The police practice of “stop, question and frisk,” or “stop-and-frisk” — randomly checking citizens for drugs or guns — was used heavily by New York City police starting in the 1990s to deter or stop crime from happening. A combination of court rulings and the end of the tenure of Mayor Mike Bloomberg saw the number of stops plummet from nearly 700,000 in 2011 to about 10,000 in 2017. Bloomberg is now being criticized by some for supporting the practice, which many feel was racially biased. But two professors in public policy, Charles F. Manski of Northwestern University and Daniel S. Nagin of Carnegie Mellon University, argue here that the tactic may have been the right answer for crime in the 1990s, but less so in an era of greatly reduced crime.

By Charles F. Manski and Daniel S. Nagin

Mike Bloomberg’s emergence as a candidate for president has rekindled controversy about the stop-and-frisk policing policy administered in New York during his period as mayor. Past and current discussions have taken polarized positions: Stop-and-frisk is either good policing or it is reprehensible. The truth is, there are rather obvious, yet overlooked, aspects of the policy that should be in the forefront of these debates. The use of stop-and-frisk may be justified in high-crime environments, but not when crime rates are low. We should always use less intrusive tactics when possible.

The sharp contrast in positions is readily seen in statements by former New York Police Department commissioner Raymond Kelly, who implemented the policy, and former U.S. District Judge Shira Scheindlin, who ruled it unconstitutional. Kelly wrote: “[T]he New York Police Department has taken tens of thousands of weapons off the street through proactive policing strategies … In the 11 years before [the Bloomberg administration], there were 13,212 murders … During the 11 years of his administration, there have been 5,849. That’s 7,383 lives saved.” Scheindlin wrote: “It is important to recognize the human toll of unconstitutional stops. No one should live in fear of being stopped whenever he leaves his home to go about the activities of daily life.”

Effective policing in a democratic society should balance the (sometimes) conflicting objectives of public safety and community trust. Stop-and-frisk methods may reduce crime, but the costs of intrusion on the rights and privacy of innocent people cannot be overlooked. It is not easy to weigh these costs and benefits, and it is even harder to achieve political consensus, but we have to appreciate the subtlety of the matter rather than succumb to sloganeering.

We developed a model of confrontational proactive policing and studied the implications of stop-and-frisk. The model considers the questionable aspects of the policy: How much does a police tactic reduce crime, and how much does it interfere with innocent people’s lives? And how much does it have a disproportionate impact across racial and ethnic groups?

The model’s point of departure is that stop-and-frisk can deter crime but with the cost of invading individual privacy. Another important insight is that it is always preferable to develop and employ effective policing tactics that are less intrusive.

The benefits of stop-and-frisk depend on the level of crime without the tactic. A basic point from grade school arithmetic is that X percent of a big number is larger than X percent of a small number. Therefore, stop-and-frisk may have had substantial benefits in the high-crime period of the early 1990s but much lower benefit in the present low crime environment.

We looked at New York City to understand the cost to innocent people and determine how much higher the New York City crime rate would have been without the use of stop-and-frisk. There is limited research on the crime prevention impact of stop-and-frisk. However, some estimate that the 685,000 stops conducted in 2011, the peak year of use of the policy, reduced the New York City crime rate by 2 percent. Close to 90 percent of the stops resulted in no arrest or summons, meaning that the costs of these stops were borne by innocent people.

The crime rate in New York in 2011 was relatively low. One may think that a 2 percent reduction in a low crime rate is too small of a benefit compared to the significant cost of conducting 685,000 stops — primarily of innocent people and on racial minorities — rather than spread evenly across the residents of New York. Black men ages 14 to 24 accounted for nearly 25 percent of all stops yet composed only 1.9 percent of the city’s population. Therefore, stop-and-frisk may have been bad policy in 2011.

But the trade-off looks quite different in earlier years, when the New York crime rate was much higher. If we assume that stop-and-frisk reduced crime by 2 percent in 2011, the benefit (in terms of crimes averted per capita) would have been 56 percent larger in 2000 and 220 percent larger in 1991 when crime rates were higher than in 2011.

Through the heated debates of today, and what we can expect will continue into the coming months, it is important to look back at the data, to understand the context of policy decisions, and to not overlook alternative forms of proactive policing that are less intrusive in the lives of innocent people.

Charles F. Manski is Board of Trustees professor in economics and fellow of the Institute for Policy Research at Northwestern University. Daniel S. Nagin is Teresa and H. John Heinz III university professor of public policy and statistics at Carnegie Mellon University’s Heinz College.