Introduction
Over the summer of 2020, the nation faced yet another reckoning over policing and use of force.1 Once again, the nation stood witness to mass protests against the use of force in policing and, more broadly, the fraught relationship between communities of color and the police officers that are meant to protect them.2 Once again, police reform is being seriously discussed at multiple levels of government across cities in multiple states.3 And, once again, renewed interest in police reform has led activists to propose a broad variety of policy, legislative, and judicial proposals.4
In order to investigate the racially disparate impact of policing, it is imperative to critically interrogate the nature of police-civilian interactions and how these interactions may be infected by individual bias. Traffic stops represent one of the most common forms of police-civilian contact, and thus are a fitting category of interaction to serve as a target for this type of inquiry.5 Traffic laws in the United States are generally extremely broad; they tend to “regulate the details of driving in ways both big and small, obvious and arcane.”6 The breadth of such codes and the latitude offered to officers in enforcing them make traffic violations potent openings that officers can exploit to investigate persons they would not otherwise have the requisite suspicion to investigate.7 As a result, millions of people each year interact with the criminal legal system by way of traffic enforcement.8 While traffic enforcement may seem harmless in the long run, perhaps even crucial to public safety, the reality is that traffic stops can serve as an opening to more severe penalties. From a valid traffic stop, an officer may, for example, order the driver to exit the vehicle,9 run background and related records checks on the driver,10 and conduct certain limited searches of the vehicle as well as the driver.11 Consequently, disproportionate application of traffic enforcement will necessarily reinforce racialized patterns in policing, as it will continue to bring certain groups into contact with the criminal legal system more often than others.
For decades, activists and community members have lamented the traffic stop as a vehicle for impermissible racial profiling.12 Generally defined, racial profiling is the practice of targeting individuals for crime enforcement or criminal investigation “based on [an] individual’s race, ethnicity, religion or national origin.”13 While some states currently have laws that meaningfully prohibit racial profiling, the majority do not.14 Even fewer specifically prohibit the use of race as a factor to any degree in selecting who to subject to routine investigation, like a short traffic stop.15
Laws against racial profiling may seem like a commonsense place to start in terms of addressing the racially disparate impacts of policing. In response, this Note makes two arguments. First, even the most robust bans against race-based policing are alone insufficient to address the issue of policing’s racially disparate effects, which can be demonstrated by the enduring inequality in traffic enforcement nationwide. Definitional defects, judicial interpretation, and lack of effective enforcement mechanisms work both separately and together to undercut the policy goals that are held up to justify such bans. Second, given the inefficacy of statutory bans, jurisdictions should pursue change through other state-level reforms, either through aggressive interpretation of state constitutions, or alternatively by changing their police forces’ approach to traffic stops.
The argument proceeds in three parts. Part I begins by giving an overview of the history of racial profiling and the role that pretextual stops have played in the perpetuation of racially biased patterns in policing. It then examines how the current state of the law makes traffic stops particularly vulnerable to selective application, subsequently enabling officers to bypass prohibitions on race-based policing. Part I ends by explaining why state-level reform is better suited to addressing the racially disparate impact of traffic enforcement than federal reform. Part II analyzes police traffic stop data to determine whether there has been any downward trend in the overrepresentation of Black drivers in the states that have enacted racial profiling bans as well as bans on race-based pretextual stops. Part II then goes on to identify the weaknesses in each law and then assesses how they may work individually and in tandem to undercut the intended purpose of the enacted racial profiling laws. Part III examines alternative proposals and assesses the strengths and weaknesses of each in turn.
I. Background
A. A System Primed for Abuse
In 2018, over twenty-four million people came into contact with the police as the result of a traffic stop.16 Theoretically, traffic stops are performed when a police officer has a reasonable articulable suspicion that a violation of the jurisdiction’s traffic code has occurred or is occurring.17 The problem is that traffic codes are sweeping in scope and extremely detailed; a given state’s code can contain prescriptions as straightforward as when to use a vehicle’s turn signals,18 or requirements as exacting as the precise positioning of and distance that a license plate must sit from the ground.19 Violations of either type could form the valid basis of a traffic stop. Moreover, officers need not always be correct in thinking that a violation has occurred in order for a traffic stop—and, by extension, any evidence recovered from it—to be considered valid.20 Finally, since the traffic codes are typically so comprehensive that universal enforcement would be impossible, officers are allowed to use their discretion to decide who to stop.21 This broad latitude in enforcement is what gives the pretext stop its outsized power.
