EQUALITY IN THE STREETS Using Proportionality Analysis to Regulate Street Policing

Abstract

The racially disparate impact and individual and collective costs of stop and frisk, misdemeanor arrests, and pretextual traffic stops have been well documented. Less widely noticed is the contrast between Supreme Court case law permitting these practices and the Court’s recent tendency to strictly regulate technologically enhanced searches that occur outside the street policing setting and that—coincidentally or not—happen to be more likely to affect the middle class. If, as the Court has indicated, electronic tracking and searches of digital records require probable cause that evidence of crime will be found, stops and frisks should also require probable cause that a crime has been committed (in the case of stops) or that evidence of crime will be found (in the case of post-detention searches). This equalization of regulatory regimes not only fits general notions of fairness. It is also mandated by the Fourth Amendment’s Reasonableness Clause and the Court’s cases construing it, which endorse a “proportionality principle” that requires that the justification for a search or seizure be roughly proportionate to its intrusiveness.Applying the probable cause requirement to the streets would not prevent police from carrying out investigative detentions when they believe criminal activity “may be afoot.” Rather, it would limit such detentions and subsequent searches to situations where they observe or have another good basis for believing that a person has engaged or is engaging in an attempted crime as defined by the law of the jurisdiction. While Terry v. Ohio and some of the Supreme Court’s other stop cases would still come out the same way under this formulation, judicial decisions that have permitted stops of people simply because they avoided the police, fit a “profile,” acted in a “furtive” manner, or appeared out of place would not. The equalization of street and technological policing would bring even more significant changes to the law governing searches incident to arrest, which could occur only when police have probable cause to believe the person has a weapon or possesses evidence of crime (although, given the fact that an arrest has been made, handcuffing would be permitted). Imposing the probable cause standard developed in technological search cases on street policing can promote equality without sacrificing public safety.In Terry v. Ohio1 the Supreme Court held that, despite the Fourth Amendment’s reference to “probable cause,” police may subject a person to a brief seizure (a stop) on a lesser “reasonable suspicion” showing that “criminal activity may be afoot” and then may conduct a pat-down (a frisk) if they develop reasonable suspicion that the person is armed and dangerous. The Terry Court gave two reasons for its invention of the reasonable-suspicion standard. First, stops and frisks are less intrusive than arrests and full searches and therefore can take place on less than the probable cause required to take people into custody or to search them for evidence.2 Second, police need some mechanism, short of arrest, that allows them to nip crime in the bud and protect themselves and others while doing so.3 These explanations notwithstanding, critics have claimed that Terry is not supported by the Fourth Amendment’s language and history,4 hands too much power to the police,5 and opens the door to laxer standards whenever government can argue that individual interests are weaker or government interests stronger than in the typical case.6Terry has been particularly controversial among those concerned about racialized policing, who have lambasted the decision’s relaxation of the probable-cause standard as an invitation for police to harass people of color. The available data back up these claims. In many American cities, tens of thousands of Black people are subjected to stops every year, well out of proportion to their presence in the city’s population or their engagement in illegal activity.7 Yet in cities like New York during the height of its stop-and-frisk campaign, the percentage of the millions of stopped people who were found to have a gun was well below one percent, the percentage of those arrested for any crime after a stop was well below ten percent, and these “hit rates” were significantly lower for Black people than for white people.8While stopping and frisking pedestrians is the aspect of street policing most maligned by those concerned about racial justice, stops are certainly not street policing’s only, or even primary, component. Millions of times a year, police arrest people outright (rather than merely stopping them based on reasonable suspicion), claiming to have probable cause to arrest for a misdemeanor or traffic violation. And, once again, people of color are disproportionately subjected to such arrests.9 Black people are also more likely than members of other racial groups to be taken into custody for a misdemeanor crime10 and more likely to have their car unsuccessfully searched after a traffic violation.11The individual toll of street policing can be significant. A stop and frisk on the street or a search of one’s car is humiliating and can easily turn into a violent event unless one complies meticulously with every police command (especially when one is Black).12 Detention on a misdemeanor charge, even one that does not result in conviction, can last for weeks or even months (especially for people of color), with all the personal and familial disruption that entails.13Street policing’s collective toll is also huge. Subjection of urban communities to routine stops and arrests for the offenses of “walking while Black” or “driving while Latino” raises tensions,14 sends the message that people of color are inferior human beings,15 ruptures relationships with the police,16 and undermines the legitimacy of the government.17 Police patrolling practices are a key reason that calls