'The siege downstairs': Data needed to regulate SWAT overuse

The Austin Statesman's Tony Plohetski had a great story over the weekend about the use of no-knock raids by the Austin PD SWAT team. The story provides a compelling example of SWAT excesses, but we'll need for SWAT raids the kind of data Texas has begun to require on police shootings before it's possible to make valid, systemic critiques of SWAT or propose sweeping solutions.

Plohetski features a case of a raid on the home of a teen living with his parents. The kid was reportedly selling pot and had legally purchased an AK-47 for protection after an attempted robbery at the house. Reported Plohetski:

The teen, who later told detectives that he didn’t know the intruders were police officers and feared for the safety of himself and his family, began shooting with an AK-47, striking an officer in the leg. The family’s dog was killed in the carnage.

And, in the end, the massive operation yielded minimal reward: Officers recovered 1.2 ounces of marijuana. ...

The incident raises questions about the use of risky military-style tactics by municipal police, occasionally for lower-level offenses. The issue has been part of a national policing conversation for several years, but the case illustrates how Austin police deploy SWAT resources in what some critics say is a troubling trend.

A little more than once a week, on average, generally under the cover of darkness, heavily armed Austin officers force their way into a home in search of evidence to help build cases. The majority of such cases involve suspicion of narcotics.

Grits has given these questions some thought and research over the years and at every turn one immediately runs into the same problem. Anecdotes are easy to find but there's almost no data available to analyze SWAT-related issues. Radley Balko made the most valiant effort to date. But at the end of  the day, you can't manage what you can't measure, and policymakers have no metrics for confronting these sorts of police militarization issues.

Grits would like to see reporting by police agencies regarding SWAT deployments the same way the Texas Lege last year required them to report police shooting episodes to the Attorney General. Make me Philosopher King and for every SWAT deployment, police would be required to report the reason for deploying a SWAT team rather than another police engagement strategy, the underlying alleged offense, the number of officers deployed, whether the suspect was present when the raid occurred, any special equipment deployed including flash-bangs, whether there was damaged property due to the use of SWAT equipment (preferably with photos), whether use of force was used, whether contraband was confiscated, and whether there were any arrests. That would go a long way toward remedying the present information imbalance.

Once data collection has been in place for a time, we'd have a much better grasp on the scope of SWAT-related problems. Right now, the debate tends to grope around in the dark. If you're anti-SWAT you're focused on overkill episodes like the one in the paper; pro-SWAT interests, by contrast, are busy touting the response to the Orlando shootings. But it's not an either-or issue. We need SWAT teams in larger jurisdictions for extraordinary episodes, but there's also a legitimate need to limit mission creep and prevent their use in situations where lesser levels of force would suffice.

One other idea to which Grits keeps coming back: Perhaps judges should provide higher levels of oversight for SWAT warrants? Right now, a judge issues a warrant and cops decide whether to use SWAT, whether to use no-knock tactics, etc.. Why not make them justify using those tactics to the judge? It'd be a small thing, but perhaps it would limit SWAT use in marginal cases like the one portrayed in the Statesman.

There are surely other reforms needed for unfettered and unregulated SWAT programs - and I hope readers will suggest some in the comments - but without more data, the debate remains frustratingly imprecise and one-sided. That's probably the first thing to fix if we truly want to get a handle on the problem.