Cameras, crime reduction and cost

Do surveillance cameras in public spaces reduce crime and are they worth the cost?

I've been highly skeptical of claims by Dallas police that new surveillance cameras downtown have radically reduced crime, mainly because their results dramatically contradict the experience in other jurisdictions that have used such cameras from a long time. Typically, DPD officials simply state that crime has declined because of cameras with no documentation and the Dallas News dutifully reports it without delving into specifics.

So I was unsurprised to see this story in the Dallas News yesterday by Steve Thompson, "Dallas surveillance cameras cut crime but are costly to monitor," claiming that in areas covered by cameras downtown, "crime there so far this year is down 11 percent from 2008."

That sounds like a big number, but it wasn't until the News published a story today by Tanya Eiserer ("Dallas crime drops 18.7% in first three months of 2009") that we gained enough context to know if that means cameras "worked." She reports that, overall, reported crime in Dallas is down a whopping 18.7% compared to last year!!

So if crime in the areas with cameras declined LESS than in areas without them, how can anyone credibly claim cameras are reducing crime? It's just not true. Simply repeating officials' claims that cameras "work" doesn't mitigate actual statistical evidence that they don't.

Meanwhile, the expense from camera systems is too much for the city to handle - a quarter million dollars a year to monitor 25 cameras 24/7 - so the city will begin to ask neighborhood groups to foot the monitoring bill, in many cases after they've already ponied up to buy the cameras in the first place. Reported Thompson:

The cameras were bought with money from neighborhood, business and philanthropic groups. They cost $8,000 to $10,000 each and thousands more dollars to link their feeds to the police dispatch center.

Police, meanwhile, have provided the manpower to monitor them. Retired and light-duty officers watch feeds 24 hours a day, zooming in on people of interest.

Industry standards recommend one pair of eyes for each 25 cameras. But in Dallas, usually only two employees juggle the nearly 100 feeds at any given time.

"There's different staffing on each watch, but generally right now, we do not have the ideal staffing down there for the cameras," Deputy Chief Tom Lawrence said.

This, as dozens more cameras have been proposed and are expected in places like Fair Park, Uptown, White Rock Lake, and along Jefferson Boulevard in Oak Cliff. Money raised by neighborhood groups is helping pay for them.

"I believe they're seeing the effectiveness of the cameras, and they want to see if they can partner with the police department," Lawrence said.

At Monday's meeting, Dallas police recommended that the city pay to monitor the cameras only in designated violent-crime hot spots. In other areas, neighborhood and business groups would have to pay for the manpower themselves.

The committee members agreed and voted to take the recommendation to the full council.

Police say it costs about $250,000 a year in manpower to monitor one station of 25 camera feeds around the clock.

This is so back-assward it's incredibly frustrating. "Violent crime hotspots" aren't the appropriate place for cameras because police can't respond quickly enough. What you need in those places are boots on the ground - actual street patrols or even walking beats in targeted areas have a much greater impact on safety because a camera cannot intervene to stop a crime. Instead, the best security studies say cameras are most useful when targeting specific, high-value assets, but not for general public surveillance.

There's no cost-benefit analysis that justifies $8-10K per camera and $250K per year for monitoring 25 of them, which is why the City of Dallas is unwilling to foot the bill for their installation or expansion to new areas. And Thompson's cost analysis doesn't even take into account the hundreds of hours police must spend watching useless video that in most cases won't actually help to solve the crime. (After all, cameras can be easily defeated by high-tech means like hats, sunglasses and hoodies.)

Dallas' experience is producing some important lessons about surveillance cameras - ones they've long ago learned in the UK: Surveillance cameras in public spaces don't reduce crime, are costly to operate, and are never a substitute for police officers actually patrolling crime-ridden areas.

MORE: A commenter points out that Dallas PD's crime reduction numbers likely have more to do with the fact that they recently altered their crime reporting protocols to reclassify or omit many reported offenses - particularly burglaries. Given that, it's clear these "reductions" in crime aren't comparing apples to apples and it's impossible to tell from these data whether crime has declined overall, downtown, or anywhere else in Dallas, much less whether surveillance cameras "worked."