for defunding the police or abolishing police departments outright have become popular with groups like the Movement for Black Lives.18The police will say that street policing is an important way of keeping weapons off the streets, forestalling incipient crime, discovering people with outstanding warrants, serendipitously finding evidence of more serious crime, and, in general, keeping a handle on the neighborhood.19 They are backed up by research suggesting that “aggressive policing” produces higher arrest rates for robbery, decreases various types of thefts and gun crimes, and increases seizures of guns.20 But there are also numerous counter-studies suggesting that this type of policing is not very effective at reducing crime or gun violence.21Proposals to mitigate the negative effects of street policing are legion. In an effort to deter the use of street encounters as pretexts to harass or conduct searches incident to arrest, reformers have advocated for decriminalization of misdemeanors,22 the conversion of misdemeanors to citation-only offenses,23 or the reduction of custodial arrests more generally.24 For the same reason, some propose that police should no longer have the authority to make traffic stops (a task that would be off-loaded to unarmed government officials)25 or should be prohibited from asking for consent to search a stopped car.26 And, most significantly, reformers want stop and frisk to be eliminated; Terry, they contend, should be reversed.27No jurisdiction has seriously pursued any of these ideas, however. And far from contemplating these moves, courts have facilitated the current state of affairs. Undoubtedly Terry opened the door here, but that decision could have been interpreted narrowly. Instead, the Supreme Court and lower courts have pushed it in the opposite direction. Stops are allowed for conduct that falls far short of criminal activity, and frisks are permitted virtually automatically if a stop has occurred.28 And the Court has relied on Terry’s reasoning and the Fourth Amendment’s Reasonableness Clause to bolster decisions that not only permit but incentivize the abuse of misdemeanor arrests and traffic stops as pretexts for acting on mere hunches, racially charged or otherwise.29This much has been well-documented. Less widely noticed is that Terry and related case law stand in telling contrast to the Supreme Court’s recent tendency to put a damper on searches that occur outside the street policing setting and that—coincidentally or not—are much more likely to affect the middle class (and, concomitantly, white people). In Jones v. United States,30 for instance, the Court required a court order before prolonged GPS tracking, and in United States v. Carpenter,31 the Court required a warrant before cell site location data may be obtained. Counterposed to the Court’s Terry jurisprudence, these cases expose a glaring inequality in Fourth Amendment jurisprudence. The types of searches that took place in Jones, Carpenter, and other Court cases involving technologically enhanced investigations can work significant infringements of privacy. But they do so remotely and covertly and thus do not trigger the humiliation, stigmatization, and fear that accompany stops and frisks and other types of publicly visible street policing. Not surprisingly, survey research (some of which I have conducted) strongly suggests that, on a spectrum of intrusiveness, stops and frisks are perceived to be similar in impact to the types of searches addressed in Jones and Carpenter.32If technological tracking and searches of digital records require probable cause that evidence of crime will be found, stops and frisks should require probable cause that a crime has been committed (in the case of stops) or that evidence of crime will be found (in the case of frisks). This equalization of regulatory regimes not only fits general notions of fairness. It can also be derived directly from the Fourth Amendment. For some time, I have been promoting the argument that the amendment’s Reasonableness Clause and the Court’s cases construing it, including Terry, endorse a “proportionality principle,” which posits that the justification for a search or seizure should be roughly proportionate to its intrusiveness.33 Following that reasoning, if stops and frisks are as intrusive as searches that the Court has said require probable cause, they should no longer be permissible on mere reasonable suspicion.The proposition that, in Fourth Amendment terms, street policing and the types of searches involved in Jones and Carpenter are on the same footing sits in some tension with Terry’s two rationales for not requiring probable cause on the streets: the fact that short detentions and frisks are less intrusive than custodial arrests and full searches of the person, and the need for a mechanism that allows police to act before crime occurs.34 But neither rationale justifies Terry’s result. While a stop is not an arrest, it still visits constraints on liberty and autonomy that are as intrusive as the privacy invasions associated with the Court’s technological search cases. And applying the probable cause requirement of those cases to the streets would not prevent police from carrying out investigative detentions when they believe “criminal activity may be afoot.” Rather, it would limit such detentions and subsequent searches to situations where they observe or have another good basis for believing that a person has engaged or is engaging in an attempted crime as defined by the law of the jurisdiction.As applied to detentions on the street, this rule would not veer significantly from the original Terry holding, which involved an attempted robbery.35 The difference would be that rather than applying the amorphous reasonable suspicion standard that Terry conjured out of thin air, police would be implementing the substantive criminal law—specifically, the law of attempts, the primary purpose of which is to criminalize the inchoate behavior