [https://‌perma.cc‌/9KBV-G8D4]. ↑
109. .Rick Callahan, Some Police Departments Shelve Body Cameras, Cite Data Costs, AP News (Sept. 10, 2016), https://‌apnews.com‌/f19d7270535f4202b92de1872278330f [https://‌perma.cc‌/5FL7-RDXJ]. ↑
110. .David Yokum, Anita Ravishankar & Alexander Coppock, A Randomized Control Trial Evaluating the Effects of Police Body-Worn Cameras, 116 Proc. Nat’l Acad. Sci. 10329, 10331 (2019). ↑
111. .Barak Ariel, Alex Sutherland, Darren Henstock, Josh Young, Paul Drover, Jayne Sykes, Simon Megicks & Ryan Henderson, 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, 459 (2016). ↑
112. .Charles M. Katz, David E. Choate, Justin R. Ready & Lidia Nuño, Evaluating the Impact of Officer Body Worn Cameras in the Phoenix Police Department 31 (2015) (finding that BWC-equipped officers increased their daily arrests by over 42%); Anthony Braga, James R. Coldren Jr., William Sousa, Denise Rodriguez & Omer Alper, The Benefits of Body-Worn Cameras: New Findings from a Randomized Controlled Trial at the Las Vegas Metropolitan Police Department 11 (2017) (“The findings of this study suggests that BWCs . . . seem to increase police productivity. . . .”). ↑
113. .See Developments in the Law—Policing (ch. 4), 128 Harv. L. Rev. 1794, 1803 (2015) (citing a recent survey in which 96% of prosecutors said that video evidence improved their ability to prosecute cases). ↑
114. .See generally Clare Garvie, Alvaro Bedoya & Jonathan Frankle, The Perpetual Line-Up: Unregulated Police Face Recognition in America, Geo. Law Ctr. on Privacy & Tech. (Oct. 18, 2016), https://‌www.perpetuallineup.org‌/ [https://‌perma.cc‌/AC7G-54K2]. ↑
115. .See, e.g., Chad Marlow & Jay Stanley, We’re Updating Our Police Body Camera Recommendations for Even Better Accountability and Civil Liberties Protections, ACLU (Jan. 25, 2017, 10:15 AM), https://‌www.aclu.org‌/blog‌/privacy-technology‌/surveillance-technologies‌/were-updating-our-police-body-camera [https://‌perma.cc‌/5LGZ-PLDW] (introducing updated version of ACLU model legislation for body-worn cameras); Body-Worn Cameras, Elec. Frontier Found., https://‌www.eff.org‌/pages‌/body-worn-cameras [https://‌perma.cc‌/7L48-3WA8] (“If combined with facial recognition or other technologies, thousands of police officers wearing body-worn cameras could record the words, deeds, and locations of much of the population at a given time, raising serious First and Fourth Amendment concerns.”). ↑
116. .See, e.g., Don Babwin, More Body-Cam Footage Is Sought After Chicago’s Police Shooting of Harith Augustus, Pittsburgh Post-Gazette (July 16, 2018, 6:20 PM), https://‌www.post-gazette.com‌/news‌/nation‌/2018‌/07‌/16‌/More-body-cam-footage-is-sought-after-Chicago-s-police-shooting-of-Harith-Augustus‌/stories‌/201807160160 [https://‌perma.cc‌/53C4-X6SS] (describing a community activist’s request that the Chicago police release body-camera video from every officer on the scene of a fatal shooting); Eric Flack, DC Mayor Rejects Calls to Release Body Cam Video from Police Clashes, WUSA9 (July 31, 2018, 9:25 AM), https://‌www.wusa9.com‌/article‌/news‌/local‌/dc‌/dc-mayor-rejects-calls-to-release-body-cam-video-from-police-clashes‌/65-578753689 [https://‌perma.cc‌/MA9L-EHAD] (reporting on the District of Columbia mayor’s refusal to publicly release body-camera video of two confrontations between local police and neighborhood residents); Tribune News Services, Charlotte Police Refuse to Release Shooting Video as 3rd Night of Protests Stay Largely Peaceful, Chi. Trib. (Sept. 23, 2016, 7:28 AM), https://‌www.chicagotribune.com‌/nation-world‌/ct-charlotte-protests-20160922-story.html [https://‌perma.cc‌/T2WA-42JA] (reporting on protestors’ demands that the Charlotte police release footage of a police shooting). ↑
117. .See Police Body Worn Cameras: A Policy Scorecard, Leadership Conf. on Civ. & Hum. Rights & Upturn (Nov. 2017), https://‌www.bwcscorecard.org‌/ [https://‌perma.cc‌/NF33-45RP] (reviewing the BWC policies of twenty-five police departments and evaluating, as one criterion, the extent to which each policy allows individuals filing complaints to view footage). ↑
118. .See, e.g., Minn. Stat. § 13.825 (2019) (“Data collected by a portable recording system [worn by a peace officer] are private data on individuals or nonpublic data,” subject to several exceptions); N.H. Rev. Stat. § 91-A:5 (2017) (exempting, except in certain circumstances, “[v]ideo and audio recordings made by a law enforcement officer using a body-worn camera” from chapter providing for public access to government records); N.D. Cent. Code. § 44-04-18.7 (2015) (“An image taken by a law enforcement officer or a firefighter with a body camera or similar device and which is taken in a private place is an exempt record.”); 2016 N.C. Sess. Laws 88 (“An act to provide that recordings made by law enforcement agencies are not public records . . . .”). ↑
119. .See, e.g., Rachel E. Barkow, Federalism and the Politics of Sentencing, 105 Colum. L. Rev. 1276, 1278–83 (2005) (describing the tough-on-crime politics in sentencing); 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. 1079, 1086–92 (1993) (detailing the relationship between voters, legislators, and criminal procedure decisions); William J. Stuntz, The Pathological Politics of Criminal Law, 100 Mich. L. Rev. 505, 576–79 (2001) (describing the political economy of defining crimes). ↑
120. .See Barkow, supra note 119, at 1278–81 (discussing the inspiration for and implementation of such policies). ↑
121. .See id. at 1280 (detailing the media’s skewed presentation of crime). ↑
122. .Rachel E. Barkow, Administering Crime, 52 UCLA L. Rev. 715, 748–50 (2005) (explaining how voters tend to think of the most salient examples of crime, which largely come from inaccurate media sources). ↑
123. .Barkow, supra note 119, at 1280. ↑
124. .Barkow, supra note 122, at 728–30. ↑
125. .See Barry Friedman & Maria Ponomarenko, Democratic Policing, 90 N.Y.U. L. Rev. 1827, 1864 (2015) (“There are few labels in American politics more damning than ‘soft on crime.’”). ↑
126. .See, e.g., Stuntz, supra note 119, at 530–34 (explaining how legislators and prosecutors aim to please voters through simple and harsh laws, such as the “three strikes” law). ↑
127. .See Barkow, supra note 122, at 726–27 (describing how current sentencing policies disproportionately affect certain minority groups); Dripps, supra note 119, at 1089–90 (noting that the targets of the criminal justice system are primarily young men of color). ↑
128. .See, e.g., Shaila Dewan & Carl Hulse, Republicans and Democrats Cannot Agree on Absolutely Anything. Except This., N.Y. Times (Nov. 14, 2018), https://‌www.nytimes.com‌/2018‌/11‌/14‌/us‌/prison-reform-bill-republicans-democrats.html [https://‌perma.cc‌/A4JT-CWNB] (arguing that both liberals and conservatives agree that there is a need to repair the nation’s criminal justice system); German Lopez, After Increases in 2015 and 2016, America’s Violent Crime Rates Seems to Be Back on the Decline, Vox (Feb. 28, 2019, 12:10 PM), https://‌www.vox.com‌/policy-and-politics‌/2019‌/2‌/28‌/18244633‌/america-violent-crime-murder-rate-2018 [https://‌perma.cc‌/U9A9-CM8A] (explaining how although politicians emphasize crime rates to justify “tough-on-crime” policies, crime rates in America have been on a decline since the 1990s). This new narrative is particularly powerful in an environment in which crime rates are at an all-time low. ↑
129. .In 2014, California voters passed Prop. 47, which reclassified many drug and property crimes from felonies to misdemeanors. Magnus Lofstrom, Mia Bird & Brandon Martin, Pub. Policy Inst. of Cal., California’s Historic Corrections Reforms 7 (2016). In 2017, the California state legislature passed a package of bills that reduced prison and jail sentences for both juveniles and adults. Marisa Lagos, Jerry Brown Signs Criminal Justice Reforms, Eases Prison Terms, KQED (Oct. 11, 2017), https://‌www.kqed.org‌/news‌/11622620‌/jerry-brown-signs-criminal-justice-reforms-eases-prison-terms [https://‌perma.cc‌/T2VP-U96M]. The Louisiana state legislature also passed a package of laws in 2017, including changes to the state’s reentry practices, increased opportunities for early release from prison, and reductions in sentence lengths for certain crimes. See 2017 La. Acts 611, 612–13 (authorizing courts to waive or create payment plans for offenders’ financial obligations in an effort to prevent financial barriers to successful rehabilitation and reentry into society); 2017 La. Acts 692, 707, 709–10 (expanding parole eligibility for non-violent offenders); 2017 La. Acts 725, 726–27 (eliminating or reducing mandatory minimum sentences for certain crimes such as arson and home invasion). ↑
130. .See, e.g., Edith Brady-Lunny, J.B. Pritzker Wants Criminal Justice Reform. Here’s How It Could Work., Herald & Rev. (Dec. 26, 2018), https://‌herald-review.com‌/news‌/state-and-regional‌/crime-and-courts‌/j-b-pritzker-wants-criminal-justice-reform-here-show‌/article‌_e5b2e0bf-8cc5-5ee2-977a-2a72d9ffd578.html [https://‌perma.cc‌/LGG6-3W5L] (reporting efforts to create an office of criminal justice reform in Illinois); Tim Mak, Over 1 Million Florida Felons Win Right to Vote with Amendment 4, NPR (Nov. 7, 2018, 2:46 AM), https://‌www.npr.org‌/2018‌/11‌/07‌/665031366‌/over-a-million-florida-ex-felons-win-right-to-vote-with-amendment-4 [https://‌perma.cc‌/VF39-4XX3] (reporting passage of a ballot initiative in Florida to restore the voting rights of some citizens convicted of felonies). ↑
131. .See, e.g., Holly Otterbein, “Completely Unelectable” Progressive Larry Krasner Wins DA’s Race, Phila. Mag. (Nov. 7, 2017, 9:25 PM), https://‌www.phillymag.com‌/news‌/2017‌/11‌/07‌/larry-krasner-wins-district-attorney-general-election‌/ [https://‌perma.cc‌/A3BA-5JQC] (noting that the winner of Philadelphia’s 2017 district attorney election “promised never to seek the death penalty or bring cases based on illegal searches”). ↑