A pretext stop occurs when an officer who wishes to investigate an unrelated crime uses a minor traffic infraction as a pretext to stop a citizen and begin their investigation.22 The United States Supreme Court has famously declined to prohibit this practice under the Fourth Amendment, reasoning that the Fourth Amendment does not take an individual officer’s reasons for conducting a stop into account.23 As such, under the current law, traffic stops are generally permissible under the Fourth Amendment as long as the officers making the stop have probable cause to believe that a traffic violation has occurred.24 This is true regardless of how minor the violation, or any ulterior motives that the officers might have in making those stops.25
Scholars have consistently criticized this aspect of the Court’s Fourth Amendment jurisprudence as virtual carte blanche for police to stop whomever they choose, often with inequitable results.26 Although selective enforcement based on race is decidedly unconstitutional, critics have asserted that the scope of the traffic code necessarily leaves so much discretion to police officers that factors such as race, gender, or national origin could be used, either consciously or not, to target individuals for enforcement.27 Critics worried that the Whren decision would effectively leave victims of discriminatory police tactics with no recourse.28
B. The Racial Profiling Problem
Around the same time that the Supreme Court took up the question of pretextual stops, state governments faced a public reckoning over their racial profiling practices.29 Although racially discriminatory policing practices have existed as long as the institution of policing itself,30 racial profiling as it is traditionally understood first gained widespread use by the federal Drug Enforcement Administration (DEA) during the Reagan era.31 In the 1980s, the DEA began creating “profiles” of typical drug couriers, which included race-neutral factors such as driving patterns or particular stopping points, but also listed characteristics such as race, age, and gender.32 These profiles were then disseminated to local law enforcement officers in trainings on drug interdiction best practices.33 Scholars and civil rights organizations began challenging the use of racial profiles during routine traffic stops as a form of discrimination in the 1990s after several high-profile incidents of police brutality drew broader attention to the issue.34 These challenges, supported by studies at the local level demonstrating the inefficiency of racial profiling practices,35 resulted in the passage of legislation that banned the use of race in determining which individuals to subject to more intrusive police behavior.36 This legislative project, which began in the late 1990s, is ongoing today.37 Currently, existing bans come in an assortment of structures; some prohibit racial profiling but omit any enforcement mechanism for officers who engage in prohibited behavior,38 while others merely impose data collection requirements.39 Some statutes, like Missouri’s, impose consequences for noncompliance.40 Only seven states both expressly prohibit the use of racial profiling in selecting individuals for enforcement and actually give a specific definition of what racial profiling is.41 These statutes consist of most or all of the following provisions: a definition of racial profiling that includes reliance on race, ethnicity, national origin, sexual orientation, gender, or religion in deciding whether to subject an individual to investigatory proceedings;42 a mandate requiring law enforcement agencies in the state to draft a policy prohibiting such practices and outlining requirements for such policies;43 officer training requirements;44 collection of demographic data for those who are subjected to traffic stops or submit complaints of profiling;45 and the creation or identification of a distinct entity to whom the data or complaints will be submitted for review.46 At first glance, these statutes appear to be comprehensive, and thus the most likely to succeed. However, traffic enforcement continues to have racially disparate effects nationwide, even in these states where race may not be used as a factor in activity as routine as traffic enforcement.47
State-level reform is the best vehicle to address enduring inequalities in policing for two related reasons. First, the vast majority of police personnel in this country operate under state and local management, subject to rules—and sanctions—set by their commanding departments as well as their state and local governments.48 As such, department policy changes, state judicial review, and specific legislative action are the most direct means of influencing police behavior. Second, while it is true that police are generally constrained by federal equal protection statutes as well as the Constitution, the federal government’s authority to regulate state and local police practices is limited.49 Federal power in this sphere is confined to indirect influence. The federal government can certainly incentivize change—for example, it could condition federal grants to departments on certain policy commitments, or it could leverage the Department of Justice’s enforcement power and sue noncompliant departments.50 Incentives, however, only work at the department level. They are not guaranteed to engender change in individual officers’ behavior the way that state- and department-specific policy changes or judicial review can. It is generally acknowledged that changes in behavior are essential to the successful implementation of any type of policing reform.51 Despite the fact that racially disparate policing is a nationwide problem, in this instance, it is actually the states that occupy the better position to find workable solutions.
II. Analysis
The difficulty in drafting statutory bans on race-based pretextual stops that can address racial disparities in police enforcement is multifaceted.52 The statutes from the states highlighted in this Part are helpful in that they require departments to collect raw data, but for reasons discussed in Section II.A, drawing causal conclusions from raw data alone is complicated. However, what the data can and does show, as discussed in Section II.B, is that racial disparities persist in states that have express statutory bans in place, underscoring their inefficacy as currently drafted.53 The majority of statutory bans operate under a handicap: not only do they suffer from mechanical infirmities, such as a failure to include enforcement mechanisms and definitional gaps, but their force is also undercut by judicial review. As discussed in Section II.C, these impediments handicap the statutes and make it impossible for them to bring about their intended result.54
A. Limitations of Traffic Stop Data