that precedes crime. Compared to the courts’ interpretation of Terry, attempt jurisprudence would cabin police preemptive practices relatively precisely by requiring observation of conduct that amounts to a substantial step toward or dangerous proximity to commission of a crime.36 While Terry and a few other Supreme Court stop cases would come out the same way under this formulation, the many cases approving stops of people because they avoided the police, fit a “profile,” acted in a “furtive” manner, or appeared out of place would not.37The proposed equalization of street policing with technological policing would bring even more significant changes to the law governing post-detention searches. If the probable-cause standard currently gaining ascendancy in technological policing cases were also applied to post-detention searches, not only the law of frisks but the law of searches incident to arrest would need to change. Under today’s jurisprudence, if a person is arrested for a custodial offense, police may automatically conduct a full search of the arrestee for evidence and weapons, not just frisk for weapons.38 In contrast, converting Terry’s frisk doctrine into one that requires probable cause to carry out a search, as I propose, would mean that a search incident could not be automatic. Rather, it could occur only when police have probable cause to believe the person has a weapon or possesses evidence of crime. To minimize the temptation for police to manufacture a more serious crime after they discover a weapon or evidence during a suspicionless search of a misdemeanant, they should also have to announce the offense of arrest at the time it occurs.While this would work a significant change in search-incident law, note that because the predicate for detention would be an arrest, police could protect themselves and others with handcuffing or other types of restraints.39 Further, when the arrest is for an attempt to commit or commission of a violent or serious crime (such as the robbery suspected in Terry), probable cause to believe a weapon is present would usually exist. Similarly, if police arrest someone in connection with a drug crime, they usually would have probable cause to search for evidence of drug possession or sale. It is only when police arrest someone for a nonviolent crime, such as jaywalking, trespass, or a traffic infraction, that such cause would usually be lacking. And that is precisely when neither a frisk nor a full search should be permitted. The Supreme Court itself has come close to recognizing that norm in the traffic-stop context by prohibiting searches of cars incident to arrest unless there is reason to believe that evidence of the offense of arrest will be found.40Thus, the equalization of street policing with technological search jurisprudence would not only significantly restrict stop and frisk practices. It would also have an impact on the other two most-abused aspects of street policing: misdemeanor arrests and traffic stops. In doing so, it would help rectify the racial imbalance inherent in today’s street-policing practices. Even if other reforms of these two components of street policing remain stymied, recognizing that equality norms require probable cause for nonconsensual detentions and post-detention searches could go a long way toward accomplishing the same goals.After rehearsing current law under Terry, this article makes clear how disproportionate that law is to the law governing other types of searches and seizures. It then explains further the implications of that comparison for street policing.The Fourth Amendment places very few constraints on policing of the streets. Beginning with the Terry decision, the Supreme Court has approved a wide array of practices that give the police close-to-unlimited discretion to do as they will. While Terry jurisprudence on stops and on frisks is the prime example of this permissive attitude, the Court’s subsequent decisions on the exclusionary rule, search incident to arrest doctrine, vagueness doctrine, and the (ir)relevance of police motivations have also broadened and deepened that power.Start with Terry and stop and frisk. At the Supreme Court level alone, over two dozen cases address the definition of reasonable suspicion. In all of them, the Court has tried to distinguish between “hunches,” on the one hand, and, on the other, “a particularized and objective basis for suspecting the particular person stopped of criminal activity.”41 Applying this distinction, the Court has said that police cannot stop a car near the border solely on the basis that it is full of people of apparent Mexican ancestry42 but may stop a vehicle near the border that is full of people of apparent Mexican ancestry whose occupants fail to acknowledge the agent and wave “oddly” at him.43 It has said that police may not stop individuals in a “high drug problem area” who walk away from one another when they see the police44 but may stop an individual in a high-crime area who runs away from the police.45 It has invalidated an airport stop of a person who gets off a plane from a supposed drug-source city with no luggage other than a shoulder bag and an apparent desire to conceal that he is traveling with another person46 but permitted an airport stop of a person coming from a drug-source city who was pale and nervous, appeared to have heavy luggage, used cash to pay for his ticket, and did not provide full identifying information on his luggage tags.47Although there are differences between these pairs of cases, they are ephemeral and have nothing to do with whether the reasons police give for a stop are “particularized” or “objective.”48 Factors such as ancestry, the criminality of a given location, and the city of departure are general characteristics that apply to hundreds of thousands of people. And whether a person has failed to acknowledge the police, waved “oddly,” left the scene quickly, or appeared nervous are all subjective judgments, as