132. .German Lopez, Congress Just Passed the Most Significant Criminal Justice Reform Bill in Decades, Vox (Dec. 20, 2018, 1:55 PM), https://‌www.vox.com‌/policy-and-politics‌/2018‌/12‌/20‌/18148482‌/first-step-act-criminal-justice-reform-house-congress [https://‌perma.cc‌/BJX8-9VQG]. ↑
133. .See id. (mentioning that President Trump advocated for “tough-on-crime” policies during the 2016 campaign); see also Maggie Koerth & Amelia Thomson-DeVeaux, Many Americans Are Convinced Crime Is Rising in the U.S. They’re Wrong., FiveThirtyEight (Aug. 3, 2020, 6:00 AM), https://‌fivethirtyeight.com‌/features‌/many-americans-are-convinced-crime-is-rising-in-the-u-s-theyre-wrong‌/ [https://‌perma.cc‌/3YVH-WZMN] (explaining that the majority of Trump’s presidential campaign advertisements aired in July 2020 dealt in some way with the decline in public safety). ↑
134. .See German Lopez, American Policing Is Broken. Here’s How to Fix It., Vox (Sept. 1, 2017, 10:53 AM), https://‌www.vox.com‌/policy-and-politics‌/2016‌/11‌/29‌/12989428‌/police-shooting-race-crime [https://‌perma.cc‌/A3WX-MNKP] (noting that while Black people made up about 13% of the population, they made up more than half of reported murder victims in 2015). ↑
135. .In response to rising crime rates in Baltimore, for instance, the John Hopkins University president asked the state legislature to create an independent police force comprised of over 100 officers for the university, even though there was already a Baltimore Police Department unit devoted to patrolling the area. Luke Broadwater, Hopkins President Pitches Maryland Lawmakers on Need for Police Force; Students, Community Advocates Push Back, Balt. Sun (Feb. 22,
     2019, 7:15 PM), https://‌www.baltimoresun.com‌/politics‌/bs-md-hopkins-police-hearing-20190222-story.html [https://‌perma.cc‌/RS7W-52GW]. ↑
136. .William J. Stuntz, The Political Constitution of Criminal Justice, 119 Harv. L. Rev. 781, 795 (2006). ↑
137. .Id. at 816 (“[P]olice are more likely to inflict physical and dignitary harms on poor suspects than on rich ones.”). ↑
138. .Editorial Board, Racial Discrimination in Stop-and-Frisk, N.Y. Times (Aug. 12,
     2013), http://‌www.nytimes.com‌/2013‌/08‌/13‌/opinion‌/racial-discrimination-in-stop-and-frisk.html
     ?mcubz=1 [https://‌perma.cc‌/C6GW-6ZSA] (finding that about 83% of those stopped-and-frisked in New York City were Black or Hispanic, and over 90% of those stopped were innocent). ↑
139. .Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness 59, 96 (2010) (discussing how the criminal justice system, by targeting men of color, functions as a system of racial control). ↑
140. .Emily Ekins, Policing in America: Understanding Public Attitudes Toward the Police. Results from a National Survey, Cato Inst. (Dec. 7, 2016), https://‌www.cato.org‌/survey-reports‌/policing-america [https://‌perma.cc‌/2CLL-WCC6] (noting that 38% of Americans report being stopped by the police at least once between 2011 and 2016). ↑
141. .Christopher Slobogin, Panvasive Surveillance, Political Process Theory, and the Nondelegation Doctrine, 102 Geo. L.J. 1721, 1723 (2014). ↑
142. .See supra subpart I(A). ↑
143. .See generally Slobogin, supra note 141 (discussing the types of panvasive search‑
     and‑seizure tools available to policing institutions). ↑
144. .See Barkow, supra note 119, at 1280 (noting that tough-on-crime politics, which emerged as a response to rising crime rates and social unrest, continue to persist through the will of the people). ↑
145. .See, e.g., Allie Bohm, Status of 2014 Domestic Drone Legislation in the States, ACLU (Apr. 22, 2014, 10:30 AM), https://‌www.aclu.org‌/blog‌/technology-and-liberty‌/status-2014-domestic-drone-legislation-states [https://‌perma.cc‌/4K2V-8REB] (tracking the status of drone legislation in states as of 2014). ↑
146. .Cal. Civ. Code § 1798.90 (West 2016). ↑
147. .See, e.g., Robyn Greene, How Cities Are Reining in Out-of-Control Policing Tech, Slate (May 14, 2018, 1:58 PM), https://‌slate.com‌/technology‌/2018‌/05‌/oakland-california-and-other-cities-are-reining-in-out-of-control-police-technologies.html [https://‌perma.cc‌/K9G6-36V8] (detailing new local city laws overseeing police technology). ↑
148. .See Friedman & Ponomarenko, supra note 125, at 1844–45 (detailing the lack of regulation in policing). ↑
149. .Id. at 1844. ↑
150. .See, e.g., Kaveh Waddell & National Journal, Agencies Say They Need Access to Americans’ Emails Without a Warrant, Atlantic (Sept. 16, 2015), https://‌www.theatlantic.com‌/politics‌/archive‌/2015‌/09‌/agencies-say-they-need-access-to-americans-emails-without-a-warrant‌/458193‌/ [https://‌perma.cc‌/77UY-8QTP] (noting that while a bipartisan proposal to reform the ECPA had the support of technology companies and the Executive Branch, federal law enforcement officials opposed it, saying it would hamper civil prosecutions). The Cloud Act did amend the ECPA, but its primary purpose was to moot a litigation battle. See Lawrence Hurley, U.S. Top Court Rules that Microsoft Email Privacy Dispute Is Moot, Reuters (Apr. 17, 2018, 9:20 AM), https://‌www.reuters.com‌/article‌/us-usa-court-microsoft‌/u-s-top-court-rules-that-microsoft-email-privacy-dispute-is-moot-idUSKBN1HO23S [https://‌perma.cc‌/334M-4JCQ] (referencing an unsigned opinion by the Supreme Court on this issue). ↑
151. .See, e.g., Friedman & Ponomarenko, supra note 125, at 1843–44 (detailing the extensive code devoted to the Department of Citrus in comparison to the “skeletal” law overseeing searches without warrants). ↑
152. .Samuel Walker, Governing the American Police: Wrestling with the Problems of Democracy, 2016 U. Chi. Legal F. 615, 619 (“Each of the U.S. local [police] agencies has full control over most of its policies, procedures, and choice of law enforcement strategies . . . .”). ↑
153. .Stuntz, supra note 136, at 795. ↑
154. .Id. ↑
155. .Id. at 793. ↑
156. .Id. ↑
157. .Friedman & Ponomarenko, supra note 125, at 1832 (“[F]ew believe it makes sense for courts to be the primary supervisors of police agencies, particularly because judicial review is almost exclusively about constitutionality. Governing policing involves a host of prior questions . . . .” (footnote omitted)). ↑
158. .471 U.S. 1, 11 (1985). ↑
159. .490 U.S. 386, 397 (1989). ↑
160. .See, e.g., 2015 Cal. Legis. Serv. Ch. 469 (S.B. 29) (West) (requiring police to undergo de-escalation training specific to mental illness and intellectual disabilities); Presidential Meeting with Community Leaders and Law Enforcement, C-SPAN (Dec. 1, 2014), https://‌www.c-span.org‌/video‌/?322995-2‌/president-obama-meeting-community-leaders-ferguson-missouri [https://‌perma.cc‌/Z3AU-9TNG] (announcing the Obama administration’s call for federal aid to purchase 50,000 additional police body cameras and funding of additional police training). ↑
161. .Maria Ponomarenko, Rethinking Police Rulemaking, 114 Nw. U. L. Rev. 1, 6, 12–13 (2019). ↑
162. .See Berger v. New York, 388 U.S. 41, 63 (1967) (suggesting the use of eavesdropping devices must be subject to Fourth Amendment standards). Wiretapping is regulated, but that occurred in the face of Supreme Court insistence that absent regulation it was off limits. See id. (discussing that techniques and practices of eavesdropping may be developed “without attending illegality”). ↑
163. .See Slobogin, supra note 141, at 1726 (asserting courts hold most types of panvasive searches as not justiciable because the searches do not implicate the Fourth Amendment). ↑
164. .See Smith v. Maryland, 442 U.S. 735, 745–46 (1979) (holding that petitioner’s Fourth Amendment rights were not violated by warrantless use of a pen register); Lopez v. United States, 373 U.S. 427, 440 (1963) (holding that recording by an informant did not violate defendant’s Fourth Amendment rights); Slobogin, supra note 141, at 1753 (noting that CCTV surveillance is subject to spotty regulation at best); You Are Being Tracked, supra note 88, at 25–26 (discussing the widespread use of ALPRs and the government’s use of grants to encourage use of the technology). ↑
165. .See supra section II(A)(3). ↑
166. .Weihua Li & Humera Lodhi, The States Taking on Police Reform After the Death of George Floyd, FiveThirtyEight (June 18, 2020, 3:00 PM), https://‌fivethirtyeight.com‌/features‌/which-states-are-taking-on-police-reform-after-george-floyd‌/ [https://‌perma.cc‌/W47B-SG5P]. ↑
167. .Factbox: What’s the Difference Between Three U.S. Plans for Police Reform,
     Reuters (June 17, 2020, 4:56 PM), https://‌de.reuters.com‌/article‌/us-minneapolis-
     police-congress-legislati‌/factbox-whats-the-difference-between-three-u-s-plans-for-police-reform-idUSKBN23O3KW [https://‌perma.cc‌/K789-B7ZC] (detailing the differences between the Republican, Democratic, and executive plans for police reform). ↑
168. .See, e.g., Samuel Walker, Why Police Reform Isn’t Dead: The Case for Optimism, Crime Rep. (Jan. 23, 2019), https://‌thecrimereport.org‌/2019‌/01‌/23‌/why-policing-reform-isnt-dead-the-case-for-optimism‌/ [https://‌perma.cc‌/CKD8-8RZM] (“[S]ince Ferguson there has been an outburst of police reform ordinances and laws across the country.”). ↑
169. .Press Release, Attorney General Kamala D. Harris, Attorney General Kamala D. Harris Announces New Law Enforcement Reporting Requirements for Officer-Involved Shootings and Use of Force (Dec. 28, 2015), https://‌oag.ca.gov‌/news‌/press-releases‌/attorney-general-kamala-d-harris-announces-new-law-enforcement-reporting [https://‌perma.cc‌/D9YQ-SL4U]. ↑
170. .Ram Subramanian & Leah Skrzypiec, Vera Inst. of Justice, To Protect
     and Serve: New Trends in State-Level Policing Reform, 2015-2016 13
     (2017), https://‌storage.googleapis.com‌/vera-web-assets‌/downloads‌/Publications‌/protect-and-serve-policing-trends-2015-2016‌/legacy‌_downloads‌/041417-PolicingTrendsReport-web.pdf [https://‌perma.cc‌/Y3MX-QQJK]. ↑
