Police Reform Needs to Come From Within

People are rightly worried about rising crime. We know how to tackle that problem, and it’s not by defunding.

By: William Bratton as seen in The Atlantic.

No one wants authoritarian law-enforcement agencies that violate civil rights or oppress minority communities. Neither does anyone want high rates of crime, especially violent crime. Those who advocate police reform need to accept both truths, because if people feel the streets are unsafe, then the reforms—regardless of their merits—become right-wing talking points. A criminal-justice system that fails to check rising crime will, as we’ve seen, face a backlash.

So what’s a police reformer to do? Some may wonder at my weighing in on this. In certain circles, I am regarded as the worst kind of police authoritarian, largely because I was New York Police Commissioner under Mayor Rudy Giuliani in the 1990s and I supported the strategy of “broken windows” policing. But in my experience, lasting police reform comes from within. Cops understand better than anyone that when you set out to remake a police agency, you don’t want to undercut what it does well in an effort to correct what it does poorly.

Smart reform starts with winning trust from two different groups. First: neighborhoods. You want to serve local communities, respond to their requests for police action, and maintain civil order, but you don’t want to over-police and foster a sense of officers as an occupying army. At the NYPD, starting in 2014, I and my deputies sent a clear message to the rank and file to use intelligent discretion. We were looking for quality of arrests, not quantity. My successors continued this strategy. As a result, the enforcement footprint shrank—especially in the city’s communities of color. Arrests of Black and Hispanic New Yorkers fell by nearly 195,000 from 2013 to 2021, according to data from the NYPD. Meanwhile, most types of crime in the city continued on the downward trend that began in mid-1990s––until the surge that started in the pandemic spring of 2020.

Second: the cops. Since I first became a police officer, in Boston in 1970, I’ve been aware that cops tend to cocoon themselves, becoming isolated from the communities they serve. I have seen how a low work ethic and a curdled cynicism can become entrenched in a department. That’s dangerous. A complacent, unproductive department is one in which abusive and violent individual officers can thrive.

So, yes, hammer police abuse of force. Hammer police misconduct. But don’t hammer police departments and their personnel in general. Indiscriminate rhetorical attacks on police as violent and racist have a withering effect on morale. When the chorus of criticism grows too intense, you will see officers falling into lassitude, neglecting the mission, walking away from the profession. This is a path to neither successful law enforcement nor good relations between police and neighborhoods.

Violent crime is on the rise across the country, so what can be done about it? The answer lies in the sort of policing that has served New York, and other cities, well in the past but that we lost sight of amid the turmoil of the past few years. The essence of the approach is well-managed, motivated police departments that target criminal behavior in problem locations in order to provide the public safety that communities themselves rightly demand.

The data show that most street criminals—especially those responsible for robbery, burglary, larceny, and auto theft—are repeat offenders. In the mid-’90s, when we instituted the CompStat system in New York City, we started doing a much better job of identifying and apprehending these career offenders. The effect on crime rates in the ensuing decades was extraordinary.

CompStat broke down the silos in the NYPD that had beat patrols, the narcotics division, and the Detective Bureau operating separately and without coordination. Still in use by the NYPD, the system relies on mapping and other tracking technologies to gather accurate, real-time intelligence on street crime. Police can see which tactics prove effective in one precinct and then export them to others. Perhaps most important, CompStat makes it possible to see exactly how the department is performing, citywide, from day to day.

Precision policing, as the CompStat-enabled approach is known, works by targeting a specific small, recidivist criminal population. To some, the very idea of a criminal population is controversial, even reprehensible. They argue for treating each arrestee as a perpetual first offender, regardless of prior offenses, and they oppose placing chronic reoffenders in pretrial custody to protect the public. Across the country, progressive district attorneys have been declining to prosecute lesser offenses, including theft and misdemeanor assaults, and downgrading charges and possible sentences for many felony crimes.