is the relevance of these factors to whether the person acted “suspiciously.” Perhaps realizing this, time and again the Court has stated that police decisions to stop a person should be viewed from the perspective of those who are “versed in the field of law enforcement”49 and have “specialized training.”50 In effect, this formulation hands the definition of “reasonable suspicion” over to the police.David Harris’s survey of how lower-court decisions applied Terry twenty-five years ago supports that view.51 Harris found that while the courts usually concluded that mere presence in a high-crime area is an insufficient basis for a stop (presaging the Supreme Court’s 2000 decision in Illinois v. Wardlow52), many also decided that such presence combined with evasion can constitute reasonable suspicion, and a few held that avoiding the police, by itself, is a ground for a stop.53 Harris also noted that many courts allowed police to take “racial incongruity” into consideration “along with other factors.”54 Finally, he found that “almost any time that the crime suspected in a Terry situation involves drugs, courts routinely allow a frisk following a stop as a matter of course” without distinguishing, for instance, between a stop on suspicion of drug trafficking or of simple possession.55It is unlikely that courts have tightened things up in the past quarter century. In fact, in 2018, Harris and David Rudofsky, relying on data from New York and Philadelphia, noted how frequently the police relied on factors such as a “bulge,” uncooperativeness, hands in pockets, presence in a high-crime neighborhood, nervousness, furtive movements, and “flight,” or some combination thereof, in stopping individuals, and yet how infrequently—well under one percent of the time—the stops produced a weapon or evidence of serious crime.56 Although they did not report how New York and Philadelphia courts reacted to these stops, presumably the police did not fear any judicial repercussions in these cities.Police may be unconcerned about judicial pushback not only because the courts’ interpretation of Terry has granted them considerable discretion but because the threat of exclusion, which is the courts’ only real tool for monitoring street policing, has no direct effect on them, something Terry itself recognized;57 while the police may occasionally get feedback about the constitutionality of their stops during a suppression hearing, their main concern is meeting their stop quotas and getting weapons and drugs off the streets.58 Additionally, any deterrent effect that exclusion may have vis-à-vis illegal stops has been further weakened by the Supreme Court’s decision in Utah v. Strieff,59 which held that even evidence found during an unconstitutional stop is admissible if, after the stop, police discover an outstanding arrest warrant that authorizes a custodial arrest and a search incident to that arrest. Since in many jurisdictions tens of thousands of these warrants exist, often for picayune offenses like failure to pay court fees or traffic violations,60Strieff incentivizes stops on less than reasonable suspicion.61As Strieff illustrates, the law of searches incident to arrest makes matters worse. Under the Court’s decisions, if the police have probable cause to arrest, they can automatically conduct a full search of the person, not just a frisk, regardless of the crime of arrest.62 The only limitation the Fourth Amendment imposes on such searches is that the officer must be willing to take the person into custody for an offense for which custody is authorized,63 the latter an issue usually left entirely up to the officer or the police department.64 Although some state courts have interpreted their own constitutions to prohibit searches incident to arrest in connection with minor traffic infractions and the like,65 the Supreme Court has refused to do so, providing additional incentive to detain people for minor infractions.66Using vagueness doctrine developed under the Due Process Clause rather than the Fourth Amendment, the Court has struck down loitering laws—laws that, ironically, criminalized the type of suspicious conduct that Terry jurisprudence says can form the basis for a stop—on the ground that these statutes give the police too much discretion!67 Read generously, these cases at least prevent a full search on mere suspicion of wrongdoing. But vagueness doctrine is of virtually no help in regulating most street policing. As William Stuntz pointed out,68 most of the statutes on which police rely to go after petty antisocial conduct are not vague; rather, they are found in criminal codes that, relatively precisely, penalize an inordinately broad array of routine conduct from which the police can pick and choose. Indeed, traffic violations have been called the new general warrant—not because they fail to particularly describe when police may act but because their ubiquity means that all of us violate them all the time (think not just of speeding and red-light laws, but seat-belt, stop-sign, cell-phone, traffic-lane, turn-signal, and equipment violations).69 Misdemeanors such as criminal trespass, jaywalking, subway turnstile jumping, simple drug possession, loitering, vandalism, and disorderly conduct serve a similar function with respect to pedestrians.70 In league with Terry and Strieff, these statutes give police authority to stop people based on reasonable suspicion (or less) and carry out tens or hundreds of thousands of unnecessary frisks and searches each year.Finally, lurking in the background of all of this, but highly influential at both the street and the court level, is the Supreme Court’s refusal to investigate police motives or beliefs. Most famous in this vein is the unanimous decision in Whren v. United States71 holding that the Fourth Amendment has nothing to say about searches triggered by a mere hunch (or worse) as long as police can also point to a legitimate reason for their arrest or stop. But the Court has