171. .Id. at 31. ↑
172. .Why Traffic Stops Cause More Harm than Good, Innocence Proj. (July 19,
     2018), https://‌www.innocenceproject.org‌/why-traffic-stops-cause-more-harm-than-good‌/ [https://‌perma.cc‌/ZG86-YFUB]. ↑
173. .Racial Profiling Law Passes in Connecticut, CNN (May 8, 2012, 5:38 AM), https://‌www.cnn.com‌/2012‌/05‌/07‌/us‌/connecticut-racial-profiling‌/index.html [https://‌perma.cc‌/9FU2-9D94]. ↑
174. .ACLU Praises Lawmakers as Missouri Becomes Fourth State in Nation to Pass Racial Profiling Legislation, ACLU (May 12, 2000), https://‌www.aclu.org‌/news‌/aclu-praises-lawmakers-missouri-becomes-fourth-state-nation-pass-racial-profiling-legislation?redirect=racial-justice‌/aclu-praises-lawmakers-missouri-becomes-fourth-state-nation-pass-racial-profiling-leg [https://‌perma.cc‌/8KPS-AZXG]. ↑
175. .See, e.g., Terry v. Ohio, 392 U.S. 1, 27 (1968) (urging courts to defer to police “in light of . . . [their] experience”). The public also trusts the police, who rank above the Supreme Court,
     the President, and church or organized religion in terms of Americans’ confidence in United States’ institutions. Lydia Saad, Military, Small Business, Police Still Stir Most Confidence,
     Gallup (June 28, 2018), https://‌news.gallup.com‌/poll‌/236243‌/military-small-business-police-stir-confidence.aspx [https://‌perma.cc‌/EL8X-R2NE]. Of course, at this current moment, trust in the police is decreasing and there is growing, understandable outrage about policing tactics and policies. See, e.g., Rashawn Ray, What Does ‘Defund the Police’ Mean and Does It Have Merit?, Brookings (June 19, 2020), https://‌www.brookings.edu‌/blog‌/fixgov‌/2020‌/06‌/19‌/what-does-defund-the-police-mean-and-does-it-have-merit‌/ [https://‌perma.cc‌/8BZY-AVGH] (explaining the movement to defund the police). But whether this movement results in long-lasting change and reduced deference to police remains to be seen. See Maurice Mitchell & Sochie Nnaemeka, The Missing Element for Police Accountability: Political Will, Nation (June 5, 2020), https://‌www.thenation.com‌/article‌/politics‌/police-accountability-protests‌/ [https://‌perma.cc‌/8VST-WVU7] (explaining how the “posturing and rhetoric of police and elected officials . . . stands in stark contrast to their actions”). ↑
176. .Anna Lvovsky, The Judicial Presumption of Police Expertise, 130 Harv. L. Rev. 1995, 2016–22 (2017) (detailing the rise of the police expert in trials). ↑
177. .Joseph Goldstein, ‘Testilying’ by Police: A Stubborn Problem, N.Y. Times (Mar. 18, 2018), https://‌www.nytimes.com‌/2018‌/03‌/18‌/nyregion‌/testilying-police-perjury-new-york.html [https://‌perma.cc‌/M5WT-B8SM]. ↑
178. .Id. ↑
179. .Lvovsky, supra note 176, at 2003. ↑
180. .Blair Stenvick, Mayor Wheeler Defends Portland Police Bureau’s False “Cement Milkshake” Tweet, Portland Mercury (July 8, 2019, 3:41 PM), https://‌www.portlandmercury.com‌/blogtown‌/2019‌/07‌/08‌/26771738‌/mayor-wheeler-defends-portland-police-bureaus-false-cement-milkshake-tweet [https://‌perma.cc‌/64PP-7KJQ]. ↑
181. .Id. ↑
182. .Kelvin Henry, Elkhart Mayor Defends Elkhart’s Police Pursuit Policy After Two Deadly Police Pursuits, ABC (June 19, 2019, 12:47 AM), https://‌www.abc57.com‌/news‌/elkharts-mayor-neese-defends-elkharts-police-and-police-pursuit-policy-after-two-deadly-police-pursuits [https://‌perma.cc‌/5DYX-SSTU]. ↑
183. .Brentin Mock, How Rahm Emanuel Blew It on Police Reform, Bloomberg Citylab (Sept. 7, 2018, 3:19 PM), https://‌www.citylab.com‌/equity‌/2018‌/09‌/how-rahm-emanuel-blew-it-on-police-reform‌/569323‌/ [https://‌perma.cc‌/39E6-AHFD]. ↑
184. .Taylor Gee & Ben Wofford, America’s Mayors: ‘We’re Afraid for Our Police,’ Politico (Aug. 8, 2016), https://‌www.politico.com‌/magazine‌/story‌/2016‌/08‌/mayors-survey-police-shooting-law-enforcement-dallas-214152 [https://‌perma.cc‌/ZN2X-YMVX]. ↑
185. .Even in the face of tremendous public pressure to defund the police, the federal government was reluctant to confront the powerful law enforcement lobby. See Luke Broadwater & Catie Edmondson, Police Groups Wield Strong Influence in Congress, Resisting the Strictest Reforms, N.Y. Times (June 25, 2020), https://‌www.nytimes.com‌/2020‌/06‌/25‌/us‌/politics‌/police-reforms-congress.html [https://‌perma.cc‌/X638-HPJU]. ↑
186. .Barkow, supra note 122, at 728. ↑
187. .Seth W. Stoughton, The Incidental Regulation of Policing, 98 Minn. L. Rev. 2179, 2206–07, 2209–12, 2214–17 (2014) (detailing the collective bargaining power of police unions); Ross Douthat, Our Police Union Problem, N.Y. Times (May 2, 2015), https://‌www.nytimes.com‌/2015‌/05‌/03‌/opinion‌/sunday‌/ross-douthat-our-police-union-problem.html [https://‌perma.cc‌/6CMU-V4DS] (describing the influence of police unions in politics); Lobbying: Top Spenders, MapLight (2018), http://‌maplight.org‌/data‌/passthrough‌/#legacyurl=http://‌classic.maplight.org‌/us-congress‌/lobbying?client=Police [https://‌perma.cc‌/FFA4-W69F] (listing campaign contributions by police lobbying groups). ↑
188. .James Surowiecki, Why Are Police Unions Blocking Reform?, New Yorker (Sept. 12, 2016), https://‌www.newyorker.com‌/magazine‌/2016‌/09‌/19‌/why-are-police-unions-blocking-reform [https://‌perma.cc‌/A6LW-4SUM]. ↑
189. .Radley Balko, The Police Officers’ Bill of Rights, Wash. Post (Apr. 24, 2015, 2:58 PM), https://‌www.washingtonpost.com‌/news‌/the-watch‌/wp‌/2015‌/04‌/24‌/the-police-officers-bill-of-rights‌/?utm‌_term=.4f727b487437 [https://‌perma.cc‌/S2KW-QQTV]. ↑
190. .See supra note 150 and accompanying text. ↑
191. .Matthew Stabley, Berwyn Heights Mayor Fights for SWAT Oversight After Dogs’ Deaths, NBC Wash. (Mar. 30, 2009, 9:00 PM), https://‌www.nbcwashington.com‌/news‌/local‌/Berwyn-Heights-Mayor-Fights-for-SWAT-Oversight-After-Dogs-Deaths.html [https://‌perma.cc‌/T56L-TAHC]. ↑
192. .Radley Balko, Maryland Legislature Mulls New SWAT Transparency Bill, Wash. Post (Mar. 19, 2015, 3:32 PM), https://‌www.washingtonpost.com‌/news‌/the-watch‌/wp‌/2015‌/03‌/19‌/maryland-legislature-mulls-new-swat-transparency-bill‌/?utm‌_term=.ae525ffc5806 [https://‌perma.cc‌/6GKG-NERH]. ↑
193. .Radley Balko, Shedding Light on the Use of SWAT Teams, Wash. Post (Feb. 17, 2014, 10:30 AM), https://‌www.washingtonpost.com‌/news‌/the-watch‌/wp‌/2014‌/02‌/17‌/shedding-light-on-the-use-of-swat-teams‌/?noredirect=on&utm‌_term=.e1027b0febd1 [https://‌perma.cc‌/9D9V-ZNJD]. ↑
194. .Liam Dillon, Q&A: Here’s How California Officials Are Trying to Prevent Racial Profiling by Police, L.A. Times (Aug. 3, 2017, 12:05 AM), https://‌www.latimes.com‌/politics‌/la-pol-ca-police-racial-profiling-rules-revised-20170803-htmlstory.html [https://‌perma.cc‌/3PDB-PP4V]. ↑
195. .Joseph Wenner, Note, Who Watches the Watchmen’s Tape? FOIA’s Categorical Exemptions and Police Body-Worn Cameras, 2016 U. Chi. Legal F. 873, 878, 881, 891. ↑
196. .Id. at 873. ↑
197. .See Friedman & Ponomarenko, supra note 125, at 1863–64. ↑
198. .E.g., Community Control, Movement for Black Lives, https://‌m4bl.org‌/policy-platforms‌/community-control‌/ [https://‌perma.cc‌/8D2K-R2ZP] (explaining its vision for impacted communities to control the laws, institutions, and policies within communities); Policing Reform Campaign, NAACP (Jan. 1, 2015), http://‌www.naacpldf.org‌/case-issue‌/policing-reform-campaign [https://‌perma.cc‌/PK5W-HZ4M] (introducing the NAACP’s Legal Defense Fund’s campaign for police reform); Reforming Police, ACLU, https://‌www.aclu.org‌/issues‌/criminal-law-reform‌/reforming-police-practices [https://‌perma.cc‌/NTH9-P6UD] (explaining the ACLU’s initiative to reform the police); Street-Level Surveillance, Elec. Frontier Found., https://‌www.eff.org‌/issues‌/street-level-surveillance [https://‌perma.cc‌/R23V-CVTT] (explaining its Street-Level Surveillance Project, which calls attention to the surveillance technologies employed by law enforcement agencies). These groups have notable influence in legislative deliberations. See, e.g., Jeremy Gorner, ACLU, Chicago Police Agree to Changes on Controversial Street Stops, Chi. Trib. (Aug. 7, 2015, 6:01 AM), http://‌www.chicagotribune.com‌/news‌/ct-chicago-police-aclu-street-stops-met-20150807-story.html [https://‌perma.cc‌/9KWH-76SS] (detailing an agreement between the ACLU and Chicago Police Department to reform the department’s street stops); Governor Brown Signs Landmark Police Reform Legislation Sponsored by ACLU of California, ACLU
     (Sept. 30, 2018), https://‌www.aclunc.org‌/news‌/governor-brown-signs-landmark-police-reform-legislation-sponsored-aclu-california [https://‌perma.cc‌/4UBL-XBZM] (explaining how two bills sponsored by the ACLU will reform policing by increasing transparency and accountability). ↑
199. .The obvious analog here is the famous campaign ad against Michael Dukakis featuring Willie Horton, which while not about policing so much as parole, makes the point quite vividly. See Beth Schwartzapfel & Bill Keller, Willie Horton Revisited, Marshall Proj. (May 13, 2015, 6:37 PM), https://‌www.themarshallproject.org‌/2015‌/05‌/13‌/willie-horton-revisited [https://‌perma.cc‌/5WJC-WPVJ] (“[T]he threat of being ‘Willie Horton’ed’ has shaped the politics of crime and punishment.”). ↑
200. .See Barry Friedman, Secret Policing, 2016 U. Chi. Legal F. 99, 99 (“[T]here cannot be democratic engagement with policing if what those agencies do and how they do it is kept from the people.”); Rachel Harmon, Why Do We (Still) Lack Data on Policing?, 96 Marq. L. Rev. 1119, 1129–30 (2013) (discussing why so little information is collected by police departments). ↑
201. .Harmon, supra note 200, at 1130 n.32. ↑
202. .See, e.g., Police Mandatory Reporting, ACLU (Oct. 13, 2016), https://‌action.aclu.org‌/secure‌/we-need-data [https://‌perma.cc‌/37NH-CLDR] (noting that “police departments don’t often volunteer any of this data, even if they’re collecting it, because they’re concerned with their image and liability”). ↑