At the outset, it should be noted that race-based pretextual stops are difficult to identify even when one has the benefit of police testimony, let alone in the vacuum of quantitative data.55 It is often impossible to know when the given reason for a stop is pretextual, save for an officer testifying as such.56 Identifying race-based pretextual stops is doubly difficult; an officer is unlikely to admit they stopped a civilian because that person was of a certain racial group, as such an admission would automatically make the stop unconstitutional.57 For that reason, much of the evidence that race-based pretext stops continue to occur is anecdotal.58 One way to make up for this empirical deficit is by looking at the proportion of each population that is subject to a stop, whether or not they were searched, and on what grounds they were subjected to that search.59 For example, if a larger proportion of a racial group is subjected to a search once the driver has been stopped, that could be indicative of bias in decisions regarding who to search.60 Even better is to discern the proportion of the population that is subjected to a search when the traffic infraction justifying the stop is relatively minor, such as a failure to signal.61 In states that provide data on the basis for a search, one could potentially estimate the number of stops that are pretextual by isolating the number of traffic stops that result in searches that were conducted pursuant to consent rather than probable cause.62 Since consent searches require the least amount of articulable evidence to justify the search,63 they are the strongest indicators that the desire to investigate another crime motivated the stop outside of a direct admission to that effect.64 This conclusion could be further bolstered by evaluating the search “hit rate” for each racial group.65 For example, in some jurisdictions in 2020, Hispanic drivers were searched twice as often as white drivers, but searches of Hispanic drivers were less likely to turn up contraband.66 This disparity could potentially support a claim that Hispanic drivers were unjustifiably targeted for stops and subsequent searches in those jurisdictions.67
However, traffic stop data can be difficult to interpret, especially when used to make comparisons between states. Differences between each state in specific data collection requirements can create the impression of bias or discrimination where none exists.68 The data provided often spans different date ranges.69 These complexities are further compounded by varying departmental policies on how to collect and record data.70 Moreover, the quality of collected data can vary between departments within a state.71 Finally, conclusions drawn from stop rates alone can be problematic, because differences between groups of people are not necessarily attributable to discrimination, but rather could be due to race- or gender-specific driving habits.72
Using data from the Stanford Open Policing Project can help account for such difficulties.73 To date, the project has aggregated traffic stop data from over 255 million records.74 The project collects data by filing public records requests with the individual state patrol agencies as well as municipal police departments.75 The advantage of using data from the Stanford Open Policing Project is that the project standardizes the collected data such that more accurate conclusions can be drawn about policing on both a departmental and national level.76 Such standardization is necessary because of idiosyncrasies in each department’s submitted data, which, as explained above, can vary based on department policy, as can the time period covered by the data.77 The relatively vague language in most statutes’ data collection mandates, along with weak enforcement mechanisms in each, compounds this issue.78 Furthermore, not only does each state differ at the statutory level regarding what type of data departments are supposed to collect, but the degree of each department’s compliance, even within a state, varies as well. For example, although Section 12525.5(b) enumerates the specific data points that officers are required to record upon making a traffic stop,79 the Los Angeles Police Department did not provide data on each enumerated category to the Stanford Open Policing Project’s public records request.80 New Mexico’s police departments do not collect traffic stop data at all, and therefore could not be included in the dataset published by the project.81 For those reasons, analysis in this Note is limited to intrastate trends within individual departments to the extent possible, rather than extending to interstate or interdepartmental comparisons.82
B. Traffic Stop Demographics
At time of writing, there are only seven states that have statutes that approach the NAACP’s definition of a comprehensive racial profiling law83 and apply it to prohibit racial profiling in pretextual stops.84 In theory, these statutes represent some of the most thorough prohibitions on racial profiling. A closer look at the demographics collected from these states belies the statutes’ impotence.
The purported goal of these racial profiling and race-based pretextual stop statutes was to help reduce the disparate impact of policing across racial groups.85 If these statutes were, in fact, having that effect, one would expect to see stop and search rates that closely mirror population demographics in those states, or at least to see demographic overrepresentation diminish over time.86 Instead, the data shows relative stagnancy across years in stop rates, with Black drivers consistently overrepresented in stop data compared to their proportion of the state population.87 This holds true for almost every department examined in this Note.88 In Wichita, Kansas, Black people made up 11.5% of the population in 2015.89 However, between 2006 and 2016, Black drivers made up between 12% and 16% of traffic stops in the city.90 In Missouri, Black people made up 11.5% of the state’s population in 2015,91 but between 2012 and 2019, Black drivers represented between 17% and 20% of stops made by the Missouri police departments statewide.92 In an extreme case, in Oakland, California, Black people constituted just 26% of the population in 2015.93 However, between 2013 and 2017, Black drivers made up between 57% and 60% of the stops made by the Oakland Police Department.94 With the exception of California,95 all of these state statutes have been effective since at least 2011.96 Even allowing for a grace period as departments begin the initial process of collecting data, agencies have had ample time to use the data they have collected to inform their practices. The fact that disparities in enforcement consistently remain high, and in fact have increased over time, is indicative of the statutes’ inability to achieve that goal, especially since some of them have remained unamended since their initial enactment.97
Data from Connecticut, Missouri, and California, which all include the necessary information to calculate hit rates, support the conclusion that despite the enactment of racial profiling and race‑based pretextual stop bans, traffic enforcement is still being applied inequitably.98 In Connecticut, the demographics of the population in 2015 were as follows: 77.3% white, 10.3% Black, 4.2% Asian, and 14.7% Hispanic or Latino.99 These demographics remained relatively stable between 2013 and 2015, with only minimal variations.100 For those years, the traffic stop data indicates that the stop rate actually reflects population demographics rather closely: white drivers made up between 69% and 73% of stops, while stops of Black drivers constituted between 13% and 14% of stops.101 However, Black drivers made up about 26% of total searches.102 In comparison, despite making up the vast majority of total stops, searches of white drivers made up, on average, only 50% of the total search count.103 The Stanford Open Policing Project’s analysis of officers’ decisions to search indicates that this disparity is most likely due to discrimination in selecting who to search.104 In Missouri, the data published by the state indicates that since 2000, Black and Hispanic drivers have consistently been searched more often than white drivers with little meaningful shift, yet have consistently had lower hit rates.105 This evidence indicates that, despite the presence of the racial profiling and race‑based pretextual stop bans, there is a good chance that pretextual stops are still being applied in ways that have racially disparate effects. Moreover, this trend could be present in all the departments in the aforementioned jurisdictions but is simply not visible due to the lack of data reporting.