The progressives secured a big victory in New York State’s sweeping 2019 bail reforms, which abolished bail conditions for many crimes without giving state judges the discretion to impose custody in particular cases. Noble in spirit perhaps, these measures reduced the deterrent effect of arrests, because offenders know they will face no immediate consequences if apprehended. Of the nearly 100,000 people charged but released pretrial in 2020 under New York’s reformed policy, about one-fifth were rearrested while their first case was pending. And last year, according to the New York State Division of Criminal Justice Services, more than 90 percent of the 60,000 felony arrestsresulted in no jail or prison time, or even probation; a scant 3 percent of arrests ended in prison sentences—compared with 8 percent in 2017, and 7 percent in 2018 and 2019.

This is not a blow against mass incarceration; it’s a blow against controlling felony crime. New York’s recent move to undo the damage of this failing policy—by partially restoring judges’ ability to impose bail for some crimes and by allowing bail to be set for those rearrested while a prior case is pending—is welcome but insufficient. Real reform is about leading and inspiring police organizations to fulfill both their crime control and their service missions. We need good police leaders to instill in young cops a sense of responsibility and service, and we need them to be willing to discipline those who go astray. But we must also ensure they have the tools they need to do the job.

Critics like to portray such policing technologies as DNA databases, photo-recognition software, automatic license-plate readers, and, in New York City, the gang database as instruments of Orwellian government surveillance. They are nothing of the kind: DNA, photo recognition, and license-plate readers are all more reliable identification tools than the traditional reliance on eyewitnesses (DNA, in particular, has exonerated many innocent people). The gang database, which allows police to track the membership of some 500 violent criminal organizations, is a reasonable response to the fact that in 2019, for example, gang-related shootings accounted for nearly half of shootings in the city.

Aside from gang activity, shootings in New York have always correlated closely with disorderly streets. Back in 1990, a historic high-water mark of crime in the city, there were about 5,200 shootings, or about 100 every week. This was none too surprising given that, when I joined the department in 1994, some 5,000 to 6,000 locations identified by the NYPD were plagued by open-air drug dealing or by disorderly conduct such as street prostitution, aggressive begging, squeegee nuisance, and illegal peddling. From that time, the NYPD was able to use CompStat to track and map shootings in the city; within a few years, we had also largely cleared the open-air drug markets and problems associated with them. By reasserting police authority in these hot spots, we cut shootings to approximately 1,700 by 1998, and to fewer than 800 by 2018. The rise in gun deaths and murders since 2020 has many causes, but a contributory factor has been that the department had, under political pressure, either abandoned or trimmed the units responsible for homeless outreach and for enforcement against street narcotics, prostitution, peddling, and fare evasion.

The only reform that makes sense, because of this connection between disorderly streets and violent crime, is one that aims to control broken-windows offenses. Some have caricatured this approach as state repression of minority communities. In fact, those local residents are most often the most vocal in support of tackling minor offenses that directly affect their neighborhood’s quality of life. Others will also say that a renewed focus on broken-windows policing will result in the sort of abuse we saw with the stop-and-frisk tactic in New York.

Unquestionably overused until about a decade ago, stop-and-frisk arguably amounted to harassment in neighborhoods where many Black and brown people lived. But in contrast to stop-and-frisk, for which officers require only reasonable suspicion to make a stop, broken-windows arrests and summonses require the much higher standard of probable cause. When residents report quality-of-life offenses to the 911 or 311 services, they expect action. If police departments want to serve their communities, they must do this kind of policing.

Many today have come to view the relatively low crime rates of recent decades almost as an American birthright. Police professionals, by contrast, see it neither as a fixed condition nor as a permanent accomplishment. We view crime as a day-to-day challenge, a problem that must be managed, suppressed, and controlled by law-enforcement agencies in order to achieve those low rates. We can reverse the current crime trends, because we know what’s worked before—but we won’t succeed by defunding the police.