203. .Friedman, supra note 200, at 118. ↑
204. .Id.; see also Mitch Smith, Timothy Williams & Monica Davey, Jason Van Dyke Killed Laquan McDonald in 2014. Now Chicago Awaits a Verdict., N.Y. Times (Oct. 4, 2018), https://‌www.nytimes.com‌/2018‌/10‌/04‌/us‌/laquan-mcdonald-van-dyke-trial.html [https://‌perma.cc‌/UQ9W-62FQ] (asserting that a white police officer killed a Black teenager who posed no threat to him and the city’s response was to cover it up); Michael Wilson, Jesse McKinley, Luis Ferré-Sadurní, Troy Closson & Sarah Maslin Nir, Daniel Prudeʼs Death: Police Silence and Accusations of a Cover-Up, N.Y. Times (Sept. 10, 2020), https://‌www.nytimes.com‌/2020‌/09‌/04‌/nyregion‌/rochester-police-daniel-prude.html [https://‌perma.cc‌/X6YL-EU6T] (explaining that the police and city officials in Rochester withheld information about their handling of Prude’s death). ↑
205. .Friedman, supra note 200, at 120. ↑
206. .Timothy Williams, Study Supports Suspicion That Police Are More Likely to Use Force on Blacks, N.Y. Times (July 7, 2016), https://‌www.nytimes.com‌/2016‌/07‌/08‌/us‌/study-supports-suspicion-that-police-use-of-force-is-more-likely-for-blacks.html [https://‌perma.cc‌/DAQ2-P9AT]. ↑
207. .Matt Kiefer, Illinois Attorney General to Chicago Police: Fulfill Chicago Reporter’s FOIA Request, Chi. Rep. (May 3, 2019), https://‌www.chicagoreporter.com‌/illinois-attorney-general-to-chicago-police-fulfill-chicago-reporters-foia-request‌/ [https://‌perma.cc‌/8JXT-TCCM]. ↑
208. .UCR is now supplemented by the National Incident-Based Reporting System (NIBRS), which collects more detailed crime data. National Incident-Based Reporting System (NIBRS), FBI, https://‌www.fbi.gov‌/services‌/cjis‌/ucr‌/nibrs [https://‌perma.cc‌/KT63-MZVH]. ↑
209. .See Nathan James & Logan Rishard, Cong. Reseach Serv., RL34309, How Crime in the United States Is Measured 19 (2008) (“In some instances, a local law enforcement agency will submit only a few months’ worth of data . . . In other instances, a local law enforcement agency will submit offense data but will not submit any of the other data . . . .”); see also Michael D. Maltz, U.S. Dep’t of Justice, NCJ 215343, Analysis of Missingness in UCR Crime Data, 9-13 (2006) (measuring differing degrees of “missingness” in reported UCR data across jurisdictions); Clayton J. Mosher, Terance D. Mieth & Timothy C. Hart, The Mismeasure of Crime, 71 (2d ed. 2011) (finding that differences in how jurisdictions interpret and record crimes create ambiguity in UCR data). ↑
210. .See Wendy Ruderman, Crime Report Manipulation Is Common Among New York Policy, Study Finds, N.Y. Times (June 28, 2012), https://‌www.nytimes.com‌/2012‌/06‌/29‌/nyregion‌/new-york-police-department-manipulates-crime-reports-study-finds.html [https://‌perma.cc‌/4CXM-7PC2] (finding that data manipulation of crime reports—specifically, downgrading crimes to lower offenses and discouraging victims from filing complaints—has been an accepted practice in the NYPD for years). ↑
211. .Lopez, supra note 132 (explaining that the federal government’s collection efforts on police killing data misses almost half of all people killed by police in America). ↑
212. .Cynthia Lum, Cody W. Telep, Christopher S. Koper & Julie Grieco, Receptivity to Research in Policing, Just. Res. & Pol’y, June 2012, at 61, 64 (detailing the “large body of research” on policing interventions and police evaluations). ↑
213. .Id. at 63. ↑
214. .Patricio Domínguez & Steven Raphael, The Role of the Cost-of-Crime Literature in Bridging the Gap Between Social Science Research and Policy Making: Potentials and Limitations, 14 Criminology & Pub. Pol’y 589, 604 (2015) (explaining that cost-benefit analysis for a given policy requires a variety of costs, including the “collateral consequences of the policy for offenders, potential offenders, family of offenders, and so on”); Maria Ponomarenko & Barry Friedman, Benefit-Cost Analysis of Public Safety: Facing the Methodological Challenges, 8 J. Benefit-Cost Analysis 305, 314–15 (2017) (arguing that the social costs of policing, such as effects on privacy and community trust, have largely been ignored by social scientists); Brandon C. Welsh, David P. Farrington & Sema A. Taheri, Effectiveness and Social Costs of Public Area Surveillance for Crime Prevention, 11 Ann. Rev. L. & Soc. Sci. 111, 112 (2015) (“All too often the general discussions and evaluations of surveillance measures do not include any meaningful analysis of social costs.”). ↑
215. .Weisburd & Neyroud, supra note 1, at 9 (noting that policing research is often not easily transferable for use by law enforcement agencies). ↑
216. .Lum et al., supra note 212, at 66. ↑
217. .Weisburd & Neyroud, supra note 1, at 9. ↑
218. .Sherman, supra note 3, at 395 (“The direct pathway from specific demands for police innovations usually travels to police leaders through theory and precedent from other jurisdictions, but not through new research.”). There is also evidence that when cops know about research, they become more receptive to it. Lum et al., supra note 212, at 69. ↑
219. .See, e.g., Gary Cordner, The U.S. Needs a National Police University, in Contemporary Issues in Criminal Justice Policy: Policy Proposals from the American Society of Criminology Conference 279 (Natasha Frost et al. eds. 2010) (“We lack an effective system by which police leaders might keep up with the latest knowledge in their field.”); Lum, supra note 2, at 3 (“There is little indication that most American police leaders and their agencies systematically or regularly use tactics that are evidence based.”). ↑
220. .Lawrence W. Sherman, Evidence-Based Policing, Ideas Am. Policing, July 1998, at 1, 4. ↑
221. .Id. at 2–3. ↑
222. .E.g., Am. Soc. of Evidence-Based Policing, https://‌www.americansebp.org‌/ [https://‌perma.cc‌/NT8A-MQYF]; Canadian Soc. of Evidence-Based Policing, https://‌www.can-sebp.net‌/ [https://‌perma.cc‌/KPZ9-Q9WT]; Soc. of Evidence Based Policing, https://‌www.sebp.police.uk‌/ [https://‌perma.cc‌/TG34-2NT8]. ↑
223. .See Sherman, supra note 3, at 438 (“Evidence-based policing in 2013 still had a long way to go.”). ↑
224. .Chief Erik Josephson (@Chief‌_Josephson), Twitter (Oct. 6, 2018, 3:41 PM), https://‌twitter.com‌/Chief‌_Josephson‌/status‌/1048674590476656640 [https://‌perma.cc‌/R34L-7HW6]. ↑
225. .Barry Friedman & Kate Mather, Policing U.S. Style: With Little Idea of What Really Works, Just Security (July 10, 2019), https://‌www.justsecurity.org‌/64865‌/policing-u-s-style-with-little-idea-of-what-really-works‌/ [https://‌perma.cc‌/6QTJ-QXM7]. ↑
226. .Lum, supra note 2, at 3. ↑
227. .Weisburd & Neyroud, supra note 1, at 3. ↑
228. .Lum, supra note 2, at 3. ↑
229. .Sherman, supra note 3. ↑
230. .Id. at 381–82. ↑
231. .In New York (as elsewhere) there was a longstanding battle even to get officers’ past disciplinary records. Liz Robbins, Discipline Records of New York Police to Stay Secret, Court Rules, N.Y. Times (Dec. 11, 2018), https://‌www.nytimes.com‌/2018‌/12‌/11‌/nyregion‌/police-disciplinary-records-court-ruling-nyc.html [https://‌perma.cc‌/V4NN-4BUM] (explaining that the state’s highest court ruled against an effort to make police officers’ disciplinary records public). That legal barrier finally was removed after the protests in New York following the death of George Floyd and other Black individuals at the hands of police. See Ashley Southall, 323,911 Accusations of N.Y.P.D. Misconduct Are Released Online, N.Y. Times (Aug. 20, 2020), https://‌www.nytimes.com‌/2020‌/08‌/20‌/nyregion‌/nypd-ccrb-records-published.html [https://‌perma.cc‌/8FFJ-AL2S] (outlining content of police records that were released after state lawmakers repealed a law that kept them secret). ↑
232. .See, e.g., Justin Fenton, Baltimore Police Review Board Called Irrelevant, Ineffective, Balt. Sun (June 2, 2013, 6:44 PM), https://‌www.baltimoresun.com‌/news‌/maryland‌/crime‌/bs-md-ci-police-civilian-review-board-20130602-story.html [https://‌perma.cc‌/2K78-VRSP] (discussing the issues with the city’s review board, including the fact that its recommendations are rarely followed and that the complaints it reviews are not public information); Tim Lynch, How Mayors, Police Unions and Cops Rig Civilian Review Boards?, Hill (Oct. 24, 2016, 2:33 PM), https://‌thehill.com‌/blogs‌/pundits-blog‌/crime‌/302521-how-mayors-police-unions-and-cops-rig-civilian-review-boards [https://‌perma.cc‌/JCR5-S2AB] (noting how in Portland, the review board can only review the findings of the police department’s own internal affairs investigations, which investigates only a small fraction of civilian complaints). ↑
233. .See Barry Friedman, Unwarranted 81 (2017) (“[T]he perceived problems with the exclusionary rule have led judges to approve a lot of bad policing just to avoid imposing it.”); Lindsey Devers, Bureau of Justice Assistance, U.S. Dep’t of Justice, Plea and Charge Bargaining: Research Summary 1 (Jan. 24, 2011), https://‌www.bja.gov‌/Publications‌/PleaBargainingResearchSummary.pdf [https://‌perma.cc‌/V2DK-BUP9] (noting that about 90–95% of federal and state court cases are resolved through a plea bargain). ↑
234. .William Baude, Is Qualified Immunity Unlawful?, 106 Calif. L. Rev. 45, 46 (2018). ↑
235. .See Hans v. Louisiana, 134 U.S. 1, 21 (1890) (establishing the doctrine of sovereign immunity in which states cannot be sued by citizens). ↑
236. .See Monell v. Dep’t of Soc. Servs., 436 U.S. 658, 694 (1978) (holding that local governments can be sued under § 1983 for monetary, declaratory, and injunctive relief in situations where the alleged action executes a policy, ordinance, regulation, or decision officially adopted by the government). ↑