C. Self-Sabotaging Statutes
A preliminary analysis of the stop data indicates that these statutes have not had the desired effect of reducing the use of racially discriminatory policing tactics. There are at least three reasons that this could be true. First, not all of these statutes mandate the type of data needed to create effective policy solutions.106 Collecting the most basic demographics can be helpful in establishing that disparities exist but will not explain why those disparities exist.107 To be useful, the data needs to eliminate potentially obfuscating factors, like criminality and race‑specific driving habits.108 It also needs to cover enough departments in a state such that policymakers have an adequate sample size from which to make inferences. Finally, the data must span a sufficient number of years to detect long-term trends.109 Without this information, state legislatures and local law enforcement departments will not be able to obtain a clear picture of what is happening from department to department, and therefore could never adequately address failures. The fact that each department in the states above is free to draft its own policies regarding training and stop procedures can bring a powerful element of flexibility to law enforcement—after all, the challenges facing law enforcement in each state can vary widely, and department policy decisions reflect that variety.110 However, gaps in data collection will always create a blind spot in policy creation, preventing police from serving their communities equitably. While some of the datasets examined in this Note satisfy the above-mentioned criteria, the majority do not.111 The extent of the problem is made clear by revisiting the data from Connecticut. Recall that Connecticut’s traffic stop demographics parallel the population fairly closely.112 If one looked only at the demographics of traffic stops, one might believe that police departments in Connecticut had succeeded in eliminating racial profiling as a policing tactic. It is only by looking to the search data that one can observe the bias in its policing tactics. If Connecticut did not collect the search data—and, indeed, many states are in such a position—that statistic would be wholly invisible except in anecdotal form, and department policies would not be able to respond appropriately.
However, there is evidence that data collection alone cannot resolve the racial profiling problem and that the statutes, as enacted, are inherently incapable of doing so as well. Section 590.650 of Missouri’s Code is relatively robust in comparison to the other states’ statutes. It includes an enforcement mechanism by which the governor may withhold state funds from noncompliant agencies.113 It has typically had fairly universal compliance from all departments that are bound by the statute to report their traffic stop data.114 Finally, the data it requires police departments to collect is thorough; departments are obligated to collect stop data, search data, and hit rates, and review such data periodically to look for patterns among individual officers.115 The Attorney General’s report is obligated to include a “disparity index”: a proportion equaling the actual number of drivers stopped in a particular demographic over the number that is expected based on their proportion of the population.116 But even though Section 590.650 addresses the majority of the concerns raised above, Missouri’s stop and search rates of Black and Hispanic drivers started high in 2000 when the state first began collecting data and have stayed high in subsequent years.117 Furthermore, despite its ostensibly powerful racial profiling and pretext stop ban, Missouri has had a number of high-profile police brutality and racial profiling incidents in recent years stemming from traffic stops; Ferguson, a city in the Greater St. Louis metropolitan area, has been under a consent decree with the Department of Justice since 2016, following an investigation revealing persistent patterns of prejudiced policing and abuses of power.118 Analysis of Missouri’s traffic stop data has indicated that the statute has inspired no significant change in racial disparities, and in fact racial gaps have widened over the past twenty years since Section 590.650 became effective.119 This data reveals over- or underrepresentation of groups within the population of those subjected to stops.120 Since Missouri has begun collecting traffic stop data, Black and Hispanic drivers have consistently been overrepresented in the total number of stops and searches.121 Given the relative strength of Missouri’s statute and the lack of resulting change, there is strong evidence that racial profiling and pretextual stop bans, without further guidance, simply cannot equalize the racially disparate impacts of pretextual stops.
There are at least three reasons why, despite seemingly robust language, a statutory ban on pretextual stops and racial profiling may be impotent. First, judicial interpretation of these statutes is lacking. Claims of racial profiling are generally raised as a matter of equal protection under the Federal Constitution rather than as a question of state law.122 As such, there has been scant opportunity for state judiciaries to apply an aggressive interpretation of these statutes, or to articulate any binding interpretation at all. Even when claims of racial profiling are raised as a matter of state constitutional law, all of the aforementioned states apart from New Mexico have applied some version of the Supreme Court’s holding in Whren v. United States.123 In Whren, the Court held that so long as there is an objectively reasonable basis for conducting the stop— for example, a witnessed traffic violation or belief that one has occurred—an inquiry into the factors that led a police officer to decide to stop the individual is inappropriate.124 The Court argued that the Fourth Amendment only required probable cause to believe that a violation of law occurred.125 Since violations of the traffic code qualify as violations of law, the Court reasoned, the Fourth Amendment required no further inquiry.126
The states that have adopted this approach weaken their statutes’ potential impact. Whether or not a racially motivated pretextual stop has occurred necessarily requires a subjective inquiry, because the question examines a particular officer’s motivation for executing the stop. Because the Whren approach forecloses analysis based on an officer’s subjective motivations in the Fourth Amendment context, courts are unable to detect when a pretextual stop has occurred contrary to existing laws. Without undertaking an inquiry into subjective intent, the statutes—and any potential consequences for violating them—can be bypassed entirely. As a result, citizens are left vulnerable to profiling even in states where public servants are expressly forbidden from considering race in carrying out their duties. Under this framework, the bans on racial profiling become little more than data collection requirements with strong language that purport to protect citizens from misconduct without actually doing so.
