A Watchdog Accused Officers of Serious Misconduct. Few Were Punished.

In the more than five years since New York's governor ordered the attorney general to take charge of investigations and prosecutions when police kill unarmed people, only three police officers have been charged with crimes and none has been convicted. By taking control of such cases away from local prosecutors, officials hoped to instill greater trust in outcomes, because of the close ties local prosecutors have with local police. But the narrow scope of the rule establishing the special prosecutions unit, plus a host of other systemic and legal barriers, have meant that the outcomes have not changed.

When Ann Arbor created a citizen panel overseeing its police department, it chose the approach that research shows is the one best suited to having real authority, and thus the most likely to reduce racial disparities in arrests and police shootings. It's too soon to know if the agency's investigations of complaints against police and review of police budgets and policies will achieve the ultimate goal of improving community trust in the police. But its chair says it is in a position to press for more accountability and transparency. East Lansing has just adopted the same model.

Five years after the FBI started tracking how often police use force, the majority of police departments still fail to comply and the FBI refuses to release publicly what information it has collected. The policy was enacted in response to the realization that no one had definitive data on how often the police kill people, use teargas, or other incidents of force. What little data exists showed racial disparities in whom police use force against. But compliance was made voluntary and the FBI made public release of the data contingent on 80% of police departments complying, a goal it's nowhere near.

More than a year after taking effect, California's law restricting when police can use deadly force has had some effects on police accountability and training, but a number of flaws illustrate how long and difficult the process of change will be. The Act to Save Lives, which limits the use of deadly force to cases when it is needed to defend a life, has been cited by prosecutors in two homicide prosecutions. A number of police departments have followed the law's training dictates. But many others have been slow to roll out the training, and the state is not requiring officers to take it to stay certified.

After the South Carolina legislature in 2005 began requiring police to enforce a seat-belt law with traffic stops, it imposed on all police departments a duty to track and report traffic-stop data on drivers' race. The law was meant as a way to prevent biased policing, through public disclosure of disparities. But only about one-third of the state's law enforcement agencies have consistently complied with the law, and some never have. Some blame ignorance of how the system should work. But another explanation is the state has done little or nothing to enforce the law.

After East Haven, Connecticut, police officers were caught harassing residents based on race, the Obama Justice Department took the police department to court and won a consent decree requiring a long list of reforms, in hiring, training, discipline, and use of force. The oversight, rare for a small city, changed the department's culture and won praise from many residents, who now trust the police more. Such federal action waned in the Trump years, but is expected to revive in the Biden administration, though perhaps under a more collaborative, less coercive model.

Once viewed as one of the nation's most brutal and corrupt police departments, New Orleans Police Department has earned steadily improving public support with a host of reforms. One reform that it took voluntarily, and in contrast with common practice in Louisiana, is to quickly release body-camera videos of police shootings and other uses of force. Though its effect is hard to untangle from other initiatives, video releases have become routine. In one case, a video prompted an official apology after proving rubber bullets were used against protesters. This helped lead to new restrictions on crowd control.

One of the most powerful tools used to reform policing practices, widely credited with restoring public faith in such troubled departments as the Los Angeles Police Department, is called a consent decree. The U.S. Justice Department sues cities where police abuses are seen as rampant. Then, under the watchful eye of a judge and independent monitor, the department agrees to a package of reforms. Under the Trump Justice Department, though, the tool has gone unused in new cases. In existing cases, the government has become passive, allowing cities to flout their agreements without consequence.

New Jersey’s responses to police misconduct offer stark lessons in how certain failures can allow troubled officers to hop from job to job undetected. The state lacks a central database of police misconduct and remains one of only five states that cannot revoke an officer’s credentials because of performance. In one case, an officer with serial failures at nine departments, involving 16 incidents in which people he arrested got injured, the red flags that should have been raised never were, allowing him to mistreat residents of majority-Black neighborhoods despite a record indicating racism and brutality.

Derek Chauvin's journey through the Minneapolis Police Department’s officer-disciplinary system illustrates the weaknesses of that system and the failure of efforts to fix it. Chauvin, the officer charged in the death of George Floyd, survived at least 17 misconduct complaints before he was fired for killing Floyd by kneeling on his neck. Critics charge the department never complied with recommendations by federal analysts to improve the tracking and disciplining of problem officers. That and other administrative failures are coupled with political and cultural barriers to neuter many reform ideas.