237. .See, e.g., Yoav Gonen, Julia Marsh & Bruce Golding, NYC Has Shelled Out $384M in 5 Years to Settle NYPD Suits, N.Y. Post (Sept. 4, 2018, 12:22 AM), https://‌nypost.com‌/2018‌/09‌/04‌/nyc-has-shelled-out-384m-in-5-years-to-settle-nypd-suits‌/ [https://‌perma.cc‌/2PFM-CKVG] (finding that over half of the suits brought against NYPD officers don’t go to trial). ↑
238. .Friedman & Ponomarenko, supra note 125, at 1839. ↑
239. .Ponomarenko, supra note 161, at 2 (discussing the lack of administrative rulemaking procedures in place despite numerous scholars calling for their implementation). ↑
240. .Amy Sinden, Cass Sunstein’s Cost-Benefit Lite: Economics for Liberals, 29 Colum. J. Envtl. L. 191, 199 (2004). ↑
241. .Exec. Order No. 12291, 3 C.F.R. 128 (1982). ↑
242. .John D. Graham, Saving Lives Through Administrative Law and Economics, 157 U. Pa. L. Rev. 395, 453 (2008). ↑
243. .Id. ↑
244. .Cass R. Sunstein, Cognition and Cost-Benefit Analysis 10 (Univ. Chi. Law Sch. Coase-Sandor Inst. for Law & Econ., Working Paper No. 85, 1999). ↑
245. .Michael A. Livermore, Can Cost-Benefit Analysis of Environmental Policy Go Global?, 19 N.Y.U. Envtl. L.J. 146, 160 (2011). ↑
246. .One of the country’s most prominent criminologists makes this point repeatedly. See, e.g., Anthony A. Braga, William H. Sousa, James R. Coldren Jr. & Denise Rodriguez, The Effects of Body-Worn Cameras on Police Activity and Police-Citizen Encounters: A Randomized Controlled Trial, 108 J. Crim. L. & Criminology 511, 513 (2018) (noting how little scientific evidence there is on the effects of police technology); Braga et al., supra note 38, at 659 (2012) (“[W]e were surprised that none of the 19 hot spots policing evaluations reviewed here [in this meta-analysis] conducted formal cost-benefit assessments.”). ↑
247. .Harmon, supra note 5, at 901–05. Harmon offers some provisional ways to begin monetizing these important costs. Id. at 905–12; see also Barkow, supra note 16 (advocating for CBA to reduce mass incarceration); Richard A. Bierschbach & Stephanos Bibas, Rationing Criminal Justice, 116 Mich. L. Rev. 187, 187 (2018) (stating that governments should approach the administration of punishment just like other public-law problems of regulation). ↑
248. .Barkow, supra note 119, at 1277. ↑
249. .Id. at 1284. ↑
250. .See Friedman, supra note 233, at 59–60 (explaining that police agencies are granted broad authority under their enabling statutes, and that there are few constraints on how they exercise
     that authority). ↑
251. .See supra section I(B)(2). ↑
252. .Another issue with police being the sole decision-makers is that they often look for the fastest, rather than the best, solution to a problem. Lum et al., supra note 212, at 62 (explaining that practitioners and researchers operate on very different timelines, with police looking for quick solutions and academics emphasizing deliberation). ↑
253. .Compare Cynthia Lum & Daniel S. Nagin, Reinventing American Policing, 46 Crime & Just. 339, 340 (2017) (“Crimes averted, not arrests made, should be the primary metric for judging police success . . . .”), with Malcolm K. Sparrow, Measuring Performance in a Modern Police Organization, New Persps. Policing 13 (2015) (stating that when New York City implemented aggressive order maintenance tactics, high arrest rates “seem[ed] to have been a source of pride”). ↑
254. .Michael Tonry, The Fog Around Cost-of-Crime Studies May Finally Be Clearing: Prisoners and Their Kids Suffer Too, 14 Criminology & Pub. Pol’y 653, 660 (2015) (explaining that the costs of incarceration include, among other things, offenders’ lost earnings, the reduced life chances and living standards for their children). ↑
255. .See Billy Henson & Bradford W. Reyns, The Only Thing We Have to Fear Is Fear Itself . . . and Crime: The Current State of the Fear of Crime Literature and Where It Should Go Next, 9 Soc. Compass 91, 99 (2015) (arguing that “fear of crime” is a serious problem that reduces quality of life); Charles F. Manski & Daniel S. Nagin, Assessing Benefits, Costs, and Disparate Racial Impacts of Confrontational Proactive Policing, 114 Proc. Nat’l Acad. Sci. 9308, 9308 (2017) (providing a model of optimal policing that assesses the benefit of a policing technique in terms of its crime reduction). ↑
256. .See Rachel A. Harmon, Why Arrest?, 115 Mich. L. Rev. 307, 359 (2016) (arguing that arrests are not an effective way to deter crime). ↑
257. .See Bierschbach & Bibas, supra note 247, at 196–97 (describing how the fragmentation of law enforcement among local, state, and federal authorities allows institutions to avoid “feel[ing] the full bite of their costs”). ↑
258. .See William J. Stuntz, Accountable Policing 51 (Harvard Law Sch. Harvard Public
     Law, Working Paper No. 130, 2006) (“Local taxpayers get the large majority of the deterrent benefits of the prison terms local police and prosecutors seek, and they pay for the large majority of police forces’ and district attorneys’ budgets. But they pay only a small fraction of the cost of
     prison beds.”). ↑
259. .See Bierschbach & Bibas, supra note 247, at 196–97. ↑
260. .See Lisa A. Robinson, Harvard Kennedy Sch. Ash Ctr. for Democratic Governance and Innovation, Benefit-Cost Analysis and the Cities (2015) (“Conducting benefit-cost analysis has its own benefits and costs.”). ↑
261. .Cass R. Sunstein, Congress, Constitutional Moments, and the Cost-Benefit State, 48 Stan. L. Rev. 247, 292 (1996). ↑
262. .See Cass R. Sunstein, The Real World of Cost-Benefit Analysis: Thirty-Six Questions (and Almost as Many Answers) 3 (Harvard Law Sch. Pub. Law & Legal Theory Working Paper Series, Paper No. 13-11, 2013) (praising CBA as a “mini-constitution for the regulatory state” that helps decisionmakers structure their thinking). ↑
263. .See, e.g., Mary Spicuzza, City Aims to Crack Down on Prostitution, Milwaukee J. Sentinel, July 11, 2017, at A2 (describing complaints from neighbors as spurring a police investigation into an alleged brothel). ↑
264. .See The State of Intelligence Reform 10 Years After 9‌/11: Hearing Before the H. Permanent Comm. on Intelligence, 109th Cong. 9–10 (2011) (statement of Robert S. Mueller III, Dir., Fed. Bureau of Investigations) (discussing the transformation of the FBI in the aftermath of 9‌/11). ↑
265. .See, e.g., Carol Morello & Jamie Stockwell, No Attacks, No Arrests, No Shortage
     of Anxiety: Area Faces New Week Under the Sniper’s Gun, Wash. Post, Oct. 14, 2002,
     at A14 (describing the police response to the Beltway Sniper, which included choking off roads
     and creating traps for the shooter); Jon Silman & Amy B. Wang, Fourth Slaying Stokes Fear
     of a Tampa Serial Killer—and Police Say Suspect May Be an Area Resident, Wash. Post
     (Nov. 14, 2017, 9:43 PM), https://‌www.washingtonpost.com‌/news‌/true-crime‌/wp‌/2017‌/11‌/14‌/fourth-death-stokes-fear-of-a-tampa-serial-killer-and-police-say-suspect-may-live-among-them‌/?
     utm‌_term=.afc14c97de77 [https://‌perma.cc‌/BQ8Q-H97F] (explaining that in the wake of four unsolved murders in the Tampa area, investigators were “working nonstop to track the killer”). ↑
266. .See Jeffrey Bellin, The Inverse Relationship Between the Constitutionality and Effectiveness of New York City “Stop and Frisk”, 94 B.U. L. Rev. 1495, 1507 (2014) (stating that the goal of New York’s stop-and-frisk program was to eradicate unlawful gun possession). ↑
267. .See Catherine L. Carpenter, Legislative Epidemics: A Cautionary Tale of Criminal Laws that Have Swept the Country, 58 Buff. L. Rev. 1, 66–67 (2010) (arguing that criminal laws created hastily in response to crises can have troubling consequences). ↑
268. .See Friedman, supra note 200, at 103–05 (describing the exchange of federal subsidization of Stringray purchases in exchange for law enforcement silence). ↑
269. .See Ted R. Miller, Mark A. Cohen & Brian Wiersema, Nat’l Inst. of Justice, Victim Costs and Consequences: A New Look 9 (1996) (citing the types of victim costs that have been identified in past research). ↑
270. .Barkow, supra note 119, at 1277–79 (noting politics and tough-on-crime rhetoric has led to three decades of longer sentences and more incarceration). ↑
271. .See, e.g., Patrick Svitek, Koch Network Targets Texas in Push for Prison Reform in Trump Era, Tex. Trib. (Feb. 1, 2018, 12:00 AM), https://‌www.texastribune.org‌/2018‌/02‌/01‌/koch-network-targets-texas-push-prison-reform-trump-era‌/ [https://‌perma.cc‌/RHA7-ZY9A] (reporting on various efforts of the billionaire brothers to push prison reform); State Reforms Reverse Decades of Incarceration Growth, Pew Charitable Tr. (Mar. 21, 2017), https://‌www.pewtrusts
     .org‌/en‌/research-and-analysis‌/issue-briefs‌/2017‌/03‌/state-reforms-reverse-decades-of-incarceration-growth [https://‌perma.cc‌/JWD2-6UTV] (“Since 2007, more than 30 states have enacted policies to reverse corrections growth and contain costs . . . .”). ↑
272. .See Kevin Strom, Nat’l Inst. of Justice, Research on the Impact of Technology on Policing Strategy in the 21st Century, Final Report 6–32 (2016) (finding that ongoing costs associated with data storage and handling for technology such as body-worn cameras and license plate readers are a common concern for agencies, and have been a barrier to full implementation). ↑
273. .See supra note 109 and accompanying text. ↑
274. .See supra notes 6–8 and accompanying text. ↑
275. .See Mark A. Cohen, The “Cost of Crime” and Benefit-Cost Analysis of Criminal Justice Policy: Understanding and Improving upon the State-of-the-Art 16–17 (Aug. 31, 2016) (manuscript), https://‌papers.ssrn.com‌/abstract=2832944 [https://‌perma.cc‌/RJQ6-7XGQ] (noting the importance of considering social benefits in benefit–cost analyses of criminal justice policy). ↑
276. .See Cass R. Sunstein, The Cost-Benefit Revolution 163 (2018) (defining opportunity cost as the resources used or benefits forgone as a result of choosing the given policy). ↑
