To the extent that Mayor Eric Garcetti can be said to have an agenda, crime is nowhere near the top of it. The issue barely came up during the mayor's race last year, and aside from appearing at press conferences touting the latest drop in the crime rate, Garcetti has had little to say on the subject as mayor.
So it's no surprise that LAPD Chief Charlie Beck was reappointed for another five-year term on Tuesday. Despite a late-inning flurry of controversies, some more substantive than others, Beck's job was never in jeopardy. And as long as crime remains a back-burner issue for most voters, and Beck doesn't screw anything up too badly, Garcetti seems likely to give him a wide berth.
But the reappointment process did allow one issue to surface that may have implications for the broader Garcetti agenda. That issue, expertly uncovered by the L.A. Times, is the integrity of the LAPD's CompStat system.
The CompStat system came in with ex-Chief Bill Bratton a dozen years ago. Bratton brought it from the New York Police Department, where he launched it — to great acclaim — in the mid-1990s. The system uses crime statistics to determine where to deploy resources and to measure police performance. It gets a lot of credit — deserved or not — for the precipitous drop in urban crime in the last two decades.
Garcetti likes CompStat a lot, so much so that he wants to bring it to other city departments. Earlier this month, the mayor announced that introducing “CompStat-style management” citywide will be a key goal in his second year in office.
So if there's a problem with CompStat, that's now a citywide concern.
And there is a problem, according to the Times. The Times' data analysis showed that the LAPD misclassified 1,100 aggravated assaults over a one-year period, logging them as simple assaults. That made the violent crime rate appear lower than it was in reality.
The LAPD and the mayor’s office have characterized this as “human error” of the sort that crops up in any complex statistical undertaking. But the errors, according to the Times, almost always pushed the crime rate down, not up. Some LAPD veterans said this was the result of intense pressure from higher-ups to reduce the crime rate.
This phenomenon will be familiar to anyone who has seen The Wire, the HBO show about a large urban police department. On the show, which aired from 2002 to 2008, characters complained about having to “juke stats” to please department bosses and politicians. In one episode, a commander even discussed reclassifying aggravated assaults. This is not a new issue.
As the show’s creator, David Simon, expounded in an interview:
One of the themes of The Wire really was that statistics will always lie. Statistics can be made to say anything. You show me anything that depicts institutional progress in America: school test scores, crime stats, arrest reports, anything that a politician can run on, anything that somebody can get a promotion on, and as soon as you invent that statistical category, fifty people in that institution will be at work trying to figure out a way to make it look as if progress is actually occurring when actually no progress is.
As it happens, Garcetti is a fan of the show. Nevertheless, he is very keen on using data as a management tool.
“We want to hold all departments accountable with the help of solid data,” Garcetti spokeswoman Vicki Curry said on Tuesday. “And we want all departments to break down silos and bring managers together to solve problems when they are revealed by data analysis or otherwise.”
Curry added that the mayor wants to make sure that all departments, including the LAPD, are verifying their data and fixing problems as soon as they are identified.
Easier said than done. CompStat is 12 years old, and the LAPD has auditors and inspectors whose job is to make sure the data is “solid.” Nevertheless, problems like the one IDed by the Times haven't gone away. It's not going to be a simple thing to generate better new data systems from scratch for other departments.
In an age when voters are adamantly against tax hikes, cities have to try to squeeze more performance out of the workforce they have. So it makes sense that elected officials turn to data to improve efficiency.
The risk is that managers come to rely on statistics to the exclusion of sound judgment, and that underlings start to game the numbers or engage in outright fraud to get ahead.
If Garcetti has internalized that lesson, he’s keeping quiet about it.