A second problem with these statutes is that four out of the seven states discussed in this Note have express exceptions in which race may still be employed in policing, and the language of Connecticut’s statute is ambiguous as to whether race may be used in combination with other factors.127 In these states, allowing the use of race in combination with other factors creates room for race to play an otherwise impermissible role in deciding which individuals to subject to search.128 For example, the Arkansas, California, Kansas, and New Mexico statutes all create an exception to the proscription on the use of race when race is identified with other characteristics in a specific suspect description.129 However, suspect descriptions can be extremely unreliable and can themselves be the product of racial biases.130 Moreover, even when police have a number of descriptive characteristics for an individual suspect, the race descriptor often bears an outsized effect on who they choose to investigate.131 The result is that innocent people have the shadow of probable cause cast over them for no reason other than their perceived race, directly contravening the purpose of the statutes. One notorious example is in Brown v. City of Oneonta.132 That case involved the burglary and assault of an elderly white woman at night.133 Despite not being able to see clearly due to poor lighting, the victim alleged that she could tell the assailant’s arm was brown and that he moved quickly.134 Based on that information alone, she identified her assailant as a young Black man and told police that he had cut his hand in the struggle of the assault.135 Using this description, police officers stopped over two hundred Black university students and residents to inspect their hands.136 At the time, there were less than three hundred Black people total living in Oneonta.137
These exceptions also compound the risk that innocent people will be swept up in the net of the criminal legal system. There is limited evidence of legislative intent associated with these statutes,138 which means that there are fewer guiding principles for judges to follow when interpreting these statutes.139 That lack of context can translate into an unwillingness to invalidate searches based on race and exclude the evidence that results from them. The exclusionary rule is an incentive for police officers to operate within the bounds of the Constitution.140 Without it, there is no practical reason for police to change their investigatory tactics. Consequently, in its absence, individual biases can permeate enforcement decisions. Moreover, since inquiries into subjective intent in pretextual stop cases are categorically repudiated, courts may be less able to identify or remedy those harms when they do occur.141 These statutes are then even further undermined by additional exceptions to the proscription of the use of race in policing: Missouri carves out an exemption from the statute for certain types of roadblocks and checkpoints,142 and New Mexico allows police to use race-specific investigations if they get a credible tip linking persons of a certain race to a particular crime.143 These exceptions threaten to swallow the rule against racial profiling.
Finally, the race-based policing statutes as a whole cannot be effective because they largely carry no enforcement mechanism for departments that violate them. Each statute carries with it mandates, be they data collection requirements, training requirements, or policy-drafting requirements, but out of the seven states discussed in this Note, only Missouri and New Jersey name any potential consequences for failure to comply with their respective demands.144 Without enforcement mechanisms in place, there is little statutory incentive for police departments to comply with any of the requirements. However, even with enforcement mechanisms in place, the statutes seem poorly positioned to gain real results. Missouri’s statute provides that each law enforcement agency should create a policy that “[p]rohibits the practice of routinely stopping members of minority groups,” reviews police officers for patterns, provides training for those who have developed patterns of stopping minorities, and provides counseling for them.145 However, aside from a brief description of what a training should cover, there are no details as to how the law enforcement agency can satisfy this policy requirement.146 Moreover, the singular enforcement tool—the withholding of funds—is entirely discretionary.147 Given that police departments have notoriously poor internal and interdepartment accountability structures, the lack of consequences for officers who continue to engage in race-based policing practices is particularly alarming.148
New Jersey’s statute takes a different approach by creating criminal penalties for public servants who engage in discriminatory behavior while carrying out their public duties.149 However, even that penalty is not necessarily going to be effective. First of all, the statute has never successfully been applied to a police officer.150 Second, even if an officer’s conduct was severe enough to warrant indictment, it is not immediately clear that the officer would be found guilty. An element of the crime is that the officer must subject another to unlawful arrest or detention, with the term “unlawful” defined later in the statute as an act that either (1) violates the United States Constitution or the New Jersey Constitution, or (2) is a criminal offense under the laws of New Jersey.151 Under the Supreme Court’s decision in Whren, subjecting someone to a pretextual stop is not a violation of the Fourth Amendment, even if the main impetus for the stop is race-based.152 New Jersey courts have typically applied the Whren decision in most cases of pretext stops regardless of whether race was a motivating factor, with the only exceptions being in cases where defendants were able to present incredibly detailed statistical evidence demonstrating discriminatory treatment.153 Therefore, given the state of the law, it is hard to say whether such conduct would actually violate the New Jersey Constitution as well. There is also the additional mens rea problem: for criminal liability to apply, the state would have to be able to show that the officer knowingly engaged in conduct that was unlawful with the purpose to intimidate or discriminate.154 Aside from the fact that it is not immediately clear whether making pretextual stops, even if using race as a factor, is actually unlawful as defined by N.J.S.A. 2C:30-6, it is conceivable that an officer still may not be convicted for failure to act with the requisite purpose.
III. Possible Solutions
The data suggests that, in their current forms, the type of bans on racial profiling and race-based pretextual stops in these seven states are insufficient to eliminate, or even mitigate, the role that racial profiling plays in perpetuating a racially discriminatory policing system. A commonly proposed reform is to limit the use of pretextual stops, although the form that such a ban would take is often debated.155 Many scholars and lawmakers have proposed legislative solutions, which often involve reducing, in some form, the role that police play in traffic enforcement.156
What drives disparities in policing seems largely to be police discretion.157 As such, reform-minded jurisdictions should strive to eliminate police discretion from the equation altogether, either by limiting jurisdiction over traffic enforcement or wholly disentangling traffic enforcement from criminal enforcement. There have been a number of proposals to accomplish these goals, couched both in legislative and judicial solutions. The following Sections consider these proposals in turn.
