When federal Judge Shira A. Scheindlin on Monday tamped down New York City’s stop-and-frisk policy, she cited the most popular knock against the technique—that in practice it amounts to racial profiling. The NYPD made 4.4 million stops between January 2004 and June 2012, and of those stopped more than 80 percent were black or Hispanic.
Proponents of stop-and-frisk, on the other hand, cite the efficacy of that technique in concert with other “aggressive” policing actions in reducing crime. Besides, the highest crime rates are also in areas with the highest minority population.
There’s a fair bit of academic research on stop-and-frisk, most of it looking at the racial aspect that drew the judge’s censure. “Overall,” Indiana University-Northwest’s Joseph Ferrandino writes about this research, “there is less empirical evidence on the effectiveness of NYPD stop and frisk policy than on the equity of its application.”
Ferrandino’s own contribution to the debate lies in looking not at ethnicity or effectiveness, but on the actual technical efficiency—how many guns were confiscated, drugs found, and arrests made compared to the number of encounters—in the best-performing precincts and in the worst. This is not to make sure that police hit paydirt every time they frisk someone, which currently occurs in about one of 10 “stop, question, and frisk” encounters. But it does seek to ensure that they perform in line with a realistic, yet intentionally high, baseline.
In short, if there are going to be stop and frisks—and Scheindlin made clear she was “not ordering an end to the practice”—let’s make sure each has a realistic chance of bearing fruit. Arguably, that’s already the goal, but in a paper published last November in the journal Criminal Justice Review Ferrandino notes that reality and theory don’t always mesh:
The Terry standard—that a frisk is justified if undertaken to find weapons for the sake of officer or public safety during a reasonable stop of a citizen—holds throughout all police departments and their respective officers. This suggests that, theoretically, it is a comparable input across all NYPD precincts and its application, if consistent with legal standards, should lead to similar outputs produced.
Ferrandino examined seven years of data on the actual stops in all of the city’s 76 police precincts. After each stop-and-frisk police officers are supposed to fill out a form, the UF-250, that details, among other things, the end result of the encounter and the ethnicity of the subject searched. Ferrandino’s results come from looking at all those cards filled out from 2004 to 2010. There are limitations—some officers have complained they have been given quotas of frisks, others may not fill out paperwork properly or for every encounter, and the number of hits leaves out data on later convictions, dismissals, or even reprimands.
To determine if that theory of consistency was true—and it wasn’t—Ferrandino used a technique known as “data envelopment analysis” to look at frisk efficiency over time, how many fewer frisks should have been made based on the resulting bust (and realizing “that an unsuccessful frisk has ramifications on the individual and mostly minority communities”) or conversely how many more busts should have occurred to match the number of frisks, and what precincts and boroughs were the best and worst at efficiency. The analysis provides a sort of best-possible benchmark that other precincts can emulate.
It’s worth noting that about half the department showed improvement over the duration of Ferrandino study. While “still inefficient overall,” 31 precincts and one borough improved their excessive-frisk efficiency (input) while 38 precincts and two boroughs improved their recovering contraband/making arrests efficiency (output). And if it seems like the cops aren’t going to toil with spreadsheets like some fantasy-baseball fanatics, Ferrandino notes that the data-heavy geographic information system CompStat has, for better or worse, become part of the department’s arsenal.
His paper didn’t look at outcomes based on race, since his efforts are less tearing down stop-and-frisk and more making bullet-proof. Besides, other studies, like a 2007 Rand paper sponsored by the New York City Police Foundation, have compared ethnicity and outcomes. “Regarding all contraband, such as weapons, stolen property, or drugs,” author Greg Ridgeway wrote, “when we compared white, black, and Hispanic suspects who were matched to have similar stop features, we learned that recovery rates are nearly the same (whites have slightly higher rates), suggesting that officers apply nearly the same standard of suspicion regardless of race.”
That Rand study suggested that from a third to a half of the stops were essentially superfluous. Ferrandino noted that his own “input-oriented efficiency analysis” suggests that a frisk-efficient NYPD should have reduced frisks by almost two-thirds, from 1,721,955 to 630,109, over his study period. “This would equate to roughly 90,000 frisks per year on average rather than the current average of 245,993 per year,” he writes. Assuming you axed the wrong frisks, which is in practice a big assumption, this might have partially assuaged opponents of the practice by making each stop more likely to be righteous.
That is, of course, if you seek fewer frisks. If you think they help create order and so at a minimum want to see the same number, then Ferrandino suggests the cops should have arrested 179,056 more people, found 6,306 more pistols, and nabbed 59,883 more packets of drugs to justify the count of frisks they did make.
This assumes that efficiency is the goal. “One could argue,” Ferrandino writes in demolishing this straw man, “that lower efficiency (increased inputs and decreased outputs) actually supports the deterrence position that the increased chance of being frisked lowers crime in the form of less people carrying guns or contraband though more recently scholars generally supportive of the NYPD have questioned this as illogical, fatally flawed, inequitable, and possibly illegal.”
Not surprisingly, Ferrandino argues that efficiency should be the goal regardless of where you stand on the issue: “Frisks should be reduced by some degree in accordance with the officer safety standard of Terry, and successful frisks should be increased through analysis of the most successful officers, precincts, and divisions to inform the wider department.” This has moved from an academic’s dream to a judge’s order.