277. .See Cohen, supra note 275, at 16 (“[O]ne of the tenants [sic] of benefit-cost analysis is that the analyst should consider several feasible alternative policies and choose the one that yields the highest net social benefits.”). ↑
278. .See W. David Ball, Pay-for-Performance in Prison: Using Healthcare Economics to Improve Criminal Justice, 94 Denv. L. Rev. 451, 458–59 (2017) (lamenting the failure of actors in the criminal justice system to consider downstream costs). ↑
279. .Richard L. Revesz & Michael A. Livermore, Retaking Rationality: How Cost-Benefit Analysis Can Better Protect the Environment and Our Health 55 (2008) (“In seeking to minimize certain risks, regulations can exacerbate others. A serious analysis of a regulation’s impact should pay attention not only to how it reduces the target risk, but also to how it raises these countervailing risks.”). ↑
280. .See Domínguez & Raphael, supra note 214, at 600 (noting examples of policy proposals that pass cost–benefit analysis but have questionable distributive effects); Dripps, supra note 119, at 1094–95 (“The aggregate amount of happiness is of course important . . . . But the distribution of happiness matters too, and at some point it becomes unfair to impose great hardship on an individual for a very slight gain for collective community security.”). ↑
281. .See Office of Mgmt. & Budget, Exec. Office of the President, Circular A-4 (2003), https://‌www.whitehouse.gov‌/sites‌/whitehouse.gov‌/files‌/omb‌/circulars‌/A4‌/a-4.pdf [https://‌perma.cc‌/4DYE-2Z84] [hereinafter Circular A-4] (“Transfers occur when wealth or income is redistributed without any direct change in aggregate social welfare.”). ↑
282. .See supra notes 50–56 and accompanying text. ↑
283. .See Richard L. Revesz, Regulation and Distribution, 93 N.Y.U. L. Rev. 1489, 1490–92 (2018) (describing as “dominant” the view that individual regulations need only maximize the “pie,” i.e., net benefits, and that Congress “should be in charge of figuring out the sizes of the respective slices”). ↑
284. .See id. at 1492 (explaining how the dominant view on regulatory policy implies that the executive branch “should be in the business of increasing the size of pies,” while Congress “should be in charge of figuring out the sizes of the respective slices”). ↑
285. .See id. at 1491 (stating that the prevailing academic view is that distributional issues should be dealt with through the tax system). ↑
286. .See id. at 1519–20 (arguing that the notion that Congress will take action on distribution through the tax system is faulty given congressional gridlock). ↑
287. .See Dripps, supra note 119, at 1094–95 (arguing that it is unfair to impose the burdens of excessive law enforcement on a minority of citizens). ↑
288. .See President’s Task Force on 21st Century Policing, Final Report of the President’s Task Force on 21st Century Policing 9 (2015) (reporting a decline in public confidence in police work among communities of color and attributing that low confidence to
     mass incarceration). ↑
289. .See supra notes 208–11 and accompanying text. ↑
290. .See Brandon C. Welsh, David P. Farrington & B. Raffan Gowar, Benefit-Cost Analysis of Crime Prevention Programs, 44 Crime & Just. 447, 458 (“The most convincing method of evaluating crime prevention programs is the randomized experiment.” (citations omitted)). ↑
291. .See id. at 458 (referring to the use of areas as units of randomization). ↑
292. .See Michael Abramowicz, Ian Ayres & Yair Listokin, Randomizing Law, 159 U. Pa. L. Rev. 929, 935–36 (2011) (“Since the distribution of . . . [every characteristic other than the treatment] is the same in both the control and the treatment groups, we can attribute any differences in the average group response to the difference in treatment.” (emphasis omitted)). ↑
293. .There are other problems. In the lab, one can make sure that nothing is happening in the treatment areas different from the control areas except the treatment. In the real world, this is tougher. One check on this is to have enough treatment areas that this is unlikely to be the case vis-à-vis the control areas. See Welsh et al., supra note 290, at 458 (noting that a large number of experimental units is required for effective randomized experiments but cautioning that it is difficult to reach that number when the units are areas). ↑
294. .See Rebecca Goldstein, Evaluating Proactive Police Units: A Case Study of Retrospective Benefit-Cost Analysis with Nonexperimental Data, 8 J. Benefit-Cost Analysis 348, 350 (2017) (stating that a problem with RCTs is that police must treat similarly situated high-crime neighborhoods based only on the need for randomization). ↑
295. .See, e.g., David Weisburd, Hot Spots Policing Experiments and Criminal Justice Research: Lessons from the Field, 559 Annals Am. Acad. Pol. & Soc. Sci. 220, 229 (2005) (stating that the “main problem” in executing a randomized study of hot spot policing was “gain[ing] the cooperation of a major police agency”). ↑
296. .See Jillian B. Carr, Estimating the Effects of Police Technology Using Quasi-Experimental Methods, 8 J. Benefit-Cost Analysis 360, 366 (2017) (noting that “[a]ny scenario in which parts of a city are treated at different times could provide . . . an opportunity” for a natural experiment). ↑
297. .Compare Rachel Bayefsky, Note, Dignity as a Value in Agency Cost-Benefit Analysis, 123 Yale L.J. 1732, 1765 (2014) (“Efforts to monetize dignity are misguided.”), with Richard T. Carson, Contingent Valuation: A Practical Alternative When Prices Aren’t Available, 26 J. Econ. Persp. 27, 28 (2012) (“[G]oods and impacts that cannot be quantified are valued, implicitly, by giving them a limitless value when government regulations preclude certain activities, or giving them a value of zero by leaving certain consequences out of the analysis.”). ↑
298. .See Steve Aos, What Is the Bottom Line?, 14 Criminology & Pub. Pol’y 633, 634 (2015) (noting CBA is not a substitute for program evaluation and if benefits are not found in the initial review, there is no hope that a full benefit–cost analysis can add useful information). ↑
299. .See Circular A-4, supra note 281 (recommending that policymakers exercise professional judgment in determining how important the non-quantified benefits or costs may be in the context of the overall analysis). ↑
300. .See id. (presenting qualitative descriptions as an alternative to cost–benefit measuring through physical units). ↑
301. .See Domínguez & Raphael, supra note 214, at 593 (“[C]ost–benefit analysis provides a clear set of criteria by which society and policy makers can rank alternative policy choices and pass judgment on specific proposals.”). ↑
302. .ACLU & City of Chicago Agreement on Stop and Frisk, ACLU Ill. (Aug. 5,
     2015), https://‌www.aclu-il.org‌/sites‌/default‌/files‌/wp-content‌/uploads‌/2015‌/08‌/2015-08-05-Stop-and-frisk-settlement-one-pager.pdf [https://‌perma.cc‌/MYG3-4JQD]. ↑
303. .See Aos, supra note 298, at 635 (presenting break even analysis as a helpful way to “reverse engineer” useful information). ↑
304. .See Harmon, supra note 5, at 905–12 (providing basic methods for estimating the costs of police coercion). ↑
305. .See Domínguez & Raphael, supra note 214, at 605 (noting hedonic pricing as one approach to calculating the cost of crime). ↑
306. .See, e.g., Carson, supra note 297, at 28 (“Contingent valuation studies ask questions that help to reveal the monetary tradeoff each person would make concerning the value of goods or services.”). ↑
307. .Id. at 30–31. ↑
308. .See Center for Quality Policing, RAND Corp., https://‌www.rand.org‌/jie‌/justice-policy‌/centers‌/quality-policing.html [https://‌perma.cc‌/TMJ2-N4FC]; Crime Lab, U. of Chi. Urban Labs, https://‌urbanlabs.uchicago.edu‌/labs‌/crime [https://‌perma.cc‌/T26Y-YPKL]; Policing Project, Policing Project, N.Y.U. Sch. of Law, https://‌www.policingproject.org [https://‌perma.cc‌/ST22-GBZF]. ↑
309. .See Nathan James, Cong. Research Serv., R44938, FY2018 Appropriations for Department of Justice 2 (2017) (noting $89 million in actual appropriations for research, evaluation, and statistics in 2017 and a request for $111 million in funding for 2018). ↑
310. .Norman E. Sharpless, M.D., Plans for NCI’s Fiscal Year 2019 Budget, NIH: Nat’l Cancer Inst. (Dec. 18, 2018), https://‌www.cancer.gov‌/news-events‌/cancer-currents-blog‌/2018‌/sharpless-nci-fy2019-budget [https://‌perma.cc‌/PVU2-P2FP]. ↑
311. .Compare Press Release, U.S. Dep’t of Justice, Attorney General Sessions Announces $98 Million to Hire Community Policing Officers (Nov. 20, 2017), https://‌www.justice.gov‌
     /opa‌/pr‌/attorney-general-sessions-announces-98-million-hire-community-policing-officers [https://‌perma.cc‌/Z4C7-2RF4] (offering priority in grant applications for law enforcement agencies that agree to certify their willingness to cooperate with federal immigration authorities), with Press Release, U.S. Dep’t of Justice, Department of Justice Awards $12 Million to Advance Community Policing Efforts and Collaborative Reform (Oct. 6, 2016), https://‌www.justice.gov‌/opa‌/
     pr‌/department-justice-awards-12-million-advance-community-policing-efforts-and-collaborative [https://‌perma.cc‌/3NF3-JU3L] (announcing grants for developing innovative strategies for community engagement). ↑
312. .Comprehensive estimates of private philanthropic funding of education and public health are readily available. See Yoku Shaw-Taylor, Nongovernment Philanthropic Spending on Public Health in the United States, 106 Am. J. Pub. Health 58, 58 (2016) (estimating the total amount of private philanthropic funding of public health in the United States at approximately $203 billion annually); Giving USA 2019: Americans Gave $427.71 Billion to Charity in 2018 Amid Complex Year for Charitable Giving, Giving USA (June 18, 2019, 3:07 PM), https://‌givingusa
