A couple weeks ago, WBEZ's Chip Mitchell ran a useful exercise as part of the station's series on violence in Chicago: What would it cost to just hire all the people—around 30,000—who are likely to commit violence? His solution would cost $1.1 billion.
It's a lot of money, equivalent to 13 percent of the city's budget, and close to the CPD's budget of $1.4 billion. But if it worked (under the assumption that people who are gainfully employed are less likely to get involved in criminal activity), it would in theory have economic benefits.
So the first question is: would it work? Thomas Abt, a senior research fellow at Harvard and its Kennedy School of Government, has some criticisms about job programs as violence prevention and some criticisms of the estimates in the piece that are worth taking into consideration, like the size of the population targeted by Mitchell's idea. He seems to misread significant parts of it; while he's right to be skeptical of the benefits of job-training/employment programs alone, that's not exactly what Mitchell sketches out.
Instead, more than half the $1.1 billion would be devoted to "wraparound services," which Mitchell describes as "counseling for traumatic stress from violence," "remedial education," "training in job-related 'soft skills' such as teamwork," expungement, drug-addiction therapy, and so forth. He sets aside $20,000 per person for that, which would pay for quite a bit of services.
And there's some evidence that jobs plus external support programs do help reduce crime, and Abt notes one of the latter in a piece for Vox about proven initiatives: the One Summer Plus program here in Chicago, which I've written about before. It reduced violent-crime arrests by 43 percent among the high-school-age participants. That's similar in magnitude to another well-regarded Chicago program, the cognitive-behavioral-therapy-based Becoming a Man, which as Abt notes, reduced violent-crime arrests by 50 percent. It's plausible to think that the wraparound services could include that as well.
Another of Abt's criticisms is worth taking seriously: that targeting 30,000 people is way more than is necessary, and thus counterproductively expensive. He cites findings that networks of violence are very small, such as Andrew Papachristos's work. He identified one large network of violence in the city, by identifying social ties between people who had been arrested together, of around 24,000 people at a higher risk of being homicide victims. But 41 percent of the homicides in that network were concentrated within just over 3,000 people whose risk is highly elevated.
Mitchell used unemployment stats in high-violence community areas to come up with his estimate of 30,000 people who would be good subjects for the idea. An alternative would be a smaller group identified by different metrics; on the other hand, if you did have $1.1 billion dollars, employing many of the unemployed generally would have other positive effects.
Anyway, there are lots of ways you could slice and dice Mitchell's work. The broader point was: if politicians think that jobs are the best anti-violence solution, here's what it would cost. So I was curious: Assuming it did work, would it pay for its $1.1 billion cost?
There are a couple ways of looking at it. Conveniently, the Center for American Progress specifically looked at how much money reducing violent crime by 10 or 25 percent would save.
And the answer is that in 2010 Chicago a 25 percent reduction in violent crime—which, given the results of One Summer Plus and Becoming a Man, seems at least plausible—would have saved… $1.3 billion. Weird.
That figure is complicated. They calculate only $59 million in "direct cost savings to the city budget," and those savings could be used for a 66-percent bump in community services. Other direct costs include "medical, property, and work- or productivity-related costs born by surviving victims and by victims of homicide."
But the big savings come from "indirect" costs. To calculate those, they used a study on jury awards, a common way of figuring out how society views the cost of crime to victims in something like a market. (Those are probably overestimates, but the authors assume that since much violent crime is never reported, it averages out to a reasonable figure.)
In all, they calculate that a 25 percent drop in violent crime would lead to $276 million in direct savings and $1.1 billion in indirect savings.
And they're not including other secondary effects. Mitchell, for instance, writes that "the violence is probably also a factor in the city’s population drop — a drain on the tax base." And the CAP authors take a run at that. Their calculations associate a 2.1 percent increase in housing prices with a 25 percent decrease in homicides—which in Chicago would have been, in 2011, a $2.2 billion increase in home values. These calculations aren't fine-grained enough to include in their estimated savings from a drop in crime, but they're worth considering.
Last week the Tribune reported that Austin is no longer the most populous community area in Chicago after a 45-year run, due to a recent exodus of residents. The neighborhood has a number of problems, but the Trib's Marwa Eltagouri found that "many residents say safety is the biggest issue." Examining the issue, the Metropolitan Planning Council's Alden Loury found further evidence of a correlation between violence and population loss in Chicago, if not firm causation.
The CAP's figures are in the ballpark of other estimates. In 2009 the University of Chicago's Crime Lab estimated the annual cost of gun violence in Chicago at $2.5 billion a year, based on averages of 2,500 shootings a year. That, of course, has gone up; last year, there were 3,550 shootings. Using their estimate of total societal costs of $1 million per shooting, updated for inflation, that would come out to $4.1 billion dollars. A 25 percent reduction to that gets awfully close to Mitchell's estimated cost.
These are, of course, broad estimates. And Mitchell's sketch of a plan would surely look different if it got turned into legislation, filtered through experts and evidence-based programs. Maybe it would be smaller and more targeted; maybe the wraparound services would look different. But it's also a good reminder that there are ideas out there to implement, and a sense that they're worth it.