pushed the pretext envelope even further with its decisions upholding the validity of searches that occur after an arrest for a different crime than the one committed,72 or crimes that are not even on the statute books,73 as long as a reasonable officer could have concluded that the arrest was valid based on the known facts. These cases may be understandable on formal grounds (the Fourth Amendment prohibits only “unreasonable” searches) or practical ones (reading an officer’s mind is difficult). But, together with the Supreme Court’s stipulation that proof of discriminatory intent must be shown to prove a violation of the Equal Protection Clause,74 they allow both police and courts to turn a blind eye toward biased and racialized policing practices.Street policing will continue to generate abysmally low hit rates, disproportionately affect people of color, disrupt police-community relations, and provide fodder for the anti-policing movement unless the legal bases for stops, frisks, and searches incident to arrest are radically changed. New developments in the Court’s search jurisprudence may provide a rationale for doing so.The same term it handed down Terry, the Supreme Court decided Katz v. United States,75 which has come to stand for the proposition that a Fourth Amendment search occurs whenever the police infringe an “expectation of privacy … that society is prepared to recognize as ‘reasonable.’”76 An electronic surveillance case, Katz was meant to move the scope of the Fourth Amendment beyond the property orientation that had permeated the Court’s earlier cases to a more capacious conception. Neither the phone booth that the police bugged in Katz nor Katz’s conversation over the booth’s phone was a house, person, paper, or effect (the four “constitutionally protected areas” found in the Fourth Amendment) as those terms are traditionally defined.77 Nor did the placement of a bugging device on top of the booth work a common-law trespass.78 Yet the Court’s sense was that the privacy of Katz’s communication was worth constitutional protection and that its interception required a warrant.79However, over the next twenty years, the Court developed four doctrines that pretty much confined application of Katz to government attempts to discover the content of private communications. Through its “knowing exposure” doctrine, the Court left unprotected from surveillance any activity that an individual knowingly displays to the public.80 The Court’s “general public use” doctrine allowed police to use any technology generally available to private citizens to observe not just public activities but goings-on inside the home.81 The “evidence-only” doctrine declared that an investigation that revealed only evidence of crime and no other fact (as allegedly occurs with drug-detection dogs) is not a Fourth Amendment search.82 And the “assumption of risk” or “third party” doctrine provided that information that a person knows or should know is in the possession of a third party assumes the risk that the third party will turn that information over to the government.83 If these doctrines apply, the Court pronounced, the Fourth Amendment is irrelevant; the police do not need even reasonable suspicion, much less probable cause.Beginning at the dawn of the twenty-first century, however, the Court began a slow about-face. The first doctrine to take a hit, albeit a glancing one, was the general-public-use doctrine. In Kyllo v. United States,84 the Court held that police need a warrant to use a thermal imager to gauge heat differentials inside a house, even though such devices are relatively easy to obtain. Eleven years later the Court decided Jones v. United States,85 which more directly attacked both the general-public-use and knowing-exposure doctrines in holding that planting a simple GPS device on a car and tracking it is a Fourth Amendment search, even if the car travels only on public thoroughfares. A year later, in Jardines v. Florida,86 the Court held that a warrant is needed to use a drug-sniffing dog near a house, putting a dent in the evidence-only doctrine. And, in its most momentous decision in this vein to date, in Carpenter v. United States87 it held that a warrant is required to obtain a person’s cell site location data from the person’s common carrier, a decision that the dissenters declared marked the beginning of the end for the third-party doctrine.88For present purposes, the importance of this set of decisions—what I will call the Court’s technological search cases—is in their contrast to the holdings in Terry and its progeny. Whereas a stop involves a physical confrontation that can legally last for up to twenty minutes and often goes much longer89 and a frisk involves systematically running hands over a person’s limbs and crotch, all of which takes place in full view of the public,90 technological searches occur remotely and covertly. While the searches in Kyllo, Jardines, Jones, and Carpenter all involved serious invasions of privacy—a point I have emphasized in my writing91—they did not trigger the humiliation, physical discomfort, or fear and anger that a stop and frisk does, particularly for people of color.92 Surveys that I and others have conducted asking participants to rate the “intrusiveness” of various types of policing practices confirm that a pat-down is viewed as more intrusive than short-term tracking and thermal imaging93 and as about as intrusive as dog sniffs,94 albeit somewhat less intrusive than accessing months of location data and financial records.95 Yet under the Fourth Amendment today, only technological searches require probable cause that a crime is occurring or has occurred. The Court could have adopted the reasonable suspicion standard in any of its technological search cases; in Carpenter, the government explicitly asked it to do so.96 Yet the Court has insisted on probable cause in this setting, not the reasonable-suspicion test that applies in stop-and-frisk cases.Formal explana