A. Legislative and Policy Reform
The main problem with allowing officers to engage in traffic stops is that the amount of discretion that is afforded to officers makes them extremely vulnerable to unconscious biases in deciding who to stop.158 Moreover, because crime prevention is a major function of policing and often a high priority,159 police officers will always have an incentive to instigate more stops in the hopes of detecting criminal activity. As such, some scholars and activists have argued that disentangling traffic enforcement from crime prevention, namely by replacing police officers with a wholly separate traffic enforcement body, would be the most effective solution to the pretext stop problem.160 The city of Berkeley, California has elected to take such an approach.161 In the Berkeley model, all traffic stops—infractions such as speeding or running red lights—could be outsourced to cameras, which only capture when a car violates a traffic safety law.162 So long as cameras are not disproportionately placed in neighborhoods of color, they may provide the necessary neutrality to avoid creating racially disparate traffic outcomes. For minor traffic violations, Berkeley is planning to outsource enforcement to an unsworn, unarmed group of transportation workers, called BerkDOT.163 BerkDOT workers will have no authority to enforce criminal laws; when they observe a traffic violation, such as a broken taillight, they will only be authorized to issue a “fix-it” ticket.164 This “fix-it” ticket, in turn, is not a fine in and of itself, but rather more like a warning to resolve whatever issue that the driver was pulled over for.165 Should a driver fail to comply with the ticket, the punishment is community service rather than a fine.166 This approach has received quite a bit of pushback from police unions, which raise safety and cost concerns in opposition to the proposal.167 Moreover, though the proposal tries to delay the use of fees, they may yet be imposed if the driver fails to resolve the ticket within a given time frame.168 Using fees at all introduces risk that low-income people will bear a disproportionate burden, further perpetuating a cycle of income insecurity and, potentially, contact with the criminal legal system. However, this proposal would effectively negate the possibility of pretext stops as well as avert at least some inequities resulting from contact with the criminal legal system, since police would no longer have discretionary authority to conduct traffic stops at all.169
Alternatively, some departments have chosen to limit the discretion that officers have when engaging in traffic enforcement.170 Under this model, the department continues to enforce traffic stops, but deliberately reduces the number of discretionary traffic stops it may engage in.171 The Oakland Police Department in California made such a policy shift in 2018, with the specific intention of reducing racial bias in traffic stops.172 Under the department’s policy, the number of discretionary stops—stops where police initiated the interaction, rather than being called out to help—dropped by nearly 13,000 stops.173 The program is still in its early stages, so detailed search and hit rate analysis, which could give a fuller picture of the program’s efficacy, has yet to emerge. In the meantime, advocates are wary of the racial disparities that appear to persist, since the data indicates that Black drivers are still being stopped at higher rates than every other group despite being less than a quarter of the city’s population.174 Moreover, opponents are still skeptical, as they believe that the refusal to stop drivers for low-level offenses could constitute a shirking of duty in a city that is already notorious for its dangerous roads.175
It should be noted that, in an attempt to reassure the public, the Oakland Police Department has stated that despite the new stop guidelines, it will continue to execute traffic stops when there is dangerous driving involved.176 This assurance directly undermines the purpose of the program, which is to reduce the number of discretionary stops entirely.177 This, in turn, makes it much less likely for the program to succeed. Whether or not certain behavior can be considered dangerous is extremely subjective.178 The promise to continue stopping “dangerous” drivers may simply be an attempt to address public concerns, but is also fundamentally inconsistent with the program’s premise, since officers could continue to stop drivers based on subjective judgment calls rather than according to any objective standard. This kind of ambiguity is another source of vulnerability for reforms based strictly in department policy.
Another weakness of department-based reforms is that they are not particularly capable of creating lasting change. For one, department policy decisions are inherently less durable than legislation, and police departments, while accountable on some level to local mayors and governors, are generally free to adopt policies as they see fit.179 For another, so long as a particular traffic violation remains on the books as a law with criminal penalties for violation, police may use it as a ground to conduct arrests.180 If such an arrest occurred, there would be no remedy for the arrested citizen; their arrest would be consistent with the Fourth Amendment even if the violation in question were only punishable by a fine.181 This is what gives the Berkeley model more potential to succeed as a policing reform: not only are the Berkeley reforms couched in legislation rather than mere policy choices—making them more enduring—but they also sidestep some of the doctrine that has made the pretext stop so powerful by decriminalizing traffic enforcement altogether.182
An alternative legislative proposal similarly limits police discretion in traffic stops while retaining police as the traffic enforcement body. In this model, it is the legislature, as opposed to individual police departments, that decides which offenses to remove from police officers’ jurisdiction.183 At least one state has adopted this approach, with others still considering the proposal.184 In late 2020, the Virginia General Assembly passed a bill that made it unlawful for law enforcement officers to stop cars for a number of minor traffic offenses.185 The bill expressly limits the use of traffic stops for offenses that are often abused as pretext to investigate other crimes, such as window tints, broken exhaust systems, and failure to illuminate a license plate.186 It also forbids law enforcement officers from searching or arresting someone solely on the basis of a perceived odor of marijuana, which would eliminate a particularly potent ground upon which police searches are conducted.187 As one Virginia politician who supported the bill noted, police intervention can often be extremely problematic.188 Because the bill does not decriminalize such offenses, but rather prohibits officers from conducting stops or arrests, traffic violations remain valid grounds upon which officers can conduct searches and arrests without interference from the Fourth Amendment.189 As such, while this measure creates more lasting change than a simple shift in department policy would, it is still lacking in that key regard.
