A quasi-experimental study of the generalized enforcement of low-level violations in New York City finds that proactive policing increases crime. This finding suggests the importance of taking a careful look at aggressive enforcement approaches used by police to reduce crime as they may be causing harm in urban communities.
Just a quarter of a century ago, there was wide agreement among police researchers that the police do not prevent crime. A series of studies of what the National Academy of Sciences has called the ‘standard model of police practices’ yielded little evidence of the crime control effectiveness of policing1. The standard model of policing was typified by generalized reactive policing approaches, such as random preventive patrols in a city, rapid response to emergency calls and police investigations of crimes. At that time, few police scholars would have been surprised with a finding showing no effect of policing on crime rates; though, perhaps even then, a finding that policing practices increase crime, such as that reported in a new study by Sullivan and O’Keeffe2 in Nature Human Behaviour, would have generated much interest.
Since that time, a generation of police researchers has focused on new and innovative policing approaches, and their findings stand in stark contrast with the ‘nothing works’ evidence of earlier decades. These approaches could be contrasted with the standard model of policing in that they are focused rather than generalized and extend the tools of policing beyond traditional law enforcement. They have often been grouped under the term ‘proactive policing’ because they involve the marshalling of police resources in a direct effort to prevent crime, rather than as part of the more general obligations of the police to respond to civilian requests. Hotspots policing, for example, which capitalizes on what I have termed ‘the law of crime concentration’3, focuses policing resources on the small proportion of streets that produce most of the crime in a given city. There are now more than 20 published studies, many of them randomized experiments, that show that hotspots policing will reduce crime in hotspots without significant displacement4. Indeed, strong evidence suggests that areas nearby the hotspots also improve — a phenomenon termed ‘diffusion of crime control benefits’5. Focused deterrence programmes, which target high-rate violent offenders, using both police threats of sanctions and community pressures, have also been found to be effective6.
In the context of this growing body of evidence that proactive policing ‘works’, the study by Sullivan and O’Keeffe is particularly important. The authors take advantage of a natural experiment in New York City that resulted from the strangling death of Eric Garner in Staten Island. Subsequent political events led to the New York City Police Department (NYPD) engaging in a ‘slowdown’ characterized by dramatic reductions in arrests and summonses. One would have expected crime to go up in this period if this type of proactivity was effective. Instead, analyzing several years of data obtained from the NYPD, they find that civilian complaints of major crimes decreased. Accordingly, they conclude that prior proactivity did not reduce crime, but led to increases in crime. This is an important finding and one that needs to be examined carefully. How can one reconcile a growing body of scientific literature that shows that proactive policing prevents crime with a strong study that shows the opposite effect?
One solution is to simply note the differences between this study and the more general literature on proactive policing. This study examines a particular type of proactive policing — the aggressive enforcement of low-level violations, sometimes termed broken-window policing. What the study can be seen to say accordingly is that broken-window policing applied in a generalized rather than focused way across a large urban city leads to increases in crime.
The finding that widely dispersed aggressive enforcement of low-level violations does not impact on crime is consistent with a series of quasi-experimental and experimental studies. One recent review of these studies reached the conclusion that increasing the misdemeanour arrest rate is not effective in reducing serious crime7. However, evidence that such approaches lead to increases in crime is novel. In turn, it is important to note that focused aggressive enforcement at places with high violent crime has been found to be an effective crime control approach (for a review see ref. 8).
Sullivan and O’Keeffe suggest that the cause of the positive relationship between proactivity and major crime is due to the potential negative impacts of such activities on citizens. This proposition follows a growing interest in whether fair and impartial policing, which gives citizens a voice in encounters, leads to positive evaluations of police legitimacy (see ref. 9 for a description of this set of ideas). In this context, the causal mechanism for crime increases can be seen as a result of a decline in the willingness of citizens to obey the law when they believe the police have acted in procedurally unjust ways. This is certainly one possible explanation for their findings. At the same time, however, a recent review of this literature finds little evidence for this causal process in practice, though of course a lack of evidence does not mean that the mechanism is not operating10.
This leaves us, in my view, with the question of how we should assess the evidence of a positive relationship between proactive policing and major crime. Certainly, this evidence should not be generalized across the broad array of proactive policing strategies. It is limited to a very specific approach applied very broadly in a large urban area. But at the same time, as the authors note, they have capitalized on an interesting natural experiment that allows them to overcome the causal limitations of previous non-experimental work. This is well done social science, and in this context the results have strong weight.
At the same time, absent a randomized experiment, it is difficult to draw unambiguous conclusions about the relationship between policing and crime. While the authors take advantage of the political shock in New York City policing that caused a slowdown in one type of proactivity, it is unclear what the ‘slowdown condition’ actually was. For example, New York City has a long history of identifying crime hotspots for intervention. Did the NYPD continue to identify such hotspots and send police to those places? In that context, the findings become even more interesting. Perhaps police patrol at such places without aggressive enforcement is more effective than highly aggressive strategies, which can lead to officers being pulled off the streets to process offenders. My point is simply that we don’t know what the shock actually tells us because we don’t have a detailed understanding of what the police were doing.
This gap suggests that we need experimental evidence of the impacts of proactivity at the jurisdictional level. Non-experimental studies simply cannot overcome the myriad threats to the causal interpretation of the findings. Such jurisdictional experiments are difficult and expensive to carry out. Nonetheless, if proactivity of this type is not simply ineffective but leads to increases in crime rates, we need to have reliable evidence now. The first law of treatment is ‘do no harm’. Sullivan and O’Keeffe suggest that generalized enforcement of low-level offenses does just that. We have a moral imperative to carry out well-designed experimental studies that can give us unambiguous answers to this important policy question.
National Research Council Fairness and Effectiveness in Policing: The Evidence Committee to Review Research on Police Policy and Practices (National Academies Press, Washington DC, 2004).
Sullivan, C. M. & O’Keeffe, Z. P. Nat. Hum. Behav. 1, https://doi.org/10.1038/s41562-017-0211-5 (2017).
Weisburd, D. Criminology 53, 133–57 (2015).
Braga, A. A., Papachristos, A. V. & Hureau, D. M. Justice Q. 31, 633–63 (2014).
Clarke, R. V. & Weisburd, D. Crime Prev. Studies 2, 165–84 (1994).
Braga, A. & Weisburd, D. Ann. Rev. Public Health 36, 55–68 (2015).
Braga, A. A., Welsh, B. C. & Schnell, C. J. Res. Crime Delinq. 52, 567–588 (2015).
Koper, C. S. & Mayo-Wilson, E. J. Exp. Criminol. 2, 227–261 (2006).
Tyler, T. R. Ann. Am. Acad. Pol. Soc. Sci 593, 84–99 (2004).
Nagin, D. & Telep, C. Procedural justice and legal compliance. Ann. Rev. Law Soc. Sci. (in the press).
About this article
Cite this article
Weisburd, D. Proactive policing and crime control. Nat Hum Behav 1, 707–708 (2017). https://doi.org/10.1038/s41562-017-0227-x
Czech Sociological Review (2020)
Crime count and crime harm in a post-socialist city: How does the law of crime concentration at places apply?
European Journal of Criminology (2020)