     .org‌/giving-usa-2019-americans-gave-427-71-billion-to-charity-in-2018-amid-complex-year-for-charitable-giving‌/ [https://‌perma.cc‌/SM45-RTB5] (estimating the amount of private philanthropy given in support of education in the United States at $58.72 billion for 2018). In contrast, we were unable to track down a reliable estimate of the amount of private philanthropic funding for criminal justice collectively, let alone the narrower domain of policing. Arnold Ventures is one of the only major private philanthropies funding criminal justice research. The sum of Arnold Ventures’ grants awarded for policing research from 2013 to 2019 is approximately $40.7 million. See 1,000 Projects and Counting, Arnold Ventures, https://‌www.arnoldventures.org‌/grants-list?focus=criminal-justice&issue=policing [https://‌perma.cc‌/VA8Y-TWWP] (presenting raw data of the organization’s donations to various police initiatives, the sum of which is approximately $40.7 million). ↑
313. .See Georges Benjamin, Why Our Justice System Is a Hazard to Our Health, Cal. Endowment, http://‌www.calendow.org‌/why-our-justice-system-is-a-hazard-to-our-health [https://‌perma.cc‌/78S3-3HG5] (Sept. 15, 2015) (“Ensuring justice and equity in our criminal justice system strengthens our ability to build healthy communities and create a healthier nation.”). ↑
314. .Kristina Strain, The Joyce Foundation’s Long Fight to Curb to Gun Violence, Inside Philanthropy (June 6, 2014), https://‌www.insidephilanthropy.com‌/home‌/2014‌/6‌/6‌/the-joyce-foundations-long-fight-to-curb-to-gun-violence.html [https://‌perma.cc‌/H2AH-59G5] (reporting that the “highest priority” of the Joyce Foundation’s Gun Violence Prevention program is public health-related funding). ↑
315. .Research on Gun Violence, Arnold Ventures, https://‌www.arnoldventures.org‌/work‌/research-on-gun-violence‌/ [https://‌perma.cc‌/SHF8-A2F9]. ↑
316. .See Nat’l Inst. of Justice, Paving the Way: Lessons Learned in Sentinel Event Reviews 2 (Nov. 2015), https://‌www.ncjrs.gov‌/pdffiles1‌/nij‌/249097.pdf [https://‌perma.cc‌/GNB4-9Q9H] (assessing the feasibility of implementing sentinel event reviews in the criminal justice system because such reviews have been successfully employed, presumably system-wide, in medicine and aviation). ↑
317. .Id. Batts et al. advocate for the use of “organizational accident models,” a kind of sentinel event review, to learn from and prevent wrongful convictions. See Anthony W. Batts, Maddy deLone & Darrel W. Stephens, Policing and Wrongful Convictions, New Persp. In Policing 16 (2014), https://‌www.ncjrs.gov‌/pdffiles1‌/nij‌/246328.pdf [https://‌perma.cc‌/J94Z-4AC2] (“The model accepts that mistakes are inevitable in a human-dominated system . . . . It shifts the focus to how inherent system weaknesses converge with these individual mistakes to lead to tragedies.”); James M. Doyle, A “Safety Model” Perspective Can Aid Diagnosis, Prevention, and Restoration After Criminal Justice Harms, 59 Santa Clara L. Rev. 107, 109 (2018) (proposing “Criminal Justice Safety Centers” as mechanisms “to analyze criminal justice disasters collaboratively and learn
     their lessons”). ↑
318. .See Empowering Local Communities to Advance Justice: NIJ and BJA Launch Sentinel Events Initiative National Demonstration Collaborations, Nat’l Inst. of Just. (Sept. 28, 2017), https://‌nij.ojp.gov‌/topics‌/articles‌/empowering-local-communities-advance-justice-nij-and-bja-launch-sentinel-events [https://‌perma.cc‌/8KU5-QRLK]. ↑
319. .Batts et al., supra note 317, at 16. ↑
320. .See id. at 16–17 (arguing that nonpunishment-based inquiries are better at generating information about why accidents happen and how they can be prevented). ↑
321. .Sentinel Events Initiative, Nat’l Inst. of Just., https://‌nij.gov‌/topics‌/justice-system‌/Pages‌/sentinel-events.aspx#ser [https://‌perma.cc‌/25G4-ALHU] (Nov. 1, 2017); see also D. Kim Rossmo & Joycelyn M. Pollock, Confirmation Bias and Other Systemic Causes of Wrongful Convictions: A Sentinel Events Perspective, 11 Ne. U. L. Rev. 790, 793–96 (2019) (describing NIJ’s sentinel event approach). ↑
322. .Critical Incident Review Board (CIRB), City of Tucson, https://‌www.tucsonaz.gov‌/police‌/critical-incident-review-0 [https://‌perma.cc‌/9KFV-VV3Z]. ↑
323. .Id. ↑
324. .Wash. St. Inst. for Pub. Pol’y, https://‌www.wsipp.wa.gov‌/ [https://‌perma.cc‌/LBP9-3CJ2]. ↑
325. .See Publications, Wash. St. Inst. for Pub. Pol’y, https://‌www.wsipp.wa.gov‌/Publications [https://‌perma.cc‌/MC98-GSDS] (listing several criminal justice-related projects and no public safety or policing related projects). ↑
326. .Exec. Order No. 12866, 58 Fed. Reg. 51735 (Oct. 4, 1993). ↑
327. .Benefit-Cost Results, Wash. St. Inst. for Pub. Pol’y, https://‌www.wsipp.wa.gov‌/BenefitCost [https://‌perma.cc‌/DH68-R9YC]. ↑
328. .Chad Marlow & Maryiam Saifuddin, How to Stop ‘Smart Cities’ from Becoming ‘Surveillance Cities,’ ACLU (Sept. 17, 2018, 9:45 AM), https://‌www.aclu.org‌/blog‌/privacy-technology‌/surveillance-technologies‌/how-stop-smart-cities-becoming-surveillance-cities [https://‌perma.cc‌/4RSU-WPWA]. ↑
329. .Seattle, Wash., Mun. Code § 14.18.20 (2013). ↑
330. .Community Control over Police Surveillance, ACLU, https://‌www.aclu.org‌/
     issues‌/privacy-technology‌/surveillance-technologies‌/community-control-over-police-surveillance [https://‌perma.cc‌/SZS7-S7RB] (including graphic of cities participating in CCOPS movement, both cities that have passed CCOPS laws and cities working on CCOPS legislation). ↑
331. .Cambridge Passes Law Requiring Community Control of Police Surveillance, ACLU Mass. (Dec. 10, 2018, 9:45 PM), https://‌www.aclum.org‌/en‌/news‌/cambridge-passes-law-requiring-community-control-police-surveillance [https://‌perma.cc‌/TN3H-KB8E]. ↑
332. .See Email from Farhang Heydari, Deputy Dir., Policing Project, to Barry Friedman, Jacob D. Fuchsberg Professor of Law, N.Y.U. School of Law (Nov. 29, 2018, 3:44 PM) (on file with authors); see also Jim Cleary, Racial Profiling Studies in Law Enforcement: Issues and Methodology (Minn. H.R. Res. Dep’t, June 2000), https://‌www.house.leg.state.mn.us‌/hrd‌/pubs‌/raceprof.pdf [https://‌perma.cc‌/UM9S-3LA9] (detailing the types of racial profiling data collected by states and municipalities and the prevalence of the collection); Portland Police Bureau, Plan to Address Racial Profiling (Jan. 2009), https://‌www.portlandoregon.gov‌/police‌/article‌/230887 [https://‌perma.cc‌/WY5W-KB9N] (articulating the goal of collecting “the right data on police stops” within Portland). ↑
333. .ACLU of Tennessee’s “Campaign Against Racial Profiling” Achieves Another Success with Passage of Traffic Stops Bill, ACLU (May 12, 2005), https://‌www.aclu.org‌/press-releases‌/aclu-tennessees-campaign-against-racial-profiling-achieves-another-success-passage [https://‌perma.cc‌/4CQK-UAGS] (“Data collection is critical for discerning any patterned practice of racial profiling by law enforcement . . . .”). ↑
334. .Sharon LaFraniere & Andrew W. Lehren, The Disproportionate Risks of Driving While Black, N.Y. Times (Oct. 24, 2015), https://‌www.nytimes.com‌/2015‌/10‌/25‌/us‌/racial-disparity-traffic-stops-driving-black.html?login=email&auth=login-email [https://‌perma.cc‌/H6QD-VFKG]. ↑
335. .Id. ↑
336. .Id. (noting that the City Council of Fayetteville required officers in 2012 to obtain written consent before searching a car, a practice that was endorsed by a White House task force on policing in 2015); see also What Decades of Traffic Stop Data Reveals About Police Bias, CBS News (Mar. 2, 2019, 11:43 AM), https://‌www.cbsnews.com‌/news‌/what-decades-of-traffic-stop-data-reveals-about-police-bias‌/ [https://‌perma.cc‌/LR8Q-7FPR] (detailing the data in North Carolina showing racial disparities in traffic stops and the resulting report that had recommendations adopted by various jurisdictions); Jim Wise, Written Consents Are Fayetteville Policy, Raleigh News & Observer (Sept. 3, 2014, 6:52 PM), https://‌www.newsobserver.com‌/news‌/local‌/counties‌/durham-county‌/article10049000.html [https://‌perma.cc‌/D9Z3-HJZE]. ↑
337. .Louisiana Law Enforcement Agencies Must Collect, Publish Data on Police Activity, SPLC (Nov. 15, 2018), https://‌www.splcenter.org‌/news‌/2018‌/11‌/15‌/splc-report-louisiana-law-enforcement-agencies-must-collect-publish-data-police-activity [https://‌perma.cc‌/DR5V-H4RM]. ↑
338. .Peter Hirschfeld, Many Vermont Police Agencies Aren’t Following Traffic-Stop Data Collection Law, N.H. Pub. Radio (Apr. 25, 2016, 6:05 PM), https://‌www.nhpr.org‌/post‌/many-vermont-police-agencies-arent-following-traffic-stop-data-collection-law#stream‌/0v [https://‌perma.cc‌/Z9UQ-835V]. ↑
339. .Christopher Damien, California High Court Cans Cop Union’s Attempt to Block New
     Law Requiring Police Misconduct Records Disclosure, Desert Sun (Jan. 3, 2019,
     12:25 PM), https://‌www.desertsun.com‌/story‌/news‌/crime‌_courts‌/2019‌/01‌/02‌/california-supreme-court-declines-lawsuit-delay-sb-1421-release-police-misconduct-records‌/2469806002‌/ [https://‌perma.cc‌/Q9TU-Q4TD]. ↑
340. .See Stephanie Wykstra, The Fight for Transparency in Police Misconduct, Explained, Vox (June 16, 2020, 7:30 AM), https://‌www.vox.com‌/2020‌/6‌/16‌/21291595‌/new-york-section-50-a-police-misconduct [https://‌perma.cc‌/6P82-FVCB]. ↑
341. .Sarah Breitenbach, States Pursue Varied Police Reforms Amid National Debate, Gov’t Tech. (June 1, 2015), https://‌www.govtech.com‌/state‌/States-Pursue-Varied-Police-Reforms-Amid-National-Debate.html [https://‌perma.cc‌/9YMV-VRKT]. ↑
342. .Jonathan Wolfe, Marijuana in New York: Here’s How the Laws Are Changing, N.Y. Times (June 20, 2018), https://‌www.nytimes.com‌/2018‌/06‌/20‌/nyregion‌/marijuana-laws-new-york.html [https://‌perma.cc‌/B2FK-DV3S]. ↑
343. .Grace E. Leeper, Note, Conditional Spending and the Need for Data on Lethal Use of Police Force, 92 N.Y.U. L. Rev. 2053, 2065 (2017). ↑