As illustrated in opposition points raised against the programs in Oakland and Virginia, there are legitimate safety concerns that must be addressed if legislatures choose to pursue jurisdiction removal.190 This is what the Berkeley proposal seems to acknowledge: the reason that pretextual stops are so insidious is because they are often plausibly couched in real safety concerns.191 Any proposal that strips away police authority to make those stops, without an entity to replace them, is going to be vulnerable to insinuations that proponents are not properly prioritizing road safety. For that reason, it can be particularly difficult to garner support for such a proposal, making it all the more difficult to get meaningful reforms passed.
The proposed Berkeley model addresses this vulnerability directly. In his paper Traffic Without the Police, Jordan Blair Woods outlines a number of stops that, under his proposal, police would no longer have the authority to make.192 Instead, the authority to enforce the vast majority of traffic violations would be transferred to a traffic monitoring entity that is divorced from law enforcement in the normal course of its duties.193 This includes the stops that the Oakland and Virginia plans have already phased out, such as failures to signal or low-level speeding.194 This is exactly what the Berkeley model proposes, although Berkeley identifies a slightly different set of traffic violations that would be removed from police authority.195 Woods also identifies a subset of traffic stops that could potentially involve police presence, such as driving a stolen vehicle, driving under the influence, or driving without a license.196 This proposal would address opponents’ concerns about traffic safety while still advancing the goal of disentangling competing criminal and traffic enforcement interests. Moreover, if employed in conjunction with the systematic decriminalization of certain traffic offenses, this proposal could further strengthen the movement toward racial and economic justice in the criminal justice system.197
B. Judicial Remedies
Given the resistance to eliminating police from traffic enforcement entirely, it is important to consider reforms that retain police in the traffic enforcement role, but disincentivize the overuse of the traffic stop to further criminal enforcement goals. State courts can be a powerful vehicle for this change. State supreme courts are recognized to have the authority to decide whether their state constitutions demand protection beyond what is afforded in the Federal Constitution.198 This authority allows state courts to interpret their respective state constitutions as requiring an application of the “would have” standard proposed by the petitioners in Whren.199 The petitioners in the case advanced the theory that the Fourth Amendment test of reasonableness for traffic stops should be whether or not a reasonable officer would have made the stop under the given circumstances.200 Research is conflicted on how effective a “would have” standard may be in actually protecting drivers.201 However, new research implies that it may have at least some constraining effect on police discretion.202 This suggests that an aggressive application of the standard may help equalize racial disparities while more permanent solutions are pursued in the legislature.
Before the summer of 2020, there were two states that applied the “would have” standard to traffic stops as a matter of state constitutional law: New Mexico203 and Washington.204 New Mexico does not collect stop data,205 so it is difficult to quantify any differences in police behavior. But in Washington, police departments have provided sufficient data to the Stanford Open Policing Project to evaluate the difference that expressly allowing pretextual stops makes in racial disparities.206 This type of analysis is possible in Washington because it is the only state to apply a “would have” standard, and then revert to something closer to the federal rule.207
For over ten years after the Whren decision, Washington courts consistently applied the “would have” test to traffic stops.208 However, in late 2012, the Washington Supreme Court changed course and introduced what it calls “mixed-motive” stops: stops in which the officer is motivated to make a traffic stop on both legitimate and illegitimate grounds.209 In doing so, the court narrowed its pretextual stop ban to an extremely small subset of cases, and, in the words of the dissent, effectively lifted the ban on pretextual stops.210
Researchers have found that after the Washington Supreme Court handed down Arreola, there was a statistically significant increase in stops of people of color in the years following the decision.211 Moreover, this shift regarding the treatment of drivers of color was consistent with the theory that police officers were using the mixed-motive stop to engage in racial profiling.212 While the study is only limited to a single jurisdiction, its findings make a strong case for the proposition that once officers were given more leeway to make traffic stops under Washington’s pretext stop doctrine, they might have applied that discretion in ways that adversely impacted people of color.213 Application of the “would have” standard in Washington led to evidence suppression in at least some cases where officers engaged in pretextual behavior.214 If the state were to return to a “would have” standard, police might be less likely to engage in pretextual behavior, and if not, drivers would at least have some modicum of protection against pretextual traffic stops.215
There are strong arguments that interpretation of state constitutions can be leveraged to apply the “would have” standard to traffic stops as opposed to the Whren standard. State supreme courts are recognized to have the authority to decide whether their state constitutions demand protection beyond what is afforded in the Federal Constitution.216 The language of its state constitution that the Washington Supreme Court relied upon in applying the “would have” standard is admittedly broader than that of the Fourth Amendment,217 but New Mexico’s constitutional protection against search is virtually identical to it.218 Aside from the language, the courts in those states relied on the fact that interpretation of their respective state constitutions has traditionally been more protective of individual liberties than the Federal Constitution.219 Such an interpretive principle is neither novel nor unique to those states—New Jersey has likewise held that its state constitution’s Fourth Amendment equivalent is more protective of its citizens, particularly in the motor vehicle context.220 Judicial reform of the pretext stop doctrine is available in other similarly situated states and would not even require application of the state’s racial profiling ban to the facts of a case before the courts.
However, judicial reform is constrained by precedent in a way that makes it inferior to legislative- or policy-based reforms. As explained in Section II.C, one reason that the statutes are inert is because courts in those states have consistently adhered to the federal standard regarding pretextual stops, even when a specific claim of racial profiling is raised.221 Such precedents are the main obstacle to judicial reform of state courts’ pretextual stop doctrine. While judicial doctrine can always change, principles of reasoned judicial decision-making dictate that some semblance of stare decisis must stand.222 If a given state does not have a settled precedent of departing from the federal rule, the courts are much less likely to implement a “would have” standard, even if their state’s constitutional language was similar to that of New Mexico’s or Washington’s constitutions.223 Moreover, even if the courts in a particular state had consistently interpreted their state constitution more broadly than the Federal Constitution, judicial reforms are not necessarily the most effective way to achieve lasting systemic change. Both Washington’s and New Mexico’s state supreme courts have since narrowed the application of the “would have” standard, either expressly or in practice.224 Given how the “would have” standard has changed in both New Mexico and Washington, judicial reform in this respect does not seem likely to be a long-term solution to the racial profiling problem as legislative or policy solutions could be.
However, a departure from Whren need not emerge from the search and seizure principle it articulated. In the Whren opinion, Justice Scalia expressly stated that the defendants’ insinuation that they had been subjected to selective enforcement on the basis of their race was a matter for the Equal Protection Clause rather than the Fourth Amendment.225 Recognizing the equal protection issue, the Massachusetts Supreme Judicial Court decided a case in September 2020 that lowered the standard of evidence for defendants who allege that they were subjected to a traffic stop on the basis of their race.226 In that case, the Massachusetts court overturned a prior precedent that required defendants to present statistical evidence demonstrating the officer’s bias to support an allegation of racial profiling;227 it found that imposing such a high burden of evidence was inconsistent with both state and federal formulations of equal protection.228 It also acknowledged the unique difficulties that traffic stops pose, which it cited as justification for a different approach, thus providing a roadmap to other state supreme courts that may wish to reevaluate their respective pretextual stop doctrines but do not want to adopt the “would have” standard.229 However, it should be noted that by nature of being a purely judicial reform, this approach is likewise vulnerable to the longevity issues identified above. Therefore, while it may be an effective intermediate solution, it remains less likely to create long-term reform the way that legislative or policy changes can.230
An alternative to constitutional intervention is judicial interpretation of other statutes to bolster the racial profiling and pretextual stop bans. This is the approach that Kansas has undertaken in its efforts to address the racial profiling problem. The Kansas Supreme Court has held that Kansas state law as articulated in Section K.S.A. 22-4609, the statutory prohibition on the use of race in policing, requires a subjective inquiry into the officer’s intent when an officer conducts a traffic stop. 231 The court made this determination in State v. Gray. In that case, the Kansas Supreme Court heard a motion to suppress based on Kansas’s statutory exclusionary rule, and it found that it could undertake an inquiry into an officer’s subjective intent for making a traffic stop to resolve the question.232 This approach was made possible by Kansas’s statutory exclusionary rule, which is separate from any Fourth Amendment or state constitutional equivalent and specifies that the exclusionary rule is the remedy for all unlawful searches.233 Since K.S.A. 22-4609 indicates by its plain meaning that a stop made with reliance on race as a factor is unlawful, the court reasoned that the statute required exclusion in that case.234 In doing so, the court sidestepped the constitutional question in order to give fuller protections than would otherwise be available under Whren.235
This method of interbranch teamwork only works if there is a statutory exclusionary rule to apply and if the state’s search and seizure jurisprudence otherwise supports its application in a particular case.236 In other words, this solution only works if there is a combination of legislative and judicial support for it, further supporting the idea that judicial reforms alone are not likely to result in long-term change. For the most part, states’ interpretation of the exclusionary rule can range from exactly parallel to the federal analogue to expressly contradictory, so long as the language of the state constitution supports a particular interpretation.237 In a state where the common law exclusionary rule doctrine tends to closely follow the federal doctrine, it may be harder to justify a departure from existing practice without express directive from the legislature, as the Kansas Supreme Court had done.238 Moreover, a state must have already passed a statute declaring traffic stops based on race unlawful for the Kansas approach to work.239 These preconditions make the Kansas approach a generally impracticable solution, especially when compared to the legislative reforms proposed above.240
Conclusion
When the defendants in Whren stood before the Supreme Court, they argued that the traffic code was too broad in scope for police officers to enforce it equitably.241 They were right. Study after study shows that when police officers are given the discretion to enforce the traffic laws as they see fit, they invariably do so in ways that disproportionately impact Black and Hispanic drivers.242 This may not solely be the result of conscious bias, but rather the result of centuries of systemic bias compounding upon itself. It would have been astounding if systemically ingrained biases could have been resolved simply by enacting broad statutory prohibitions on racialized policing. This analysis suggests the truth of that sentiment. Bans on racialized policing are not effective when other types of pretextual stops remain part of the policing playbook, even when combined with specific bans on race-based pretextual stops. As reform-minded jurisdictions begin to sort through the panoply of options available to them, one thing is clear: statutory bans on racial profiling and race-based pretextual stops, without more, do not go far enough. It is time for new ideas.