344. .Anthony Braga, Rod K. Brunson, Gary Cordner, Lorie Fridell, Matthew Hickman, Cynthia Lum, Stephen D. Mastrofski, Jack McDevitt, Dennis P. Rosenbaum, Wesley G. Skogan & William Terrill, Recommendations to the President’s Task Force on 21st Century Policing, Am. Soc’y of Criminology (Feb. 12, 2015) (noting that the failure “is not for lack of wanting or trying, but for lack of the resources necessary to undertake the serious and sustained program of research and development that is necessary” and recommending that Congress immediately allocate funds to the DOJ to fulfill the mandate). ↑
345. .Leeper, supra note 343, at 2088. ↑
346. .Id. ↑
347. .See Ponomarenko, supra note 161, at 45 (proposing that regulatory intermediaries “could help shift the incentive structure that discourages police rulemaking, provide more systemic oversight over police policies and practices, and even facilitate broader public input and debate”). ↑
348. .See The Function and Role of the Board of Police Commissioners, L.A. Police Dep’t, http://‌www.lapdonline.org‌/police‌_commission‌/content‌_basic‌_view‌/900 [https://‌perma.cc‌/T4Q9-NKBC]. ↑
349. .Elizabeth Zima, LA Will Iron Out Body-Cam Transparency Rules, Gov’t
     Tech (Mar. 14, 2018), https://‌www.govtech.com‌/public-safety‌/LA-Will-Iron-Out-Body-Cam-Transparency-Rules.html [https://‌perma.cc‌/8KLS-YN33] (discussing a survey of members of the public and not-for-profit organizations on the proposed video release policy). ↑
350. .Liam Dillon, California Legislature Passes Major Police Transparency Measures on Internal Investigations and Body Cameras, L.A. Times (Sept. 1, 2018, 3:25 PM), https://‌www.latimes.com‌/politics‌/la-pol-ca-police-transparency-bill-passes-20180831-story.html [https://‌perma.cc‌/8W2T-H3QC] (noting that the California bill was modeled on the LAPD’s body-camera policy). ↑
351. .Oakland, CA., Ord. 13349 C.M.S. (2016) (describing the role of the Privacy
     Advisory Commission). ↑
352. .Office of Inspector General, City of Chicago, https://‌igchicago.org‌/ [https://‌perma.cc‌/5XSY-2G3V]; Office of Inspector General, County of Los Angeles, https://‌oig.lacounty.gov [https://‌perma.cc‌/4ZSG-2V5L]; Inspector General for the NYPD, https://‌www1.nyc.gov‌/site‌/doi‌/offices‌/oignypd.page [https://‌perma.cc‌/QQ2T-EEDS]. ↑
353. .Juan Perez Jr., Police Who Patrol Chicago Schools Lack Recruitment, Training Standards, City Inspector General Says, Chi. Trib. (Sept. 14, 2018, 6:30 AM), https://‌www.chicagotribune.com‌/news‌/breaking‌/ct-met-chicago-police-school-officers-inspector-general-20180913-story.html [https://‌perma.cc‌/C9NN-SJYR]. In response, the Chicago Police Department committed to making changes such as creating a specialized training program for the police, for the 2019–20 school year, which, though a step in the right direction, was not as immediate as the IG suggested. Id. ↑
354. .Mick Dumke, Chicago’s Inspector General Finds the City’s Gang Database Is Riddled with Errors, ProPublica Ill. (Apr. 11, 2019, 10:00 AM), https://‌www.propublica.org‌/article‌/chicago-police-department-gang-database-inspector-general-report [https://‌perma.cc‌/7LRT-KL7M]; Sanjana Karanth, Inspector General Finds Chicago Gang Database Outdated, Inaccurate and Damaging, HuffPost (Apr. 11, 2019, 9:06 PM), https://‌www.huffpost.com‌/entry‌/
     chicago-inspector-general-police-gang-database‌_n_5cafd2b2e4b082aab0831bf7?guccounter=1&
     guce_referrer=aHR0cHM6Ly93d3cuZ29vZ2xlLmNvbS8&guce_referrer_sig=AQAAAGrpSBmHWKLZExO6GCDabo3QHmpDv5iu5kdnBzPvup5gYGxijXesEZMk5_gZwYd8RRJH5_I2eejfdSzTHDfjDvP6TN6Y7aZxeViyPIcoQcUur8d0IJ1qrbM3VVX6tWG0AhNrlnpqa‌_-eZZOLBTq93Upc
     64DlFR3Nk2WiwjQcwMHU [https://‌perma.cc‌/3YGY-97C4]. ↑
355. .Ben Brachfeld, City Council Evaluates Progress of NYPD Inspector General, Gotham Gazette (Nov. 16, 2017), https://‌www.gothamgazette.com‌/city‌/7325-city-council-examines-progress-under-nypd-inspector-general [https://‌perma.cc‌/PK8P-XK4P]; Steve Inskeep, NYC Police Chief Calls Inspector General’s Critical Report ‘Grandstanding,’ NPR (Oct. 2, 2015, 4:57 AM), https://‌www.npr.org‌/2015‌/10‌/02‌/445216747‌/new-york-city-police-get-new-orders-on-using-force [https://‌perma.cc‌/X543-Q78N]. ↑
356. .Barkow, supra note 16, at 182 (arguing for an IG model where the IG “evaluates the agency’s substantive policies to make sure they are using limited resources efficiently and where the IG also verifies the policies are being implemented as claimed”). ↑
357. .Mark Puente, LAPD Ends Another Data-Driven Crime Program Touted to Target Violent Offenders, L.A. Times (Apr. 12, 2019, 4:48 PM), https://‌www.latimes.com‌/local‌/lanow‌/la-me-laser-lapd-crime-data-program-20190412-story.html [https://‌perma.cc‌/JTD9-LGYK]. ↑
358. .Id. ↑
359. .See id. (noting that the LAPD ended the program due to data inconsistencies, and that the audit report found the program used inconsistent criteria). ↑
360. .Christopher Slobogin, Policing as Administration, 165 U. Pa. L. Rev. 91, 137 (2016). ↑
361. .Friedman & Ponomarenko, supra note 125, at 1856. ↑
362. .See Policing Project, N.Y.U. Sch. of Law, Report to the NYPD
     Summarizing Public Feedback on Its Proposed Body-Worn Camera Policy (2017), https://‌static1.squarespace.com‌/static‌/58a33e881b631bc60d4f8b31‌/t‌/59ce7edfb0786914ba448d82‌/1506705121578‌/Report+to+the+NYPD+Summarizing+Public+Feedback+on+BWC+Policy.pdf [https://‌perma.cc‌/PCG5-PZYM] (describing the NYPD’s request for the Policing Project to solicit public feedback and the way the organization received comments); Public Feedback on
     the CCPD’s Body Worn Camera Policy: Report on the Comments Received
     and the CCPD’s Response, Policing Proj., https://‌static1.squarespace.com‌/
     static‌/58a33e881b631bc60d4f8b31‌/t‌/59cea1543e00be785b9fa3c3‌/1506713946337‌/Report‌_Public
     +Feedback+on+the+CCPD%27s+Body-Worn+Camera+Policy.pdf [https://‌perma.cc‌/7DS5-RJEU] (describing a similar request for the Policing Project to solicit public feedback). ↑
363. .See, e.g., Communities United for Police Reform and Others Slam NYPD Body Camera Policy, Defending Rights & Dissent (Apr. 7, 2017), https://‌rightsanddissent.org‌/news‌/communities-united-police-reform-others-slam-nypd-body-camera-policy‌/ [https://‌perma.cc‌/UKK2-PBZB] (noting New York activitists’ criticism of the new policy on body cameras). ↑
364. .See Heather Cherone, Don’t Create a New Gang Database, ACLU Tells Lightfoot and Chicago Police, Block Club Chi. (May 13, 2019, 8:47 AM), https://‌blockclubchicago.org‌/2019‌/05‌/13‌/dont-create-a-new-gang-database-aclu-tells-lightfoot-and-chicago-police‌/ [https://‌perma.cc‌/2G2R-ADH4] (criticizing the Chicago Police Department’s decision to create a new
     gang database). ↑
365. .Criminal Enterprise Information System Policy, Chi. Police Dep’t, https://‌home.chicagopolice.org‌/reform‌/policy-review‌/criminal-enterprise-information-system‌/ [https://‌perma.cc‌/772J-RUC4]. ↑
366. .Heather Cherone, Deadline to Weigh In on Chicago Police’s New Gang Database Is Saturday, Block Club Chi. (May 10, 2019, 9:00 AM), https://‌blockclubchicago.org‌/2019‌/05‌/10‌/deadline-to-weigh-in-on-chicago-polices-new-gang-database-is-saturday‌/ [https://‌perma.cc‌/2VWS-8QWG]. ↑
367. .Barkow, supra note 16, at 178. ↑
368. .Id. at 183. ↑
369. .Cordner, supra note 219. ↑
370. .See id. at 285 (“Simply put, the NPU [(National Policing University)] should take the responsibility for producing and compiling knowledge about policing and making it available to all those who need it.”). ↑
371. .Lum et al., supra note 212, at 62–63. ↑
372. .About Us, College of Policing, https://‌www.college.police.uk‌/About‌/Pages‌/default.aspx [https://‌perma.cc‌/4R24-6UHV]. ↑
373. .About College What Works, College of Policing, https://‌whatworks.college.police.uk‌/About‌/Pages‌/default.aspx [https://‌perma.cc‌/32XS-ASQU]. ↑
374. .Id. ↑
375. .Gillian Hunter, Tiggey May & Mike Hough, An Evaluation of ‘What Works Centre for Crime Reduction’ Final Report 10 (2017), https://‌whatworks.college.police.uk‌/About‌/Documents‌/ICPR‌_Final‌_Evaluation‌_WWCCR.pdf [https://‌perma.cc‌/GFV8-FJ35]. ↑
376. .Cost Benefit Tool, College of Policing, https://‌whatworks.college.police.uk‌/Research‌/Pages‌/Cost‌_Benefit‌_Tool.aspx [https://‌perma.cc‌/AMJ9-JZZB]. ↑
377. .Cordner, supra note 219, at 279 (“We lack an effective system by which police leaders might keep up with the latest knowledge in the field.”). ↑
378. .About CrimeSolutions, CrimeSolutions, Nat’l Inst. of Just., https://‌www
     .crimesolutions.gov‌/about.aspx [https://‌perma.cc‌/K8SA-83DW]. ↑
379. .See About ASEBP, Am. Soc’y of Evidence-Based Policing, https://‌www
     .americansebp.org‌/about‌_asebp.php [https://‌perma.cc‌/EC39-MR4C] (describing ASEBP’s desire to use “research findings to develop those strategies that are best designed to reduce crime and disorder while maintaining the trust and confidence of the public”); Research Briefs, Am. Soc’y of Evidence-Based Policing, https://‌mms.americansebp.org‌/members‌/mlogin.php?org‌_id=SEBP&mc=Y&bounce=https%3A%2F%2Fwww.americansebp.org%2Fresearch‌_briefs.php [https://‌perma.cc‌/55GM-DAZU] (providing research briefs on policing tactics to members of ASEBP). ↑
380. .Katheryn Giglio, Helping Police Find Better Strategies to Fight Crime, RAND Corp.: RAND Blog (Dec. 14, 2018), https://‌www.rand.org‌/blog‌/articles‌/2018‌/12‌/helping-police-find-better-strategies-to-fight-crime.html [https://‌perma.cc‌/H6LF-DWS9]. ↑
381. .Evidence-Based Policing Matrix, Ctr. for Evidence-Based Crime Pol’y, https://‌cebcp.org‌/evidence-based-policing‌/the-matrix‌/ [https://‌perma.cc‌/9G7C-RMLL]. The Police Foundation also created an evidence-based policing application for iPhones, which is powered by the Matrix. The Evidence-Based Policing App, App Store Preview, https://‌apps.apple.com‌/us‌/app‌/evidence-based-policing‌/id1047674581 [https://‌perma.cc‌/2BTT-34E2]. ↑
382. .Cordner, supra note 219, at 279 (noting that it is very unlikely that police chiefs are familiar with evidence-based policing). ↑